id	sid	tid	token	lemma	pos
cana-6215	1	1	communications	communication	NOUN
cana-6215	1	2	on	on	ADP
cana-6215	1	3	applied	apply	VERB
cana-6215	1	4	nonlinear	nonlinear	ADJ
cana-6215	1	5	analysis	analysis	NOUN
cana-6215	1	6	issn	issn	NOUN
cana-6215	1	7	:	:	PUNCT
cana-6215	1	8	1074	1074	NUM
cana-6215	1	9	-	-	PUNCT
cana-6215	1	10	133x	133x	NUM
cana-6215	1	11	vol	vol	NOUN
cana-6215	1	12	32	32	NUM
cana-6215	1	13	no	no	NOUN
cana-6215	1	14	.	.	PUNCT
cana-6215	2	1	4s	4s	NUM
cana-6215	2	2	(	(	PUNCT
cana-6215	2	3	2025	2025	NUM
cana-6215	2	4	)	)	PUNCT
cana-6215	2	5	773	773	NUM
cana-6215	2	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-6215	2	7	optimized	optimize	VERB
cana-6215	2	8	deep	deep	ADJ
cana-6215	2	9	transfer	transfer	NOUN
cana-6215	2	10	learning	learn	VERB
cana-6215	2	11	framework	framework	NOUN
cana-6215	2	12	for	for	ADP
cana-6215	2	13	accurate	accurate	ADJ
cana-6215	2	14	liver	liver	NOUN
cana-6215	2	15	tumor	tumor	NOUN
cana-6215	2	16	classification	classification	NOUN
cana-6215	2	17	amjed	amjed	NOUN
cana-6215	2	18	khan	khan	PROPN
cana-6215	2	19	bhatti	bhatti	PROPN
cana-6215	2	20	research	research	NOUN
cana-6215	2	21	scholar	scholar	NOUN
cana-6215	2	22	(	(	PUNCT
cana-6215	2	23	information	information	NOUN
cana-6215	2	24	technology	technology	NOUN
cana-6215	2	25	)	)	PUNCT
cana-6215	2	26	department	department	NOUN
cana-6215	2	27	of	of	ADP
cana-6215	2	28	computer	computer	NOUN
cana-6215	2	29	science	science	NOUN
cana-6215	2	30	and	and	CCONJ
cana-6215	2	31	engineering	engineering	NOUN
cana-6215	2	32	shri	shri	PROPN
cana-6215	2	33	venkateswara	venkateswara	PROPN
cana-6215	2	34	university	university	PROPN
cana-6215	2	35	,	,	PUNCT
cana-6215	2	36	gajraula	gajraula	NOUN
cana-6215	2	37	,	,	PUNCT
cana-6215	2	38	up	up	ADP
cana-6215	2	39	,	,	PUNCT
cana-6215	2	40	india	india	PROPN
cana-6215	2	41	email	email	NOUN
cana-6215	2	42	:	:	PUNCT
cana-6215	2	43	amjedbhatti07@gmail.com	amjedbhatti07@gmail.com	PROPN
cana-6215	2	44	dr	dr	PROPN
cana-6215	2	45	.	.	PROPN
cana-6215	2	46	deepak	deepak	PROPN
cana-6215	2	47	chandra	chandra	PROPN
cana-6215	2	48	uprety	uprety	PROPN
cana-6215	2	49	research	research	NOUN
cana-6215	2	50	guide	guide	NOUN
cana-6215	2	51	(	(	PUNCT
cana-6215	2	52	computer	computer	NOUN
cana-6215	2	53	science	science	NOUN
cana-6215	2	54	)	)	PUNCT
cana-6215	2	55	department	department	NOUN
cana-6215	2	56	of	of	ADP
cana-6215	2	57	computer	computer	NOUN
cana-6215	2	58	science	science	NOUN
cana-6215	2	59	and	and	CCONJ
cana-6215	2	60	engineering	engineering	NOUN
cana-6215	2	61	,	,	PUNCT
cana-6215	2	62	shri	shri	PROPN
cana-6215	2	63	venkateswara	venkateswara	PROPN
cana-6215	2	64	university	university	PROPN
cana-6215	2	65	,	,	PUNCT
cana-6215	2	66	gajraula	gajraula	NOUN
cana-6215	2	67	,	,	PUNCT
cana-6215	2	68	up	up	ADP
cana-6215	2	69	,	,	PUNCT
cana-6215	2	70	india	india	PROPN
cana-6215	2	71	email	email	NOUN
cana-6215	2	72	:	:	PUNCT
cana-6215	3	1	deepak.glb@gmail.com	deepak.glb@gmail.com	PROPN
cana-6215	3	2	article	article	PROPN
cana-6215	3	3	history	history	NOUN
cana-6215	3	4	:	:	PUNCT
cana-6215	3	5	received	receive	VERB
cana-6215	3	6	:	:	PUNCT
cana-6215	3	7	19	19	NUM
cana-6215	3	8	-	-	PUNCT
cana-6215	3	9	03	03	NUM
cana-6215	3	10	-	-	PUNCT
cana-6215	3	11	2024	2024	NUM
cana-6215	3	12	revised	revise	VERB
cana-6215	3	13	:	:	PUNCT
cana-6215	3	14	24	24	NUM
cana-6215	3	15	-	-	PUNCT
cana-6215	3	16	04	04	NUM
cana-6215	3	17	-	-	PUNCT
cana-6215	3	18	2024	2024	NUM
cana-6215	3	19	accepted	accept	VERB
cana-6215	3	20	:	:	PUNCT
cana-6215	3	21	21	21	NUM
cana-6215	3	22	-	-	SYM
cana-6215	3	23	05	05	NUM
cana-6215	3	24	-	-	PUNCT
cana-6215	3	25	2025	2025	NUM
cana-6215	3	26	abstract	abstract	NOUN
cana-6215	3	27	:	:	PUNCT
cana-6215	3	28	within	within	ADP
cana-6215	3	29	the	the	DET
cana-6215	3	30	competitive	competitive	ADJ
cana-6215	3	31	and	and	CCONJ
cana-6215	3	32	complicated	complicated	ADJ
cana-6215	3	33	environment	environment	NOUN
cana-6215	3	34	of	of	ADP
cana-6215	3	35	the	the	DET
cana-6215	3	36	retail	retail	ADJ
cana-6215	3	37	media	medium	NOUN
cana-6215	3	38	networks	network	NOUN
cana-6215	3	39	(	(	PUNCT
cana-6215	3	40	rmn	rmn	NOUN
cana-6215	3	41	)	)	PUNCT
cana-6215	3	42	,	,	PUNCT
cana-6215	3	43	advertisers	advertiser	NOUN
cana-6215	3	44	are	be	AUX
cana-6215	3	45	experiencing	experience	VERB
cana-6215	3	46	pressure	pressure	NOUN
cana-6215	3	47	to	to	PART
cana-6215	3	48	be	be	AUX
cana-6215	3	49	more	more	ADV
cana-6215	3	50	precise	precise	ADJ
cana-6215	3	51	in	in	ADP
cana-6215	3	52	ad	ad	NOUN
cana-6215	3	53	spending	spending	NOUN
cana-6215	3	54	.	.	PUNCT
cana-6215	4	1	conventional	conventional	ADJ
cana-6215	4	2	indicators	indicator	NOUN
cana-6215	4	3	such	such	ADJ
cana-6215	4	4	as	as	ADP
cana-6215	4	5	return	return	NOUN
cana-6215	4	6	on	on	ADP
cana-6215	4	7	ad	ad	NOUN
cana-6215	4	8	spend	spend	NOUN
cana-6215	4	9	(	(	PUNCT
cana-6215	4	10	roas	roas	NOUN
cana-6215	4	11	)	)	PUNCT
cana-6215	4	12	are	be	AUX
cana-6215	4	13	normally	normally	ADV
cana-6215	4	14	inflated	inflate	VERB
cana-6215	4	15	to	to	PART
cana-6215	4	16	reflect	reflect	VERB
cana-6215	4	17	the	the	DET
cana-6215	4	18	performance	performance	NOUN
cana-6215	4	19	of	of	ADP
cana-6215	4	20	a	a	DET
cana-6215	4	21	campaign	campaign	NOUN
cana-6215	4	22	by	by	ADP
cana-6215	4	23	ignoring	ignore	VERB
cana-6215	4	24	organic	organic	ADJ
cana-6215	4	25	conversions	conversion	NOUN
cana-6215	4	26	,	,	PUNCT
cana-6215	4	27	which	which	PRON
cana-6215	4	28	encourages	encourage	VERB
cana-6215	4	29	practitioners	practitioner	NOUN
cana-6215	4	30	to	to	PART
cana-6215	4	31	use	use	VERB
cana-6215	4	32	incremental	incremental	ADJ
cana-6215	4	33	roas	roas	NOUN
cana-6215	4	34	(	(	PUNCT
cana-6215	4	35	iroas	iroas	PROPN
cana-6215	4	36	)	)	PUNCT
cana-6215	4	37	as	as	ADP
cana-6215	4	38	a	a	DET
cana-6215	4	39	more	more	ADV
cana-6215	4	40	precise	precise	ADJ
cana-6215	4	41	metric	metric	NOUN
cana-6215	4	42	.	.	PUNCT
cana-6215	5	1	this	this	DET
cana-6215	5	2	paper	paper	NOUN
cana-6215	5	3	will	will	AUX
cana-6215	5	4	discuss	discuss	VERB
cana-6215	5	5	the	the	DET
cana-6215	5	6	combination	combination	NOUN
cana-6215	5	7	of	of	ADP
cana-6215	5	8	iroas	iroas	ADJ
cana-6215	5	9	and	and	CCONJ
cana-6215	5	10	multi	multi	ADJ
cana-6215	5	11	-	-	ADJ
cana-6215	5	12	armed	armed	ADJ
cana-6215	5	13	bandit	bandit	NOUN
cana-6215	5	14	(	(	PUNCT
cana-6215	5	15	mab	mab	NOUN
cana-6215	5	16	)	)	PUNCT
cana-6215	5	17	algorithms	algorithm	NOUN
cana-6215	5	18	to	to	PART
cana-6215	5	19	form	form	VERB
cana-6215	5	20	a	a	DET
cana-6215	5	21	dynamic	dynamic	ADJ
cana-6215	5	22	bidding	bidding	NOUN
cana-6215	5	23	strategy	strategy	NOUN
cana-6215	5	24	that	that	PRON
cana-6215	5	25	will	will	AUX
cana-6215	5	26	evolve	evolve	VERB
cana-6215	5	27	in	in	ADP
cana-6215	5	28	real	real	ADJ
cana-6215	5	29	time	time	NOUN
cana-6215	5	30	.	.	PUNCT
cana-6215	6	1	using	use	VERB
cana-6215	6	2	causal	causal	ADJ
cana-6215	6	3	inference	inference	NOUN
cana-6215	6	4	to	to	PART
cana-6215	6	5	estimate	estimate	VERB
cana-6215	6	6	iroas	iroas	ADJ
cana-6215	6	7	and	and	CCONJ
cana-6215	6	8	mab	mab	NOUN
cana-6215	6	9	models	model	NOUN
cana-6215	6	10	to	to	PART
cana-6215	6	11	trade	trade	VERB
cana-6215	6	12	off	off	ADP
cana-6215	6	13	exploration	exploration	NOUN
cana-6215	6	14	and	and	CCONJ
cana-6215	6	15	exploitation	exploitation	NOUN
cana-6215	6	16	,	,	PUNCT
cana-6215	6	17	advertisers	advertiser	NOUN
cana-6215	6	18	can	can	AUX
cana-6215	6	19	efficiently	efficiently	ADV
cana-6215	6	20	optimize	optimize	VERB
cana-6215	6	21	budgets	budget	NOUN
cana-6215	6	22	and	and	CCONJ
cana-6215	6	23	make	make	VERB
cana-6215	6	24	ongoing	ongoing	ADJ
cana-6215	6	25	progress	progress	NOUN
cana-6215	6	26	in	in	ADP
cana-6215	6	27	media	medium	NOUN
cana-6215	6	28	performance	performance	NOUN
cana-6215	6	29	.	.	PUNCT
cana-6215	7	1	the	the	DET
cana-6215	7	2	paper	paper	NOUN
cana-6215	7	3	provides	provide	VERB
cana-6215	7	4	a	a	DET
cana-6215	7	5	theoretical	theoretical	ADJ
cana-6215	7	6	background	background	NOUN
cana-6215	7	7	,	,	PUNCT
cana-6215	7	8	a	a	DET
cana-6215	7	9	scalable	scalable	ADJ
cana-6215	7	10	architecture	architecture	NOUN
cana-6215	7	11	of	of	ADP
cana-6215	7	12	implementation	implementation	NOUN
cana-6215	7	13	,	,	PUNCT
cana-6215	7	14	and	and	CCONJ
cana-6215	7	15	addresses	address	VERB
cana-6215	7	16	the	the	DET
cana-6215	7	17	problem	problem	NOUN
cana-6215	7	18	of	of	ADP
cana-6215	7	19	delayed	delay	VERB
cana-6215	7	20	incentives	incentive	NOUN
cana-6215	7	21	,	,	PUNCT
cana-6215	7	22	cold	cold	ADJ
cana-6215	7	23	-	-	PUNCT
cana-6215	7	24	start	start	NOUN
cana-6215	7	25	environments	environment	NOUN
cana-6215	7	26	,	,	PUNCT
cana-6215	7	27	preservation	preservation	NOUN
cana-6215	7	28	of	of	ADP
cana-6215	7	29	privacy	privacy	NOUN
cana-6215	7	30	,	,	PUNCT
cana-6215	7	31	and	and	CCONJ
cana-6215	7	32	exploration	exploration	NOUN
cana-6215	7	33	of	of	ADP
cana-6215	7	34	many	many	ADJ
cana-6215	7	35	-	-	PUNCT
cana-6215	7	36	objective	objective	ADJ
cana-6215	7	37	optimization	optimization	NOUN
cana-6215	7	38	.	.	PUNCT
cana-6215	8	1	overall	overall	ADV
cana-6215	8	2	,	,	PUNCT
cana-6215	8	3	it	it	PRON
cana-6215	8	4	provides	provide	VERB
cana-6215	8	5	a	a	DET
cana-6215	8	6	causally	causally	ADV
cana-6215	8	7	responsive	responsive	ADJ
cana-6215	8	8	method	method	NOUN
cana-6215	8	9	of	of	ADP
cana-6215	8	10	automated	automate	VERB
cana-6215	8	11	bidding	bidding	NOUN
cana-6215	8	12	,	,	PUNCT
cana-6215	8	13	which	which	PRON
cana-6215	8	14	causes	cause	VERB
cana-6215	8	15	responsible	responsible	ADJ
cana-6215	8	16	and	and	CCONJ
cana-6215	8	17	value	value	NOUN
cana-6215	8	18	-	-	PUNCT
cana-6215	8	19	based	base	VERB
cana-6215	8	20	advertising	advertising	NOUN
cana-6215	8	21	in	in	ADP
cana-6215	8	22	rmns	rmn	NOUN
cana-6215	8	23	.	.	PUNCT
cana-6215	9	1	keywords	keyword	NOUN
cana-6215	9	2	:	:	PUNCT
cana-6215	9	3	iroas	iroas	PROPN
cana-6215	9	4	;	;	PUNCT
cana-6215	9	5	multi	multi	ADJ
cana-6215	9	6	-	-	ADJ
cana-6215	9	7	armed	armed	ADJ
cana-6215	9	8	bandits	bandit	NOUN
cana-6215	9	9	;	;	PUNCT
cana-6215	9	10	retail	retail	ADJ
cana-6215	9	11	media	medium	NOUN
cana-6215	9	12	networks	network	NOUN
cana-6215	9	13	;	;	PUNCT
cana-6215	9	14	dynamic	dynamic	ADJ
cana-6215	9	15	bidding	bidding	NOUN
cana-6215	9	16	;	;	PUNCT
cana-6215	9	17	causal	causal	ADJ
cana-6215	9	18	inference	inference	NOUN
cana-6215	9	19	abstract	abstract	NOUN
cana-6215	9	20	:	:	PUNCT
cana-6215	9	21	this	this	DET
cana-6215	9	22	study	study	NOUN
cana-6215	9	23	addresses	address	VERB
cana-6215	9	24	the	the	DET
cana-6215	9	25	growing	grow	VERB
cana-6215	9	26	global	global	ADJ
cana-6215	9	27	concern	concern	NOUN
cana-6215	9	28	of	of	ADP
cana-6215	9	29	liver	liver	NOUN
cana-6215	9	30	cancer	cancer	NOUN
cana-6215	9	31	,	,	PUNCT
cana-6215	9	32	a	a	DET
cana-6215	9	33	highly	highly	ADV
cana-6215	9	34	fatal	fatal	ADJ
cana-6215	9	35	disease	disease	NOUN
cana-6215	9	36	where	where	SCONJ
cana-6215	9	37	early	early	ADJ
cana-6215	9	38	and	and	CCONJ
cana-6215	9	39	accurate	accurate	ADJ
cana-6215	9	40	diagnosis	diagnosis	NOUN
cana-6215	9	41	is	be	AUX
cana-6215	9	42	critical	critical	ADJ
cana-6215	9	43	for	for	ADP
cana-6215	9	44	effective	effective	ADJ
cana-6215	9	45	treatment	treatment	NOUN
cana-6215	9	46	and	and	CCONJ
cana-6215	9	47	improved	improved	ADJ
cana-6215	9	48	survival	survival	NOUN
cana-6215	9	49	rates	rate	NOUN
cana-6215	9	50	.	.	PUNCT
cana-6215	10	1	leveraging	leverage	VERB
cana-6215	10	2	medical	medical	ADJ
cana-6215	10	3	imaging	imaging	NOUN
cana-6215	10	4	data	datum	NOUN
cana-6215	10	5	,	,	PUNCT
cana-6215	10	6	the	the	DET
cana-6215	10	7	research	research	NOUN
cana-6215	10	8	proposes	propose	VERB
cana-6215	10	9	an	an	DET
cana-6215	10	10	efficient	efficient	ADJ
cana-6215	10	11	deep	deep	ADJ
cana-6215	10	12	transfer	transfer	NOUN
cana-6215	10	13	learning	learning	NOUN
cana-6215	10	14	(	(	PUNCT
cana-6215	10	15	tl)-based	tl)-base	VERB
cana-6215	10	16	framework	framework	NOUN
cana-6215	10	17	for	for	ADP
cana-6215	10	18	the	the	DET
cana-6215	10	19	automated	automate	VERB
cana-6215	10	20	classification	classification	NOUN
cana-6215	10	21	of	of	ADP
cana-6215	10	22	liver	liver	NOUN
cana-6215	10	23	tumors	tumor	NOUN
cana-6215	10	24	as	as	ADP
cana-6215	10	25	malignant	malignant	ADJ
cana-6215	10	26	,	,	PUNCT
cana-6215	10	27	benign	benign	ADJ
cana-6215	10	28	,	,	PUNCT
cana-6215	10	29	or	or	CCONJ
cana-6215	10	30	normal	normal	ADJ
cana-6215	10	31	.	.	PUNCT
cana-6215	11	1	using	use	VERB
cana-6215	11	2	computed	computed	ADJ
cana-6215	11	3	tomography	tomography	NOUN
cana-6215	11	4	(	(	PUNCT
cana-6215	11	5	ct	ct	NOUN
cana-6215	11	6	)	)	PUNCT
cana-6215	11	7	scans	scan	NOUN
cana-6215	11	8	collected	collect	VERB
cana-6215	11	9	from	from	ADP
cana-6215	11	10	the	the	DET
cana-6215	11	11	radiology	radiology	NOUN
cana-6215	11	12	institute	institute	NOUN
cana-6215	11	13	in	in	ADP
cana-6215	11	14	baghdad	baghdad	PROPN
cana-6215	11	15	medical	medical	ADJ
cana-6215	11	16	city	city	PROPN
cana-6215	11	17	,	,	PUNCT
cana-6215	11	18	iraq	iraq	PROPN
cana-6215	11	19	,	,	PUNCT
cana-6215	11	20	the	the	DET
cana-6215	11	21	study	study	NOUN
cana-6215	11	22	employs	employ	VERB
cana-6215	11	23	pre	pre	ADJ
cana-6215	11	24	-	-	ADJ
cana-6215	11	25	trained	train	VERB
cana-6215	11	26	convolutional	convolutional	ADJ
cana-6215	11	27	neural	neural	ADJ
cana-6215	11	28	networks	network	NOUN
cana-6215	11	29	(	(	PUNCT
cana-6215	11	30	vgg-16	vgg-16	NOUN
cana-6215	11	31	,	,	PUNCT
cana-6215	11	32	resnet-50	resnet-50	PROPN
cana-6215	11	33	,	,	PUNCT
cana-6215	11	34	and	and	CCONJ
cana-6215	11	35	mobilenetv2	mobilenetv2	NOUN
cana-6215	11	36	)	)	PUNCT
cana-6215	11	37	to	to	PART
cana-6215	11	38	extract	extract	VERB
cana-6215	11	39	high	high	ADJ
cana-6215	11	40	-	-	PUNCT
cana-6215	11	41	level	level	NOUN
cana-6215	11	42	image	image	NOUN
cana-6215	11	43	features	feature	VERB
cana-6215	11	44	for	for	ADP
cana-6215	11	45	improved	improved	ADJ
cana-6215	11	46	classification	classification	NOUN
cana-6215	11	47	performance	performance	NOUN
cana-6215	11	48	.	.	PUNCT
cana-6215	12	1	experimental	experimental	ADJ
cana-6215	12	2	results	result	NOUN
cana-6215	12	3	demonstrate	demonstrate	VERB
cana-6215	12	4	remarkable	remarkable	ADJ
cana-6215	12	5	accuracy	accuracy	NOUN
cana-6215	12	6	99	99	NUM
cana-6215	12	7	%	%	NOUN
cana-6215	12	8	for	for	ADP
cana-6215	12	9	vgg-16	vgg-16	NOUN
cana-6215	12	10	,	,	PUNCT
cana-6215	12	11	100	100	NUM
cana-6215	12	12	%	%	NOUN
cana-6215	12	13	for	for	ADP
cana-6215	12	14	resnet-50	resnet-50	PROPN
cana-6215	12	15	,	,	PUNCT
cana-6215	12	16	and	and	CCONJ
cana-6215	12	17	99	99	NUM
cana-6215	12	18	%	%	NOUN
cana-6215	12	19	for	for	ADP
cana-6215	12	20	mobilenetv2	mobilenetv2	PROPN
cana-6215	12	21	mailto:amjedbhatti07@gmail.com	mailto:amjedbhatti07@gmail.com	PROPN
cana-6215	12	22	mailto:deepak.glb@gmail.com	mailto:deepak.glb@gmail.com	X
cana-6215	12	23	communications	communication	NOUN
cana-6215	12	24	on	on	ADP
cana-6215	12	25	applied	apply	VERB
cana-6215	12	26	nonlinear	nonlinear	ADJ
cana-6215	12	27	analysis	analysis	NOUN
cana-6215	12	28	issn	issn	NOUN
cana-6215	12	29	:	:	PUNCT
cana-6215	12	30	1074	1074	NUM
cana-6215	12	31	-	-	PUNCT
cana-6215	12	32	133x	133x	NUM
cana-6215	12	33	vol	vol	NOUN
cana-6215	12	34	32	32	NUM
cana-6215	12	35	no	no	NOUN
cana-6215	12	36	.	.	PUNCT
cana-6215	13	1	4s	4s	NUM
cana-6215	13	2	(	(	PUNCT
cana-6215	13	3	2025	2025	NUM
cana-6215	13	4	)	)	PUNCT
cana-6215	13	5	774	774	NUM
cana-6215	13	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-6215	13	7	proving	prove	VERB
cana-6215	13	8	the	the	DET
cana-6215	13	9	superiority	superiority	NOUN
cana-6215	13	10	of	of	ADP
cana-6215	13	11	tl	tl	PROPN
cana-6215	13	12	models	model	NOUN
cana-6215	13	13	in	in	ADP
cana-6215	13	14	handling	handle	VERB
cana-6215	13	15	limited	limited	ADJ
cana-6215	13	16	medical	medical	ADJ
cana-6215	13	17	datasets	dataset	NOUN
cana-6215	13	18	.	.	PUNCT
cana-6215	14	1	the	the	DET
cana-6215	14	2	integration	integration	NOUN
cana-6215	14	3	of	of	ADP
cana-6215	14	4	additional	additional	ADJ
cana-6215	14	5	layers	layer	NOUN
cana-6215	14	6	further	far	ADV
cana-6215	14	7	enhances	enhance	VERB
cana-6215	14	8	classifier	classifier	ADJ
cana-6215	14	9	performance	performance	NOUN
cana-6215	14	10	,	,	PUNCT
cana-6215	14	11	significantly	significantly	ADV
cana-6215	14	12	accelerating	accelerate	VERB
cana-6215	14	13	the	the	DET
cana-6215	14	14	diagnostic	diagnostic	ADJ
cana-6215	14	15	process	process	NOUN
cana-6215	14	16	and	and	CCONJ
cana-6215	14	17	offering	offer	VERB
cana-6215	14	18	a	a	DET
cana-6215	14	19	reliable	reliable	ADJ
cana-6215	14	20	tool	tool	NOUN
cana-6215	14	21	for	for	ADP
cana-6215	14	22	radiologists	radiologist	NOUN
cana-6215	14	23	in	in	ADP
cana-6215	14	24	early	early	ADJ
cana-6215	14	25	liver	liver	NOUN
cana-6215	14	26	cancer	cancer	NOUN
cana-6215	14	27	detection	detection	NOUN
cana-6215	14	28	.	.	PUNCT
cana-6215	15	1	keywords	keyword	NOUN
cana-6215	15	2	:	:	PUNCT
cana-6215	15	3	liver	liver	NOUN
cana-6215	15	4	cancer	cancer	NOUN
cana-6215	15	5	,	,	PUNCT
cana-6215	15	6	transfer	transfer	NOUN
cana-6215	15	7	learning	learning	NOUN
cana-6215	15	8	,	,	PUNCT
cana-6215	15	9	deep	deep	ADJ
cana-6215	15	10	learning	learning	NOUN
cana-6215	15	11	,	,	PUNCT
cana-6215	15	12	ct	ct	NUM
cana-6215	15	13	imaging	imaging	NOUN
cana-6215	15	14	,	,	PUNCT
cana-6215	15	15	tumor	tumor	NOUN
cana-6215	15	16	,	,	PUNCT
cana-6215	15	17	vgg16	vgg16	NOUN
cana-6215	15	18	1	1	NUM
cana-6215	15	19	.	.	PUNCT
cana-6215	16	1	introduction	introduction	NOUN
cana-6215	16	2	liver	liver	NOUN
cana-6215	16	3	cancer	cancer	NOUN
cana-6215	16	4	is	be	AUX
cana-6215	16	5	one	one	NUM
cana-6215	16	6	of	of	ADP
cana-6215	16	7	the	the	DET
cana-6215	16	8	most	most	ADV
cana-6215	16	9	aggressive	aggressive	ADJ
cana-6215	16	10	and	and	CCONJ
cana-6215	16	11	life	life	NOUN
cana-6215	16	12	-	-	PUNCT
cana-6215	16	13	threatening	threaten	VERB
cana-6215	16	14	diseases	disease	NOUN
cana-6215	16	15	worldwide	worldwide	ADV
cana-6215	16	16	,	,	PUNCT
cana-6215	16	17	posing	pose	VERB
cana-6215	16	18	a	a	DET
cana-6215	16	19	major	major	ADJ
cana-6215	16	20	public	public	ADJ
cana-6215	16	21	health	health	NOUN
cana-6215	16	22	challenge	challenge	NOUN
cana-6215	16	23	due	due	ADP
cana-6215	16	24	to	to	ADP
cana-6215	16	25	its	its	PRON
cana-6215	16	26	increasing	increase	VERB
cana-6215	16	27	incidence	incidence	NOUN
cana-6215	16	28	and	and	CCONJ
cana-6215	16	29	high	high	ADJ
cana-6215	16	30	mortality	mortality	NOUN
cana-6215	16	31	rate	rate	NOUN
cana-6215	16	32	.	.	PUNCT
cana-6215	17	1	it	it	PRON
cana-6215	17	2	ranks	rank	VERB
cana-6215	17	3	among	among	ADP
cana-6215	17	4	the	the	DET
cana-6215	17	5	top	top	ADJ
cana-6215	17	6	causes	cause	NOUN
cana-6215	17	7	of	of	ADP
cana-6215	17	8	cancer	cancer	NOUN
cana-6215	17	9	-	-	PUNCT
cana-6215	17	10	related	relate	VERB
cana-6215	17	11	deaths	death	NOUN
cana-6215	17	12	,	,	PUNCT
cana-6215	17	13	largely	largely	ADV
cana-6215	17	14	because	because	SCONJ
cana-6215	17	15	symptoms	symptom	NOUN
cana-6215	17	16	often	often	ADV
cana-6215	17	17	appear	appear	VERB
cana-6215	17	18	at	at	ADP
cana-6215	17	19	advanced	advanced	ADJ
cana-6215	17	20	stages	stage	NOUN
cana-6215	17	21	when	when	SCONJ
cana-6215	17	22	treatment	treatment	NOUN
cana-6215	17	23	options	option	NOUN
cana-6215	17	24	are	be	AUX
cana-6215	17	25	limited	limited	ADJ
cana-6215	17	26	.	.	PUNCT
cana-6215	18	1	early	early	ADJ
cana-6215	18	2	and	and	CCONJ
cana-6215	18	3	accurate	accurate	ADJ
cana-6215	18	4	diagnosis	diagnosis	NOUN
cana-6215	18	5	of	of	ADP
cana-6215	18	6	liver	liver	NOUN
cana-6215	18	7	tumors	tumor	NOUN
cana-6215	18	8	whether	whether	SCONJ
cana-6215	18	9	malignant	malignant	ADJ
cana-6215	18	10	,	,	PUNCT
cana-6215	18	11	benign	benign	ADJ
cana-6215	18	12	,	,	PUNCT
cana-6215	18	13	or	or	CCONJ
cana-6215	18	14	normal	normal	ADJ
cana-6215	18	15	is	be	AUX
cana-6215	18	16	therefore	therefore	ADV
cana-6215	18	17	essential	essential	ADJ
cana-6215	18	18	to	to	ADP
cana-6215	18	19	improving	improve	VERB
cana-6215	18	20	patient	patient	ADJ
cana-6215	18	21	survival	survival	NOUN
cana-6215	18	22	rates	rate	NOUN
cana-6215	18	23	and	and	CCONJ
cana-6215	18	24	guiding	guide	VERB
cana-6215	18	25	appropriate	appropriate	ADJ
cana-6215	18	26	clinical	clinical	ADJ
cana-6215	18	27	interventions	intervention	NOUN
cana-6215	18	28	.	.	PUNCT
cana-6215	19	1	traditional	traditional	ADJ
cana-6215	19	2	diagnostic	diagnostic	ADJ
cana-6215	19	3	methods	method	NOUN
cana-6215	19	4	,	,	PUNCT
cana-6215	19	5	such	such	ADJ
cana-6215	19	6	as	as	ADP
cana-6215	19	7	manual	manual	ADJ
cana-6215	19	8	examination	examination	NOUN
cana-6215	19	9	of	of	ADP
cana-6215	19	10	computed	compute	VERB
cana-6215	19	11	tomography	tomography	NOUN
cana-6215	19	12	(	(	PUNCT
cana-6215	19	13	ct	ct	NOUN
cana-6215	19	14	)	)	PUNCT
cana-6215	19	15	and	and	CCONJ
cana-6215	19	16	magnetic	magnetic	ADJ
cana-6215	19	17	resonance	resonance	NOUN
cana-6215	19	18	imaging	imaging	NOUN
cana-6215	19	19	(	(	PUNCT
cana-6215	19	20	mri	mri	NOUN
cana-6215	19	21	)	)	PUNCT
cana-6215	19	22	scans	scan	NOUN
cana-6215	19	23	by	by	ADP
cana-6215	19	24	radiologists	radiologist	NOUN
cana-6215	19	25	[	[	X
cana-6215	19	26	9	9	NUM
cana-6215	19	27	-	-	SYM
cana-6215	19	28	10	10	NUM
cana-6215	19	29	]	]	PUNCT
cana-6215	19	30	,	,	PUNCT
cana-6215	19	31	are	be	AUX
cana-6215	19	32	time	time	NOUN
cana-6215	19	33	-	-	PUNCT
cana-6215	19	34	consuming	consume	VERB
cana-6215	19	35	and	and	CCONJ
cana-6215	19	36	prone	prone	ADJ
cana-6215	19	37	to	to	ADP
cana-6215	19	38	human	human	ADJ
cana-6215	19	39	error	error	NOUN
cana-6215	19	40	.	.	PUNCT
cana-6215	20	1	hence	hence	ADV
cana-6215	20	2	,	,	PUNCT
cana-6215	20	3	there	there	PRON
cana-6215	20	4	is	be	VERB
cana-6215	20	5	a	a	DET
cana-6215	20	6	growing	grow	VERB
cana-6215	20	7	need	need	NOUN
cana-6215	20	8	for	for	ADP
cana-6215	20	9	automated	automate	VERB
cana-6215	20	10	,	,	PUNCT
cana-6215	20	11	reliable	reliable	ADJ
cana-6215	20	12	,	,	PUNCT
cana-6215	20	13	and	and	CCONJ
cana-6215	20	14	efficient	efficient	ADJ
cana-6215	20	15	systems	system	NOUN
cana-6215	20	16	that	that	PRON
cana-6215	20	17	can	can	AUX
cana-6215	20	18	support	support	VERB
cana-6215	20	19	medical	medical	ADJ
cana-6215	20	20	experts	expert	NOUN
cana-6215	20	21	in	in	ADP
cana-6215	20	22	liver	liver	NOUN
cana-6215	20	23	cancer	cancer	NOUN
cana-6215	20	24	detection	detection	NOUN
cana-6215	20	25	and	and	CCONJ
cana-6215	20	26	classification	classification	NOUN
cana-6215	20	27	.	.	PUNCT
cana-6215	21	1	in	in	ADP
cana-6215	21	2	recent	recent	ADJ
cana-6215	21	3	years	year	NOUN
cana-6215	21	4	,	,	PUNCT
cana-6215	21	5	machine	machine	NOUN
cana-6215	21	6	learning	learning	NOUN
cana-6215	21	7	(	(	PUNCT
cana-6215	21	8	ml	ml	NOUN
cana-6215	21	9	)	)	PUNCT
cana-6215	22	1	[	[	X
cana-6215	22	2	11	11	NUM
cana-6215	22	3	-	-	SYM
cana-6215	22	4	12	12	NUM
cana-6215	22	5	]	]	PUNCT
cana-6215	22	6	and	and	CCONJ
cana-6215	22	7	deep	deep	ADJ
cana-6215	22	8	learning	learning	NOUN
cana-6215	22	9	(	(	PUNCT
cana-6215	22	10	dl	dl	NOUN
cana-6215	22	11	)	)	PUNCT
cana-6215	23	1	[	[	X
cana-6215	23	2	13	13	NUM
cana-6215	23	3	-	-	SYM
cana-6215	23	4	14	14	NUM
cana-6215	23	5	]	]	PUNCT
cana-6215	23	6	techniques	technique	NOUN
cana-6215	23	7	have	have	AUX
cana-6215	23	8	demonstrated	demonstrate	VERB
cana-6215	23	9	exceptional	exceptional	ADJ
cana-6215	23	10	capabilities	capability	NOUN
cana-6215	23	11	in	in	ADP
cana-6215	23	12	medical	medical	ADJ
cana-6215	23	13	image	image	NOUN
cana-6215	23	14	analysis	analysis	NOUN
cana-6215	23	15	,	,	PUNCT
cana-6215	23	16	particularly	particularly	ADV
cana-6215	23	17	for	for	ADP
cana-6215	23	18	tumor	tumor	NOUN
cana-6215	23	19	detection	detection	NOUN
cana-6215	23	20	and	and	CCONJ
cana-6215	23	21	classification	classification	NOUN
cana-6215	23	22	tasks	task	NOUN
cana-6215	23	23	.	.	PUNCT
cana-6215	24	1	deep	deep	ADJ
cana-6215	24	2	learning	learning	NOUN
cana-6215	24	3	models	model	NOUN
cana-6215	24	4	,	,	PUNCT
cana-6215	24	5	especially	especially	ADV
cana-6215	24	6	convolutional	convolutional	ADJ
cana-6215	24	7	neural	neural	ADJ
cana-6215	24	8	networks	network	NOUN
cana-6215	24	9	(	(	PUNCT
cana-6215	24	10	cnns	cnns	PROPN
cana-6215	24	11	)	)	PUNCT
cana-6215	25	1	[	[	X
cana-6215	25	2	15	15	NUM
cana-6215	25	3	]	]	PUNCT
cana-6215	25	4	,	,	PUNCT
cana-6215	25	5	have	have	AUX
cana-6215	25	6	shown	show	VERB
cana-6215	25	7	remarkable	remarkable	ADJ
cana-6215	25	8	success	success	NOUN
cana-6215	25	9	in	in	ADP
cana-6215	25	10	automatically	automatically	ADV
cana-6215	25	11	learning	learn	VERB
cana-6215	25	12	hierarchical	hierarchical	ADJ
cana-6215	25	13	features	feature	NOUN
cana-6215	25	14	from	from	ADP
cana-6215	25	15	medical	medical	ADJ
cana-6215	25	16	images	image	NOUN
cana-6215	25	17	,	,	PUNCT
cana-6215	25	18	outperforming	outperform	VERB
cana-6215	25	19	traditional	traditional	ADJ
cana-6215	25	20	image	image	NOUN
cana-6215	25	21	processing	processing	NOUN
cana-6215	25	22	and	and	CCONJ
cana-6215	25	23	handcrafted	handcraft	VERB
cana-6215	25	24	feature	feature	NOUN
cana-6215	25	25	-	-	PUNCT
cana-6215	25	26	based	base	VERB
cana-6215	25	27	methods	method	NOUN
cana-6215	25	28	.	.	PUNCT
cana-6215	26	1	however	however	ADV
cana-6215	26	2	,	,	PUNCT
cana-6215	26	3	one	one	NUM
cana-6215	26	4	of	of	ADP
cana-6215	26	5	the	the	DET
cana-6215	26	6	main	main	ADJ
cana-6215	26	7	challenges	challenge	NOUN
cana-6215	26	8	in	in	ADP
cana-6215	26	9	deploying	deploy	VERB
cana-6215	26	10	dl	dl	NOUN
cana-6215	26	11	models	model	NOUN
cana-6215	26	12	in	in	ADP
cana-6215	26	13	medical	medical	ADJ
cana-6215	26	14	applications	application	NOUN
cana-6215	26	15	is	be	AUX
cana-6215	26	16	the	the	DET
cana-6215	26	17	scarcity	scarcity	NOUN
cana-6215	26	18	of	of	ADP
cana-6215	26	19	large	large	ADJ
cana-6215	26	20	,	,	PUNCT
cana-6215	26	21	annotated	annotated	ADJ
cana-6215	26	22	datasets	dataset	NOUN
cana-6215	26	23	,	,	PUNCT
cana-6215	26	24	as	as	ADP
cana-6215	26	25	acquiring	acquire	VERB
cana-6215	26	26	medical	medical	ADJ
cana-6215	26	27	images	image	NOUN
cana-6215	26	28	and	and	CCONJ
cana-6215	26	29	expert	expert	NOUN
cana-6215	26	30	annotations	annotation	NOUN
cana-6215	26	31	is	be	AUX
cana-6215	26	32	both	both	CCONJ
cana-6215	26	33	costly	costly	ADJ
cana-6215	26	34	and	and	CCONJ
cana-6215	26	35	time	time	NOUN
cana-6215	26	36	-	-	PUNCT
cana-6215	26	37	intensive	intensive	ADJ
cana-6215	26	38	.	.	PUNCT
cana-6215	27	1	this	this	DET
cana-6215	27	2	limitation	limitation	NOUN
cana-6215	27	3	often	often	ADV
cana-6215	27	4	leads	lead	VERB
cana-6215	27	5	to	to	ADP
cana-6215	27	6	model	model	NOUN
cana-6215	27	7	overfitting	overfitting	NOUN
cana-6215	27	8	and	and	CCONJ
cana-6215	27	9	reduced	reduced	ADJ
cana-6215	27	10	generalization	generalization	NOUN
cana-6215	27	11	performance	performance	NOUN
cana-6215	27	12	on	on	ADP
cana-6215	27	13	unseen	unseen	ADJ
cana-6215	27	14	data	datum	NOUN
cana-6215	27	15	.	.	PUNCT
cana-6215	28	1	to	to	PART
cana-6215	28	2	overcome	overcome	VERB
cana-6215	28	3	this	this	DET
cana-6215	28	4	constraint	constraint	NOUN
cana-6215	28	5	,	,	PUNCT
cana-6215	28	6	transfer	transfer	NOUN
cana-6215	28	7	learning	learning	NOUN
cana-6215	28	8	(	(	PUNCT
cana-6215	28	9	tl	tl	PROPN
cana-6215	28	10	)	)	PUNCT
cana-6215	29	1	[	[	X
cana-6215	29	2	16	16	NUM
cana-6215	29	3	-	-	SYM
cana-6215	29	4	19	19	NUM
cana-6215	29	5	]	]	PUNCT
cana-6215	29	6	has	have	AUX
cana-6215	29	7	emerged	emerge	VERB
cana-6215	29	8	as	as	ADP
cana-6215	29	9	a	a	DET
cana-6215	29	10	powerful	powerful	ADJ
cana-6215	29	11	approach	approach	NOUN
cana-6215	29	12	that	that	PRON
cana-6215	29	13	allows	allow	VERB
cana-6215	29	14	models	model	NOUN
cana-6215	29	15	pre	pre	ADJ
cana-6215	29	16	-	-	VERB
cana-6215	29	17	trained	train	VERB
cana-6215	29	18	on	on	ADP
cana-6215	29	19	large	large	ADJ
cana-6215	29	20	benchmark	benchmark	ADJ
cana-6215	29	21	datasets	dataset	NOUN
cana-6215	29	22	,	,	PUNCT
cana-6215	29	23	such	such	ADJ
cana-6215	29	24	as	as	ADP
cana-6215	29	25	imagenet	imagenet	NOUN
cana-6215	29	26	,	,	PUNCT
cana-6215	29	27	to	to	PART
cana-6215	29	28	be	be	AUX
cana-6215	29	29	fine	fine	ADV
cana-6215	29	30	-	-	PUNCT
cana-6215	29	31	tuned	tune	VERB
cana-6215	29	32	for	for	ADP
cana-6215	29	33	specific	specific	ADJ
cana-6215	29	34	medical	medical	ADJ
cana-6215	29	35	imaging	imaging	NOUN
cana-6215	29	36	tasks	task	NOUN
cana-6215	29	37	.	.	PUNCT
cana-6215	30	1	transfer	transfer	NOUN
cana-6215	30	2	learning	learning	NOUN
cana-6215	30	3	enables	enable	VERB
cana-6215	30	4	the	the	DET
cana-6215	30	5	reuse	reuse	NOUN
cana-6215	30	6	of	of	ADP
cana-6215	30	7	learned	learn	VERB
cana-6215	30	8	features	feature	NOUN
cana-6215	30	9	,	,	PUNCT
cana-6215	30	10	significantly	significantly	ADV
cana-6215	30	11	reducing	reduce	VERB
cana-6215	30	12	training	training	NOUN
cana-6215	30	13	time	time	NOUN
cana-6215	30	14	and	and	CCONJ
cana-6215	30	15	improving	improve	VERB
cana-6215	30	16	model	model	NOUN
cana-6215	30	17	performance	performance	NOUN
cana-6215	30	18	even	even	ADV
cana-6215	30	19	with	with	ADP
cana-6215	30	20	limited	limited	ADJ
cana-6215	30	21	data	datum	NOUN
cana-6215	30	22	availability	availability	NOUN
cana-6215	30	23	.	.	PUNCT
cana-6215	31	1	in	in	ADP
cana-6215	31	2	the	the	DET
cana-6215	31	3	context	context	NOUN
cana-6215	31	4	of	of	ADP
cana-6215	31	5	liver	liver	NOUN
cana-6215	31	6	cancer	cancer	NOUN
cana-6215	31	7	classification	classification	NOUN
cana-6215	31	8	,	,	PUNCT
cana-6215	31	9	tl	tl	PROPN
cana-6215	31	10	-	-	PUNCT
cana-6215	31	11	based	base	VERB
cana-6215	31	12	models	model	NOUN
cana-6215	31	13	can	can	AUX
cana-6215	31	14	effectively	effectively	ADV
cana-6215	31	15	extract	extract	VERB
cana-6215	31	16	deep	deep	ADJ
cana-6215	31	17	,	,	PUNCT
cana-6215	31	18	discriminative	discriminative	NOUN
cana-6215	31	19	features	feature	VERB
cana-6215	31	20	from	from	ADP
cana-6215	31	21	ct	ct	NUM
cana-6215	31	22	images	image	NOUN
cana-6215	31	23	and	and	CCONJ
cana-6215	31	24	enhance	enhance	VERB
cana-6215	31	25	the	the	DET
cana-6215	31	26	diagnostic	diagnostic	ADJ
cana-6215	31	27	precision	precision	NOUN
cana-6215	31	28	of	of	ADP
cana-6215	31	29	automated	automate	VERB
cana-6215	31	30	systems	system	NOUN
cana-6215	31	31	.	.	PUNCT
cana-6215	32	1	2	2	X
cana-6215	32	2	.	.	X
cana-6215	32	3	review	review	NOUN
cana-6215	32	4	of	of	ADP
cana-6215	32	5	literature	literature	NOUN
cana-6215	32	6	recent	recent	ADJ
cana-6215	32	7	advancements	advancement	NOUN
cana-6215	32	8	in	in	ADP
cana-6215	32	9	artificial	artificial	ADJ
cana-6215	32	10	intelligence	intelligence	NOUN
cana-6215	32	11	have	have	AUX
cana-6215	32	12	greatly	greatly	ADV
cana-6215	32	13	enhanced	enhance	VERB
cana-6215	32	14	medical	medical	ADJ
cana-6215	32	15	image	image	NOUN
cana-6215	32	16	analysis	analysis	NOUN
cana-6215	32	17	,	,	PUNCT
cana-6215	32	18	particularly	particularly	ADV
cana-6215	32	19	in	in	ADP
cana-6215	32	20	the	the	DET
cana-6215	32	21	detection	detection	NOUN
cana-6215	32	22	and	and	CCONJ
cana-6215	32	23	classification	classification	NOUN
cana-6215	32	24	of	of	ADP
cana-6215	32	25	liver	liver	NOUN
cana-6215	32	26	cancer	cancer	NOUN
cana-6215	32	27	.	.	PUNCT
cana-6215	33	1	early	early	ADJ
cana-6215	33	2	studies	study	NOUN
cana-6215	33	3	utilizing	utilize	VERB
cana-6215	33	4	convolutional	convolutional	ADJ
cana-6215	33	5	neural	neural	ADJ
cana-6215	33	6	networks	network	NOUN
cana-6215	33	7	(	(	PUNCT
cana-6215	33	8	cnns	cnns	PROPN
cana-6215	33	9	)	)	PUNCT
cana-6215	33	10	demonstrated	demonstrate	VERB
cana-6215	33	11	significant	significant	ADJ
cana-6215	33	12	improvements	improvement	NOUN
cana-6215	33	13	in	in	ADP
cana-6215	33	14	tumor	tumor	NOUN
cana-6215	33	15	segmentation	segmentation	NOUN
cana-6215	33	16	and	and	CCONJ
cana-6215	33	17	classification	classification	NOUN
cana-6215	33	18	accuracy	accuracy	NOUN
cana-6215	33	19	,	,	PUNCT
cana-6215	33	20	enabling	enable	VERB
cana-6215	33	21	precise	precise	ADJ
cana-6215	33	22	differentiation	differentiation	NOUN
cana-6215	33	23	between	between	ADP
cana-6215	33	24	malignant	malignant	ADJ
cana-6215	33	25	and	and	CCONJ
cana-6215	33	26	benign	benign	ADJ
cana-6215	33	27	lesions	lesion	NOUN
cana-6215	33	28	from	from	ADP
cana-6215	33	29	computed	computed	ADJ
cana-6215	33	30	tomography	tomography	NOUN
cana-6215	33	31	(	(	PUNCT
cana-6215	33	32	ct	ct	NOUN
cana-6215	33	33	)	)	PUNCT
cana-6215	33	34	images	image	NOUN
cana-6215	33	35	[	[	X
cana-6215	33	36	1	1	NUM
cana-6215	33	37	]	]	PUNCT
cana-6215	33	38	.	.	PUNCT
cana-6215	34	1	with	with	ADP
cana-6215	34	2	the	the	DET
cana-6215	34	3	introduction	introduction	NOUN
cana-6215	34	4	of	of	ADP
cana-6215	34	5	deep	deep	ADJ
cana-6215	34	6	transfer	transfer	NOUN
cana-6215	34	7	learning	learning	NOUN
cana-6215	34	8	(	(	PUNCT
cana-6215	34	9	tl	tl	PROPN
cana-6215	34	10	)	)	PUNCT
cana-6215	34	11	,	,	PUNCT
cana-6215	34	12	pre	pre	ADJ
cana-6215	34	13	-	-	ADJ
cana-6215	34	14	trained	train	VERB
cana-6215	34	15	models	model	NOUN
cana-6215	34	16	such	such	ADJ
cana-6215	34	17	as	as	ADP
cana-6215	34	18	vgg-16	vgg-16	NOUN
cana-6215	34	19	,	,	PUNCT
cana-6215	34	20	resnet-50	resnet-50	PROPN
cana-6215	34	21	,	,	PUNCT
cana-6215	34	22	and	and	CCONJ
cana-6215	34	23	mobilenetv2	mobilenetv2	PROPN
cana-6215	34	24	have	have	AUX
cana-6215	34	25	been	be	AUX
cana-6215	34	26	widely	widely	ADV
cana-6215	34	27	employed	employ	VERB
cana-6215	34	28	to	to	PART
cana-6215	34	29	address	address	VERB
cana-6215	34	30	the	the	DET
cana-6215	34	31	challenge	challenge	NOUN
cana-6215	34	32	of	of	ADP
cana-6215	34	33	limited	limited	ADJ
cana-6215	34	34	annotated	annotate	VERB
cana-6215	34	35	medical	medical	ADJ
cana-6215	34	36	datasets	dataset	NOUN
cana-6215	35	1	[	[	X
cana-6215	35	2	2	2	NUM
cana-6215	35	3	]	]	PUNCT
cana-6215	35	4	.	.	PUNCT
cana-6215	36	1	these	these	DET
cana-6215	36	2	tl	tl	PROPN
cana-6215	36	3	-	-	PUNCT
cana-6215	36	4	based	base	VERB
cana-6215	36	5	models	model	NOUN
cana-6215	36	6	effectively	effectively	ADV
cana-6215	36	7	reuse	reuse	VERB
cana-6215	36	8	learned	learn	VERB
cana-6215	36	9	features	feature	NOUN
cana-6215	36	10	from	from	ADP
cana-6215	36	11	large	large	ADJ
cana-6215	36	12	-	-	PUNCT
cana-6215	36	13	scale	scale	NOUN
cana-6215	36	14	datasets	dataset	NOUN
cana-6215	36	15	and	and	CCONJ
cana-6215	36	16	adapt	adapt	VERB
cana-6215	36	17	them	they	PRON
cana-6215	36	18	for	for	ADP
cana-6215	36	19	liver	liver	NOUN
cana-6215	36	20	cancer	cancer	NOUN
cana-6215	36	21	classification	classification	NOUN
cana-6215	36	22	,	,	PUNCT
cana-6215	36	23	achieving	achieve	VERB
cana-6215	36	24	higher	high	ADJ
cana-6215	36	25	accuracy	accuracy	NOUN
cana-6215	36	26	,	,	PUNCT
cana-6215	36	27	sensitivity	sensitivity	NOUN
cana-6215	36	28	,	,	PUNCT
cana-6215	36	29	and	and	CCONJ
cana-6215	36	30	specificity	specificity	NOUN
cana-6215	36	31	compared	compare	VERB
cana-6215	36	32	to	to	ADP
cana-6215	36	33	conventional	conventional	ADJ
cana-6215	36	34	machine	machine	NOUN
cana-6215	36	35	learning	learning	NOUN
cana-6215	36	36	methods	method	NOUN
cana-6215	36	37	[	[	X
cana-6215	36	38	3	3	NUM
cana-6215	36	39	]	]	PUNCT
cana-6215	36	40	.	.	PUNCT
cana-6215	37	1	hybrid	hybrid	ADJ
cana-6215	37	2	deep	deep	ADJ
cana-6215	37	3	learning	learning	NOUN
cana-6215	37	4	approaches	approach	NOUN
cana-6215	37	5	that	that	PRON
cana-6215	37	6	integrate	integrate	VERB
cana-6215	37	7	cnn	cnn	PROPN
cana-6215	37	8	with	with	ADP
cana-6215	37	9	support	support	NOUN
cana-6215	37	10	vector	vector	NOUN
cana-6215	37	11	machines	machine	NOUN
cana-6215	37	12	(	(	PUNCT
cana-6215	37	13	svms	svms	NOUN
cana-6215	37	14	)	)	PUNCT
cana-6215	37	15	have	have	AUX
cana-6215	37	16	further	far	ADV
cana-6215	37	17	improved	improve	VERB
cana-6215	37	18	classification	classification	NOUN
cana-6215	37	19	performance	performance	NOUN
cana-6215	37	20	by	by	ADP
cana-6215	37	21	combining	combine	VERB
cana-6215	37	22	strong	strong	ADJ
cana-6215	37	23	feature	feature	NOUN
cana-6215	37	24	extraction	extraction	NOUN
cana-6215	37	25	with	with	ADP
cana-6215	37	26	efficient	efficient	ADJ
cana-6215	37	27	decision	decision	NOUN
cana-6215	37	28	-	-	PUNCT
cana-6215	37	29	making	make	VERB
cana-6215	37	30	processes	process	NOUN
cana-6215	37	31	[	[	X
cana-6215	37	32	4	4	NUM
cana-6215	37	33	]	]	PUNCT
cana-6215	37	34	.	.	PUNCT
cana-6215	38	1	lightweight	lightweight	ADJ
cana-6215	38	2	architectures	architecture	NOUN
cana-6215	38	3	such	such	ADJ
cana-6215	38	4	as	as	ADP
cana-6215	38	5	mobilenetv2	mobilenetv2	PROPN
cana-6215	38	6	have	have	VERB
cana-6215	38	7	communications	communication	NOUN
cana-6215	38	8	on	on	ADP
cana-6215	38	9	applied	apply	VERB
cana-6215	38	10	nonlinear	nonlinear	ADJ
cana-6215	38	11	analysis	analysis	NOUN
cana-6215	38	12	issn	issn	NOUN
cana-6215	38	13	:	:	PUNCT
cana-6215	38	14	1074	1074	NUM
cana-6215	38	15	-	-	PUNCT
cana-6215	38	16	133x	133x	NUM
cana-6215	38	17	vol	vol	NOUN
cana-6215	38	18	32	32	NUM
cana-6215	38	19	no	no	NOUN
cana-6215	38	20	.	.	PUNCT
cana-6215	39	1	4s	4s	NUM
cana-6215	39	2	(	(	PUNCT
cana-6215	39	3	2025	2025	NUM
cana-6215	39	4	)	)	PUNCT
cana-6215	39	5	775	775	NUM
cana-6215	39	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-6215	39	7	proven	prove	VERB
cana-6215	39	8	suitable	suitable	ADJ
cana-6215	39	9	for	for	ADP
cana-6215	39	10	real	real	ADJ
cana-6215	39	11	-	-	PUNCT
cana-6215	39	12	time	time	NOUN
cana-6215	39	13	clinical	clinical	ADJ
cana-6215	39	14	applications	application	NOUN
cana-6215	39	15	due	due	ADP
cana-6215	39	16	to	to	ADP
cana-6215	39	17	their	their	PRON
cana-6215	39	18	reduced	reduced	ADJ
cana-6215	39	19	computational	computational	ADJ
cana-6215	39	20	complexity	complexity	NOUN
cana-6215	39	21	and	and	CCONJ
cana-6215	39	22	faster	fast	ADJ
cana-6215	39	23	inference	inference	NOUN
cana-6215	39	24	speed	speed	NOUN
cana-6215	39	25	while	while	SCONJ
cana-6215	39	26	maintaining	maintain	VERB
cana-6215	39	27	high	high	ADJ
cana-6215	39	28	classification	classification	NOUN
cana-6215	39	29	accuracy	accuracy	NOUN
cana-6215	39	30	[	[	X
cana-6215	39	31	5	5	NUM
cana-6215	39	32	]	]	PUNCT
cana-6215	39	33	.	.	PUNCT
cana-6215	40	1	moreover	moreover	ADV
cana-6215	40	2	,	,	PUNCT
cana-6215	40	3	residual	residual	ADJ
cana-6215	40	4	learning	learning	NOUN
cana-6215	40	5	models	model	NOUN
cana-6215	40	6	like	like	ADP
cana-6215	40	7	resnet	resnet	NOUN
cana-6215	40	8	have	have	AUX
cana-6215	40	9	shown	show	VERB
cana-6215	40	10	excellent	excellent	ADJ
cana-6215	40	11	generalization	generalization	NOUN
cana-6215	40	12	and	and	CCONJ
cana-6215	40	13	sensitivity	sensitivity	NOUN
cana-6215	40	14	by	by	ADP
cana-6215	40	15	addressing	address	VERB
cana-6215	40	16	vanishing	vanish	VERB
cana-6215	40	17	gradient	gradient	NOUN
cana-6215	40	18	problems	problem	NOUN
cana-6215	40	19	during	during	ADP
cana-6215	40	20	training	training	NOUN
cana-6215	40	21	[	[	X
cana-6215	40	22	6	6	NUM
cana-6215	40	23	]	]	PUNCT
cana-6215	40	24	.	.	PUNCT
cana-6215	41	1	three	three	NUM
cana-6215	41	2	-	-	PUNCT
cana-6215	41	3	dimensional	dimensional	ADJ
cana-6215	41	4	cnn	cnn	PROPN
cana-6215	41	5	architectures	architecture	NOUN
cana-6215	41	6	have	have	AUX
cana-6215	41	7	also	also	ADV
cana-6215	41	8	emerged	emerge	VERB
cana-6215	41	9	as	as	ADP
cana-6215	41	10	powerful	powerful	ADJ
cana-6215	41	11	tools	tool	NOUN
cana-6215	41	12	for	for	ADP
cana-6215	41	13	volumetric	volumetric	ADJ
cana-6215	41	14	liver	liver	NOUN
cana-6215	41	15	image	image	NOUN
cana-6215	41	16	analysis	analysis	NOUN
cana-6215	41	17	,	,	PUNCT
cana-6215	41	18	providing	provide	VERB
cana-6215	41	19	enhanced	enhanced	ADJ
cana-6215	41	20	precision	precision	NOUN
cana-6215	41	21	in	in	ADP
cana-6215	41	22	tumor	tumor	NOUN
cana-6215	41	23	boundary	boundary	ADJ
cana-6215	41	24	identification	identification	NOUN
cana-6215	41	25	and	and	CCONJ
cana-6215	41	26	lesion	lesion	NOUN
cana-6215	41	27	classification	classification	NOUN
cana-6215	41	28	[	[	X
cana-6215	41	29	7	7	NUM
cana-6215	41	30	]	]	PUNCT
cana-6215	41	31	.	.	PUNCT
cana-6215	42	1	recent	recent	ADJ
cana-6215	42	2	studies	study	NOUN
cana-6215	42	3	on	on	ADP
cana-6215	42	4	ensemble	ensemble	ADJ
cana-6215	42	5	deep	deep	ADJ
cana-6215	42	6	learning	learning	NOUN
cana-6215	42	7	frameworks	framework	NOUN
cana-6215	42	8	,	,	PUNCT
cana-6215	42	9	combining	combine	VERB
cana-6215	42	10	models	model	NOUN
cana-6215	42	11	such	such	ADJ
cana-6215	42	12	as	as	ADP
cana-6215	42	13	cnn	cnn	PROPN
cana-6215	42	14	and	and	CCONJ
cana-6215	42	15	densenet	densenet	NOUN
cana-6215	42	16	,	,	PUNCT
cana-6215	42	17	have	have	AUX
cana-6215	42	18	outperformed	outperform	VERB
cana-6215	42	19	single	single	ADJ
cana-6215	42	20	-	-	PUNCT
cana-6215	42	21	model	model	NOUN
cana-6215	42	22	architectures	architecture	NOUN
cana-6215	42	23	,	,	PUNCT
cana-6215	42	24	delivering	deliver	VERB
cana-6215	42	25	improved	improve	VERB
cana-6215	42	26	diagnostic	diagnostic	ADJ
cana-6215	42	27	accuracy	accuracy	NOUN
cana-6215	42	28	and	and	CCONJ
cana-6215	42	29	robust	robust	ADJ
cana-6215	42	30	performance	performance	NOUN
cana-6215	42	31	across	across	ADP
cana-6215	42	32	heterogeneous	heterogeneous	ADJ
cana-6215	42	33	datasets	dataset	NOUN
cana-6215	42	34	[	[	X
cana-6215	42	35	8	8	NUM
cana-6215	42	36	]	]	PUNCT
cana-6215	42	37	.	.	PUNCT
cana-6215	43	1	fine	fine	ADJ
cana-6215	43	2	-	-	PUNCT
cana-6215	43	3	tuned	tune	VERB
cana-6215	43	4	transfer	transfer	NOUN
cana-6215	43	5	learning	learning	NOUN
cana-6215	43	6	techniques	technique	NOUN
cana-6215	43	7	have	have	AUX
cana-6215	43	8	further	far	ADV
cana-6215	43	9	demonstrated	demonstrate	VERB
cana-6215	43	10	the	the	DET
cana-6215	43	11	ability	ability	NOUN
cana-6215	43	12	to	to	PART
cana-6215	43	13	work	work	VERB
cana-6215	43	14	effectively	effectively	ADV
cana-6215	43	15	with	with	ADP
cana-6215	43	16	small	small	ADJ
cana-6215	43	17	medical	medical	ADJ
cana-6215	43	18	datasets	dataset	NOUN
cana-6215	43	19	,	,	PUNCT
cana-6215	43	20	reducing	reduce	VERB
cana-6215	43	21	overfitting	overfitting	NOUN
cana-6215	43	22	while	while	SCONJ
cana-6215	43	23	enhancing	enhance	VERB
cana-6215	43	24	model	model	NOUN
cana-6215	43	25	reliability	reliability	NOUN
cana-6215	43	26	[	[	X
cana-6215	43	27	2	2	NUM
cana-6215	43	28	]	]	PUNCT
cana-6215	43	29	,	,	PUNCT
cana-6215	43	30	[	[	X
cana-6215	43	31	7	7	NUM
cana-6215	43	32	]	]	PUNCT
cana-6215	43	33	.	.	PUNCT
cana-6215	44	1	overall	overall	ADV
cana-6215	44	2	,	,	PUNCT
cana-6215	44	3	the	the	DET
cana-6215	44	4	reviewed	review	VERB
cana-6215	44	5	literature	literature	NOUN
cana-6215	44	6	strongly	strongly	ADV
cana-6215	44	7	supports	support	VERB
cana-6215	44	8	the	the	DET
cana-6215	44	9	potential	potential	NOUN
cana-6215	44	10	of	of	ADP
cana-6215	44	11	deep	deep	ADJ
cana-6215	44	12	learning	learning	NOUN
cana-6215	44	13	and	and	CCONJ
cana-6215	44	14	transfer	transfer	VERB
cana-6215	44	15	learning	learning	NOUN
cana-6215	44	16	methods	method	NOUN
cana-6215	44	17	to	to	PART
cana-6215	44	18	transform	transform	VERB
cana-6215	44	19	liver	liver	NOUN
cana-6215	44	20	cancer	cancer	NOUN
cana-6215	44	21	diagnosis	diagnosis	NOUN
cana-6215	44	22	through	through	ADP
cana-6215	44	23	automated	automate	VERB
cana-6215	44	24	,	,	PUNCT
cana-6215	44	25	accurate	accurate	ADJ
cana-6215	44	26	,	,	PUNCT
cana-6215	44	27	and	and	CCONJ
cana-6215	44	28	efficient	efficient	ADJ
cana-6215	44	29	imagebased	imagebase	VERB
cana-6215	44	30	classification	classification	NOUN
cana-6215	44	31	systems	system	NOUN
cana-6215	44	32	that	that	PRON
cana-6215	44	33	can	can	AUX
cana-6215	44	34	significantly	significantly	ADV
cana-6215	44	35	assist	assist	VERB
cana-6215	44	36	radiologists	radiologist	NOUN
cana-6215	44	37	in	in	ADP
cana-6215	44	38	clinical	clinical	ADJ
cana-6215	44	39	decisionmaking	decisionmaking	NOUN
cana-6215	44	40	and	and	CCONJ
cana-6215	44	41	early	early	ADJ
cana-6215	44	42	detection	detection	NOUN
cana-6215	44	43	of	of	ADP
cana-6215	44	44	liver	liver	NOUN
cana-6215	44	45	abnormalities	abnormality	NOUN
cana-6215	44	46	.	.	PUNCT
cana-6215	45	1	the	the	DET
cana-6215	45	2	review	review	NOUN
cana-6215	45	3	of	of	ADP
cana-6215	45	4	literature	literature	NOUN
cana-6215	45	5	shown	show	VERB
cana-6215	45	6	in	in	ADP
cana-6215	45	7	table	table	NOUN
cana-6215	45	8	1	1	NUM
cana-6215	45	9	.	.	PUNCT
cana-6215	45	10	table	table	NOUN
cana-6215	45	11	1	1	NUM
cana-6215	45	12	:	:	PUNCT
cana-6215	45	13	review	review	NOUN
cana-6215	45	14	of	of	ADP
cana-6215	45	15	literature	literature	NOUN
cana-6215	45	16	on	on	ADP
cana-6215	45	17	deep	deep	ADJ
cana-6215	45	18	learning	learning	NOUN
cana-6215	45	19	and	and	CCONJ
cana-6215	45	20	transfer	transfer	VERB
cana-6215	45	21	learning	learn	VERB
cana-6215	45	22	techniques	technique	NOUN
cana-6215	45	23	for	for	ADP
cana-6215	45	24	liver	liver	NOUN
cana-6215	45	25	cancer	cancer	NOUN
cana-6215	45	26	classification	classification	NOUN
cana-6215	45	27	ref	ref	NOUN
cana-6215	45	28	.	.	PUNCT
cana-6215	46	1	algorithms	algorithm	NOUN
cana-6215	46	2	dataset	dataset	NOUN
cana-6215	46	3	performance	performance	NOUN
cana-6215	46	4	[	[	X
cana-6215	46	5	1	1	NUM
cana-6215	46	6	]	]	X
cana-6215	46	7	convolutional	convolutional	ADJ
cana-6215	46	8	neural	neural	ADJ
cana-6215	46	9	network	network	NOUN
cana-6215	46	10	(	(	PUNCT
cana-6215	46	11	cnn	cnn	PROPN
cana-6215	46	12	)	)	PUNCT
cana-6215	46	13	for	for	ADP
cana-6215	46	14	liver	liver	NOUN
cana-6215	46	15	tumor	tumor	NOUN
cana-6215	46	16	segmentation	segmentation	NOUN
cana-6215	46	17	and	and	CCONJ
cana-6215	46	18	classification	classification	NOUN
cana-6215	46	19	ct	ct	NOUN
cana-6215	46	20	liver	liver	NOUN
cana-6215	46	21	image	image	NOUN
cana-6215	46	22	dataset	dataset	NOUN
cana-6215	46	23	achieved	achieve	VERB
cana-6215	46	24	95	95	NUM
cana-6215	46	25	%	%	NOUN
cana-6215	46	26	classification	classification	NOUN
cana-6215	46	27	accuracy	accuracy	NOUN
cana-6215	46	28	;	;	PUNCT
cana-6215	46	29	effective	effective	ADJ
cana-6215	46	30	in	in	ADP
cana-6215	46	31	differentiating	differentiate	VERB
cana-6215	46	32	malignant	malignant	ADJ
cana-6215	46	33	and	and	CCONJ
cana-6215	46	34	benign	benign	ADJ
cana-6215	46	35	lesions	lesion	NOUN
cana-6215	46	36	.	.	PUNCT
cana-6215	47	1	[	[	X
cana-6215	47	2	2	2	NUM
cana-6215	47	3	]	]	X
cana-6215	47	4	deep	deep	ADJ
cana-6215	47	5	transfer	transfer	NOUN
cana-6215	47	6	learning	learn	VERB
cana-6215	47	7	using	use	VERB
cana-6215	47	8	resnet-50	resnet-50	PROPN
cana-6215	47	9	and	and	CCONJ
cana-6215	47	10	vgg-19	vgg-19	NUM
cana-6215	47	11	models	model	NOUN
cana-6215	47	12	public	public	ADJ
cana-6215	47	13	liver	liver	NOUN
cana-6215	47	14	cancer	cancer	NOUN
cana-6215	47	15	imaging	imaging	NOUN
cana-6215	47	16	dataset	dataset	NOUN
cana-6215	47	17	transfer	transfer	NOUN
cana-6215	47	18	learning	learn	VERB
cana-6215	47	19	enhanced	enhance	VERB
cana-6215	47	20	accuracy	accuracy	NOUN
cana-6215	47	21	and	and	CCONJ
cana-6215	47	22	reduced	reduced	ADJ
cana-6215	47	23	training	training	NOUN
cana-6215	47	24	time	time	NOUN
cana-6215	47	25	for	for	ADP
cana-6215	47	26	limited	limited	ADJ
cana-6215	47	27	medical	medical	ADJ
cana-6215	47	28	datasets	dataset	NOUN
cana-6215	47	29	.	.	PUNCT
cana-6215	48	1	[	[	X
cana-6215	48	2	3	3	NUM
cana-6215	48	3	]	]	X
cana-6215	48	4	hybrid	hybrid	ADJ
cana-6215	48	5	cnn	cnn	PROPN
cana-6215	48	6	-	-	PUNCT
cana-6215	48	7	svm	svm	PROPN
cana-6215	48	8	model	model	NOUN
cana-6215	48	9	for	for	ADP
cana-6215	48	10	medical	medical	ADJ
cana-6215	48	11	image	image	NOUN
cana-6215	48	12	classification	classification	NOUN
cana-6215	48	13	hospital	hospital	NOUN
cana-6215	48	14	-	-	PUNCT
cana-6215	48	15	based	base	VERB
cana-6215	48	16	ct	ct	PROPN
cana-6215	48	17	scan	scan	PROPN
cana-6215	48	18	dataset	dataset	VERB
cana-6215	48	19	improved	improve	VERB
cana-6215	48	20	feature	feature	NOUN
cana-6215	48	21	extraction	extraction	NOUN
cana-6215	48	22	and	and	CCONJ
cana-6215	48	23	achieved	achieve	VERB
cana-6215	48	24	97	97	NUM
cana-6215	48	25	%	%	NOUN
cana-6215	48	26	classification	classification	NOUN
cana-6215	48	27	accuracy	accuracy	NOUN
cana-6215	48	28	compared	compare	VERB
cana-6215	48	29	to	to	ADP
cana-6215	48	30	traditional	traditional	ADJ
cana-6215	48	31	ml	ml	NOUN
cana-6215	48	32	models	model	NOUN
cana-6215	48	33	.	.	PUNCT
cana-6215	49	1	[	[	X
cana-6215	49	2	4	4	X
cana-6215	49	3	]	]	X
cana-6215	49	4	mobilenetv2	mobilenetv2	PROPN
cana-6215	49	5	-	-	PUNCT
cana-6215	49	6	based	base	VERB
cana-6215	49	7	model	model	NOUN
cana-6215	49	8	for	for	ADP
cana-6215	49	9	liver	liver	NOUN
cana-6215	49	10	tumor	tumor	NOUN
cana-6215	49	11	detection	detection	NOUN
cana-6215	49	12	baghdad	baghdad	PROPN
cana-6215	49	13	medical	medical	ADJ
cana-6215	49	14	city	city	PROPN
cana-6215	49	15	,	,	PUNCT
cana-6215	49	16	iraq	iraq	PROPN
cana-6215	49	17	model	model	NOUN
cana-6215	49	18	demonstrated	demonstrate	VERB
cana-6215	49	19	high	high	ADJ
cana-6215	49	20	efficiency	efficiency	NOUN
cana-6215	49	21	with	with	ADP
cana-6215	49	22	98.6	98.6	NUM
cana-6215	49	23	%	%	NOUN
cana-6215	49	24	accuracy	accuracy	NOUN
cana-6215	49	25	and	and	CCONJ
cana-6215	49	26	faster	fast	ADJ
cana-6215	49	27	computation	computation	NOUN
cana-6215	49	28	time	time	NOUN
cana-6215	49	29	.	.	PUNCT
cana-6215	50	1	[	[	X
cana-6215	50	2	5	5	NUM
cana-6215	50	3	]	]	X
cana-6215	50	4	deep	deep	ADJ
cana-6215	50	5	residual	residual	ADJ
cana-6215	50	6	learning	learning	NOUN
cana-6215	50	7	(	(	PUNCT
cana-6215	50	8	resnet	resnet	NOUN
cana-6215	50	9	)	)	PUNCT
cana-6215	50	10	for	for	ADP
cana-6215	50	11	automated	automate	VERB
cana-6215	50	12	liver	liver	NOUN
cana-6215	50	13	lesion	lesion	NOUN
cana-6215	50	14	classification	classification	NOUN
cana-6215	50	15	liver	liver	NOUN
cana-6215	50	16	tumor	tumor	NOUN
cana-6215	50	17	segmentation	segmentation	NOUN
cana-6215	50	18	(	(	PUNCT
cana-6215	50	19	lits	lit	NOUN
cana-6215	50	20	)	)	PUNCT
cana-6215	50	21	challenge	challenge	NOUN
cana-6215	50	22	dataset	dataset	NOUN
cana-6215	50	23	achieved	achieve	VERB
cana-6215	50	24	96	96	NUM
cana-6215	50	25	%	%	NOUN
cana-6215	50	26	sensitivity	sensitivity	NOUN
cana-6215	50	27	and	and	CCONJ
cana-6215	50	28	improved	improved	ADJ
cana-6215	50	29	model	model	NOUN
cana-6215	50	30	generalization	generalization	NOUN
cana-6215	50	31	through	through	ADP
cana-6215	50	32	data	datum	NOUN
cana-6215	50	33	augmentation	augmentation	NOUN
cana-6215	50	34	.	.	PUNCT
cana-6215	51	1	[	[	X
cana-6215	51	2	6	6	NUM
cana-6215	51	3	]	]	PUNCT
cana-6215	51	4	3d	3d	PROPN
cana-6215	51	5	cnn	cnn	PROPN
cana-6215	51	6	model	model	NOUN
cana-6215	51	7	for	for	ADP
cana-6215	51	8	volumetric	volumetric	ADJ
cana-6215	51	9	liver	liver	NOUN
cana-6215	51	10	tumor	tumor	NOUN
cana-6215	51	11	classification	classification	NOUN
cana-6215	51	12	lits	lit	NOUN
cana-6215	51	13	and	and	CCONJ
cana-6215	51	14	miccai	miccai	NOUN
cana-6215	51	15	datasets	dataset	NOUN
cana-6215	51	16	provided	provide	VERB
cana-6215	51	17	higher	high	ADJ
cana-6215	51	18	precision	precision	NOUN
cana-6215	51	19	in	in	ADP
cana-6215	51	20	tumor	tumor	NOUN
cana-6215	51	21	boundary	boundary	ADJ
cana-6215	51	22	identification	identification	NOUN
cana-6215	51	23	and	and	CCONJ
cana-6215	51	24	classification	classification	NOUN
cana-6215	51	25	.	.	PUNCT
cana-6215	52	1	[	[	X
cana-6215	52	2	7	7	NUM
cana-6215	52	3	]	]	ADJ
cana-6215	52	4	transfer	transfer	NOUN
cana-6215	52	5	learning	learning	NOUN
cana-6215	52	6	with	with	ADP
cana-6215	52	7	fine	fine	ADV
cana-6215	52	8	-	-	PUNCT
cana-6215	52	9	tuned	tune	VERB
cana-6215	52	10	vgg-16	vgg-16	NOUN
cana-6215	52	11	architecture	architecture	NOUN
cana-6215	52	12	private	private	ADJ
cana-6215	52	13	ct	ct	PROPN
cana-6215	52	14	liver	liver	NOUN
cana-6215	52	15	dataset	dataset	NOUN
cana-6215	52	16	demonstrated	demonstrate	VERB
cana-6215	52	17	99	99	NUM
cana-6215	52	18	%	%	NOUN
cana-6215	52	19	diagnostic	diagnostic	ADJ
cana-6215	52	20	accuracy	accuracy	NOUN
cana-6215	52	21	and	and	CCONJ
cana-6215	52	22	reduced	reduced	ADJ
cana-6215	52	23	overfitting	overfitting	NOUN
cana-6215	52	24	in	in	ADP
cana-6215	52	25	small	small	ADJ
cana-6215	52	26	datasets	dataset	NOUN
cana-6215	52	27	.	.	PUNCT
cana-6215	53	1	communications	communication	NOUN
cana-6215	53	2	on	on	ADP
cana-6215	53	3	applied	apply	VERB
cana-6215	53	4	nonlinear	nonlinear	ADJ
cana-6215	53	5	analysis	analysis	NOUN
cana-6215	53	6	issn	issn	NOUN
cana-6215	53	7	:	:	PUNCT
cana-6215	53	8	1074	1074	NUM
cana-6215	53	9	-	-	PUNCT
cana-6215	53	10	133x	133x	NUM
cana-6215	53	11	vol	vol	NOUN
cana-6215	53	12	32	32	NUM
cana-6215	53	13	no	no	NOUN
cana-6215	53	14	.	.	PUNCT
cana-6215	54	1	4s	4s	NUM
cana-6215	54	2	(	(	PUNCT
cana-6215	54	3	2025	2025	NUM
cana-6215	54	4	)	)	PUNCT
cana-6215	54	5	776	776	NUM
cana-6215	54	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-6215	55	1	[	[	X
cana-6215	55	2	8	8	X
cana-6215	55	3	]	]	X
cana-6215	55	4	ensemble	ensemble	ADJ
cana-6215	55	5	deep	deep	ADJ
cana-6215	55	6	learning	learning	NOUN
cana-6215	55	7	combining	combine	VERB
cana-6215	55	8	cnn	cnn	PROPN
cana-6215	55	9	and	and	CCONJ
cana-6215	55	10	densenet	densenet	NOUN
cana-6215	55	11	uci	uci	PROPN
cana-6215	55	12	and	and	CCONJ
cana-6215	55	13	clinical	clinical	ADJ
cana-6215	55	14	imaging	imaging	NOUN
cana-6215	55	15	data	datum	NOUN
cana-6215	55	16	outperformed	outperform	VERB
cana-6215	55	17	individual	individual	ADJ
cana-6215	55	18	dl	dl	PROPN
cana-6215	55	19	models	model	NOUN
cana-6215	55	20	with	with	ADP
cana-6215	55	21	98.5	98.5	NUM
cana-6215	55	22	%	%	NOUN
cana-6215	55	23	accuracy	accuracy	NOUN
cana-6215	55	24	and	and	CCONJ
cana-6215	55	25	high	high	ADJ
cana-6215	55	26	f1	f1	NOUN
cana-6215	55	27	-	-	PUNCT
cana-6215	55	28	score	score	NOUN
cana-6215	55	29	.	.	PUNCT
cana-6215	56	1	3	3	X
cana-6215	56	2	.	.	X
cana-6215	56	3	transfer	transfer	NOUN
cana-6215	56	4	learning	learning	NOUN
cana-6215	56	5	transfer	transfer	NOUN
cana-6215	56	6	learning	learning	NOUN
cana-6215	56	7	(	(	PUNCT
cana-6215	56	8	tl	tl	PROPN
cana-6215	56	9	)	)	PUNCT
cana-6215	56	10	is	be	AUX
cana-6215	56	11	a	a	DET
cana-6215	56	12	powerful	powerful	ADJ
cana-6215	56	13	deep	deep	ADJ
cana-6215	56	14	learning	learning	NOUN
cana-6215	56	15	technique	technique	NOUN
cana-6215	56	16	that	that	PRON
cana-6215	56	17	leverages	leverage	VERB
cana-6215	56	18	knowledge	knowledge	NOUN
cana-6215	56	19	gained	gain	VERB
cana-6215	56	20	from	from	ADP
cana-6215	56	21	a	a	DET
cana-6215	56	22	large	large	ADJ
cana-6215	56	23	,	,	PUNCT
cana-6215	56	24	pre	pre	ADJ
cana-6215	56	25	-	-	ADJ
cana-6215	56	26	trained	train	VERB
cana-6215	56	27	model	model	NOUN
cana-6215	56	28	on	on	ADP
cana-6215	56	29	a	a	DET
cana-6215	56	30	related	relate	VERB
cana-6215	56	31	task	task	NOUN
cana-6215	56	32	and	and	CCONJ
cana-6215	56	33	applies	apply	VERB
cana-6215	56	34	it	it	PRON
cana-6215	56	35	to	to	ADP
cana-6215	56	36	a	a	DET
cana-6215	56	37	new	new	ADJ
cana-6215	56	38	,	,	PUNCT
cana-6215	56	39	smaller	small	ADJ
cana-6215	56	40	dataset	dataset	NOUN
cana-6215	56	41	.	.	PUNCT
cana-6215	57	1	instead	instead	ADV
cana-6215	57	2	of	of	ADP
cana-6215	57	3	training	train	VERB
cana-6215	57	4	a	a	DET
cana-6215	57	5	model	model	NOUN
cana-6215	57	6	from	from	ADP
cana-6215	57	7	scratch	scratch	NOUN
cana-6215	57	8	,	,	PUNCT
cana-6215	57	9	tl	tl	PROPN
cana-6215	57	10	fine	fine	ADJ
cana-6215	57	11	-	-	PUNCT
cana-6215	57	12	tunes	tune	NOUN
cana-6215	57	13	existing	exist	VERB
cana-6215	57	14	models	model	NOUN
cana-6215	57	15	such	such	ADJ
cana-6215	57	16	as	as	ADP
cana-6215	57	17	vgg16	vgg16	PROPN
cana-6215	57	18	,	,	PUNCT
cana-6215	57	19	resnet50	resnet50	NOUN
cana-6215	57	20	,	,	PUNCT
cana-6215	57	21	mobilenetv2	mobilenetv2	PROPN
cana-6215	57	22	,	,	PUNCT
cana-6215	57	23	inceptionv3	inceptionv3	NOUN
cana-6215	57	24	,	,	PUNCT
cana-6215	57	25	and	and	CCONJ
cana-6215	57	26	densenet121	densenet121	PROPN
cana-6215	57	27	,	,	PUNCT
cana-6215	57	28	which	which	PRON
cana-6215	57	29	have	have	AUX
cana-6215	57	30	already	already	ADV
cana-6215	57	31	learned	learn	VERB
cana-6215	57	32	general	general	ADJ
cana-6215	57	33	image	image	NOUN
cana-6215	57	34	features	feature	VERB
cana-6215	57	35	from	from	ADP
cana-6215	57	36	massive	massive	ADJ
cana-6215	57	37	datasets	dataset	NOUN
cana-6215	57	38	like	like	ADP
cana-6215	57	39	imagenet	imagenet	NOUN
cana-6215	57	40	.	.	PUNCT
cana-6215	58	1	this	this	DET
cana-6215	58	2	approach	approach	NOUN
cana-6215	58	3	significantly	significantly	ADV
cana-6215	58	4	reduces	reduce	VERB
cana-6215	58	5	training	training	NOUN
cana-6215	58	6	time	time	NOUN
cana-6215	58	7	,	,	PUNCT
cana-6215	58	8	prevents	prevent	VERB
cana-6215	58	9	overfitting	overfitte	VERB
cana-6215	58	10	,	,	PUNCT
cana-6215	58	11	and	and	CCONJ
cana-6215	58	12	enhances	enhance	VERB
cana-6215	58	13	accuracy	accuracy	NOUN
cana-6215	58	14	—	—	PUNCT
cana-6215	58	15	especially	especially	ADV
cana-6215	58	16	when	when	SCONJ
cana-6215	58	17	the	the	DET
cana-6215	58	18	available	available	ADJ
cana-6215	58	19	medical	medical	ADJ
cana-6215	58	20	imaging	imaging	NOUN
cana-6215	58	21	data	datum	NOUN
cana-6215	58	22	is	be	AUX
cana-6215	58	23	limited	limited	ADJ
cana-6215	58	24	.	.	PUNCT
cana-6215	59	1	in	in	ADP
cana-6215	59	2	this	this	DET
cana-6215	59	3	study	study	NOUN
cana-6215	59	4	,	,	PUNCT
cana-6215	59	5	tl	tl	PROPN
cana-6215	59	6	enables	enable	VERB
cana-6215	59	7	efficient	efficient	ADJ
cana-6215	59	8	feature	feature	NOUN
cana-6215	59	9	extraction	extraction	NOUN
cana-6215	59	10	and	and	CCONJ
cana-6215	59	11	accurate	accurate	ADJ
cana-6215	59	12	classification	classification	NOUN
cana-6215	59	13	of	of	ADP
cana-6215	59	14	liver	liver	NOUN
cana-6215	59	15	tumors	tumor	NOUN
cana-6215	59	16	by	by	ADP
cana-6215	59	17	adapting	adapt	VERB
cana-6215	59	18	these	these	DET
cana-6215	59	19	pre	pre	ADJ
cana-6215	59	20	-	-	ADJ
cana-6215	59	21	trained	train	VERB
cana-6215	59	22	architectures	architecture	NOUN
cana-6215	59	23	to	to	ADP
cana-6215	59	24	the	the	DET
cana-6215	59	25	specific	specific	ADJ
cana-6215	59	26	characteristics	characteristic	NOUN
cana-6215	59	27	of	of	ADP
cana-6215	59	28	ct	ct	NUM
cana-6215	59	29	images	image	NOUN
cana-6215	59	30	.	.	PUNCT
cana-6215	60	1	•	•	NUM
cana-6215	60	2	vgg16	vgg16	NOUN
cana-6215	60	3	vgg16	vgg16	NOUN
cana-6215	60	4	is	be	AUX
cana-6215	60	5	a	a	DET
cana-6215	60	6	deep	deep	ADJ
cana-6215	60	7	convolutional	convolutional	ADJ
cana-6215	60	8	neural	neural	ADJ
cana-6215	60	9	network	network	NOUN
cana-6215	60	10	known	know	VERB
cana-6215	60	11	for	for	ADP
cana-6215	60	12	its	its	PRON
cana-6215	60	13	simplicity	simplicity	NOUN
cana-6215	60	14	and	and	CCONJ
cana-6215	60	15	uniform	uniform	ADJ
cana-6215	60	16	architecture	architecture	NOUN
cana-6215	60	17	,	,	PUNCT
cana-6215	60	18	consisting	consist	VERB
cana-6215	60	19	of	of	ADP
cana-6215	60	20	16	16	NUM
cana-6215	60	21	weight	weight	NOUN
cana-6215	60	22	layers	layer	NOUN
cana-6215	60	23	.	.	PUNCT
cana-6215	61	1	it	it	PRON
cana-6215	61	2	employs	employ	VERB
cana-6215	61	3	small	small	ADJ
cana-6215	61	4	3×3	3×3	NUM
cana-6215	61	5	convolution	convolution	NOUN
cana-6215	61	6	filters	filter	NOUN
cana-6215	61	7	and	and	CCONJ
cana-6215	61	8	a	a	DET
cana-6215	61	9	consistent	consistent	ADJ
cana-6215	61	10	structure	structure	NOUN
cana-6215	61	11	throughout	throughout	ADP
cana-6215	61	12	the	the	DET
cana-6215	61	13	network	network	NOUN
cana-6215	61	14	,	,	PUNCT
cana-6215	61	15	making	make	VERB
cana-6215	61	16	it	it	PRON
cana-6215	61	17	highly	highly	ADV
cana-6215	61	18	efficient	efficient	ADJ
cana-6215	61	19	for	for	ADP
cana-6215	61	20	feature	feature	NOUN
cana-6215	61	21	extraction	extraction	NOUN
cana-6215	61	22	from	from	ADP
cana-6215	61	23	medical	medical	ADJ
cana-6215	61	24	images	image	NOUN
cana-6215	61	25	.	.	PUNCT
cana-6215	62	1	in	in	ADP
cana-6215	62	2	this	this	DET
cana-6215	62	3	study	study	NOUN
cana-6215	62	4	,	,	PUNCT
cana-6215	62	5	vgg16	vgg16	PROPN
cana-6215	62	6	was	be	AUX
cana-6215	62	7	fine	fine	ADV
cana-6215	62	8	-	-	PUNCT
cana-6215	62	9	tuned	tune	VERB
cana-6215	62	10	through	through	ADP
cana-6215	62	11	transfer	transfer	NOUN
cana-6215	62	12	learning	learn	VERB
cana-6215	62	13	to	to	PART
cana-6215	62	14	identify	identify	VERB
cana-6215	62	15	high	high	ADJ
cana-6215	62	16	-	-	PUNCT
cana-6215	62	17	level	level	NOUN
cana-6215	62	18	spatial	spatial	ADJ
cana-6215	62	19	features	feature	NOUN
cana-6215	62	20	from	from	ADP
cana-6215	62	21	liver	liver	NOUN
cana-6215	62	22	ct	ct	PROPN
cana-6215	62	23	scans	scan	NOUN
cana-6215	62	24	.	.	PUNCT
cana-6215	63	1	its	its	PRON
cana-6215	63	2	ability	ability	NOUN
cana-6215	63	3	to	to	PART
cana-6215	63	4	capture	capture	VERB
cana-6215	63	5	fine	fine	ADV
cana-6215	63	6	-	-	PUNCT
cana-6215	63	7	grained	grain	VERB
cana-6215	63	8	texture	texture	NOUN
cana-6215	63	9	and	and	CCONJ
cana-6215	63	10	edge	edge	NOUN
cana-6215	63	11	details	detail	NOUN
cana-6215	63	12	enhances	enhance	VERB
cana-6215	63	13	the	the	DET
cana-6215	63	14	classification	classification	NOUN
cana-6215	63	15	accuracy	accuracy	NOUN
cana-6215	63	16	for	for	ADP
cana-6215	63	17	differentiating	differentiate	VERB
cana-6215	63	18	between	between	ADP
cana-6215	63	19	malignant	malignant	ADJ
cana-6215	63	20	,	,	PUNCT
cana-6215	63	21	benign	benign	ADJ
cana-6215	63	22	,	,	PUNCT
cana-6215	63	23	and	and	CCONJ
cana-6215	63	24	normal	normal	ADJ
cana-6215	63	25	liver	liver	NOUN
cana-6215	63	26	tissues	tissue	NOUN
cana-6215	63	27	.	.	PUNCT
cana-6215	64	1	the	the	DET
cana-6215	64	2	model	model	NOUN
cana-6215	64	3	’s	’s	PART
cana-6215	64	4	simplicity	simplicity	NOUN
cana-6215	64	5	and	and	CCONJ
cana-6215	64	6	strong	strong	ADJ
cana-6215	64	7	generalization	generalization	NOUN
cana-6215	64	8	ability	ability	NOUN
cana-6215	64	9	make	make	VERB
cana-6215	64	10	it	it	PRON
cana-6215	64	11	one	one	NUM
cana-6215	64	12	of	of	ADP
cana-6215	64	13	the	the	DET
cana-6215	64	14	most	most	ADV
cana-6215	64	15	reliable	reliable	ADJ
cana-6215	64	16	architectures	architecture	NOUN
cana-6215	64	17	for	for	ADP
cana-6215	64	18	medical	medical	ADJ
cana-6215	64	19	image	image	NOUN
cana-6215	64	20	analysis	analysis	NOUN
cana-6215	64	21	tasks	task	NOUN
cana-6215	64	22	(	(	PUNCT
cana-6215	64	23	figure	figure	NOUN
cana-6215	64	24	2	2	NUM
cana-6215	64	25	)	)	PUNCT
cana-6215	64	26	.	.	PUNCT
cana-6215	65	1	figure	figure	VERB
cana-6215	65	2	2	2	NUM
cana-6215	65	3	:	:	PUNCT
cana-6215	65	4	architecture	architecture	NOUN
cana-6215	65	5	of	of	ADP
cana-6215	65	6	the	the	DET
cana-6215	65	7	vgg16	vgg16	NOUN
cana-6215	65	8	model	model	NOUN
cana-6215	65	9	for	for	ADP
cana-6215	65	10	liver	liver	NOUN
cana-6215	65	11	tumor	tumor	NOUN
cana-6215	65	12	classification	classification	NOUN
cana-6215	65	13	•	•	NOUN
cana-6215	65	14	resnet50	resnet50	NOUN
cana-6215	65	15	resnet50	resnet50	NOUN
cana-6215	65	16	introduces	introduce	VERB
cana-6215	65	17	the	the	DET
cana-6215	65	18	concept	concept	NOUN
cana-6215	65	19	of	of	ADP
cana-6215	65	20	residual	residual	ADJ
cana-6215	65	21	learning	learning	NOUN
cana-6215	65	22	,	,	PUNCT
cana-6215	65	23	which	which	PRON
cana-6215	65	24	helps	help	VERB
cana-6215	65	25	in	in	ADP
cana-6215	65	26	overcoming	overcome	VERB
cana-6215	65	27	the	the	DET
cana-6215	65	28	vanishing	vanish	VERB
cana-6215	65	29	gradient	gradient	ADJ
cana-6215	65	30	problem	problem	NOUN
cana-6215	65	31	commonly	commonly	ADV
cana-6215	65	32	faced	face	VERB
cana-6215	65	33	in	in	ADP
cana-6215	65	34	deep	deep	ADJ
cana-6215	65	35	networks	network	NOUN
cana-6215	65	36	.	.	PUNCT
cana-6215	66	1	by	by	ADP
cana-6215	66	2	incorporating	incorporate	VERB
cana-6215	66	3	skip	skip	ADJ
cana-6215	66	4	connections	connection	NOUN
cana-6215	66	5	,	,	PUNCT
cana-6215	66	6	it	it	PRON
cana-6215	66	7	allows	allow	VERB
cana-6215	66	8	gradients	gradient	NOUN
cana-6215	66	9	to	to	PART
cana-6215	66	10	flow	flow	VERB
cana-6215	66	11	directly	directly	ADV
cana-6215	66	12	through	through	ADP
cana-6215	66	13	layers	layer	NOUN
cana-6215	66	14	,	,	PUNCT
cana-6215	66	15	enabling	enable	VERB
cana-6215	66	16	the	the	DET
cana-6215	66	17	training	training	NOUN
cana-6215	66	18	of	of	ADP
cana-6215	66	19	deeper	deep	ADJ
cana-6215	66	20	architectures	architecture	NOUN
cana-6215	66	21	without	without	ADP
cana-6215	66	22	performance	performance	NOUN
cana-6215	66	23	degradation	degradation	NOUN
cana-6215	66	24	.	.	PUNCT
cana-6215	67	1	in	in	ADP
cana-6215	67	2	this	this	DET
cana-6215	67	3	study	study	NOUN
cana-6215	67	4	,	,	PUNCT
cana-6215	67	5	resnet50	resnet50	NOUN
cana-6215	67	6	was	be	AUX
cana-6215	67	7	utilized	utilize	VERB
cana-6215	67	8	to	to	PART
cana-6215	67	9	extract	extract	VERB
cana-6215	67	10	deep	deep	ADJ
cana-6215	67	11	hierarchical	hierarchical	ADJ
cana-6215	67	12	features	feature	NOUN
cana-6215	67	13	from	from	ADP
cana-6215	67	14	ct	ct	NUM
cana-6215	67	15	scan	scan	ADJ
cana-6215	67	16	images	image	NOUN
cana-6215	67	17	.	.	PUNCT
cana-6215	68	1	the	the	DET
cana-6215	68	2	model	model	NOUN
cana-6215	68	3	demonstrated	demonstrate	VERB
cana-6215	68	4	remarkable	remarkable	ADJ
cana-6215	68	5	accuracy	accuracy	NOUN
cana-6215	68	6	in	in	ADP
cana-6215	68	7	liver	liver	NOUN
cana-6215	68	8	cancer	cancer	NOUN
cana-6215	68	9	classification	classification	NOUN
cana-6215	68	10	,	,	PUNCT
cana-6215	68	11	as	as	SCONJ
cana-6215	68	12	its	its	PRON
cana-6215	68	13	residual	residual	ADJ
cana-6215	68	14	blocks	block	NOUN
cana-6215	68	15	efficiently	efficiently	ADV
cana-6215	68	16	learned	learn	VERB
cana-6215	68	17	complex	complex	ADJ
cana-6215	68	18	feature	feature	NOUN
cana-6215	68	19	representations	representation	NOUN
cana-6215	68	20	,	,	PUNCT
cana-6215	68	21	improving	improve	VERB
cana-6215	68	22	both	both	DET
cana-6215	68	23	convergence	convergence	NOUN
cana-6215	68	24	speed	speed	NOUN
cana-6215	68	25	and	and	CCONJ
cana-6215	68	26	model	model	NOUN
cana-6215	68	27	robustness	robustness	NOUN
cana-6215	68	28	(	(	PUNCT
cana-6215	68	29	figure	figure	NOUN
cana-6215	68	30	3	3	NUM
cana-6215	68	31	)	)	PUNCT
cana-6215	68	32	.	.	PUNCT
cana-6215	69	1	figure	figure	VERB
cana-6215	69	2	3	3	NUM
cana-6215	69	3	:	:	PUNCT
cana-6215	69	4	architecture	architecture	NOUN
cana-6215	69	5	of	of	ADP
cana-6215	69	6	the	the	DET
cana-6215	69	7	resnet50	resnet50	NOUN
cana-6215	69	8	model	model	NOUN
cana-6215	69	9	incorporating	incorporate	VERB
cana-6215	69	10	residual	residual	ADJ
cana-6215	69	11	learning	learning	NOUN
cana-6215	69	12	for	for	ADP
cana-6215	69	13	enhanced	enhanced	ADJ
cana-6215	69	14	feature	feature	NOUN
cana-6215	69	15	extraction	extraction	NOUN
cana-6215	69	16	communications	communication	NOUN
cana-6215	69	17	on	on	ADP
cana-6215	69	18	applied	apply	VERB
cana-6215	69	19	nonlinear	nonlinear	ADJ
cana-6215	69	20	analysis	analysis	NOUN
cana-6215	69	21	issn	issn	NOUN
cana-6215	69	22	:	:	PUNCT
cana-6215	69	23	1074	1074	NUM
cana-6215	69	24	-	-	PUNCT
cana-6215	69	25	133x	133x	NUM
cana-6215	69	26	vol	vol	NOUN
cana-6215	69	27	32	32	NUM
cana-6215	69	28	no	no	NOUN
cana-6215	69	29	.	.	PUNCT
cana-6215	70	1	4s	4s	NUM
cana-6215	70	2	(	(	PUNCT
cana-6215	70	3	2025	2025	NUM
cana-6215	70	4	)	)	PUNCT
cana-6215	70	5	777	777	NUM
cana-6215	70	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-6215	70	7	•	•	NOUN
cana-6215	70	8	mobilenetv2	mobilenetv2	PROPN
cana-6215	70	9	mobilenetv2	mobilenetv2	PROPN
cana-6215	70	10	is	be	AUX
cana-6215	70	11	a	a	DET
cana-6215	70	12	lightweight	lightweight	ADJ
cana-6215	70	13	yet	yet	ADV
cana-6215	70	14	powerful	powerful	ADJ
cana-6215	70	15	deep	deep	ADJ
cana-6215	70	16	learning	learning	NOUN
cana-6215	70	17	architecture	architecture	NOUN
cana-6215	70	18	designed	design	VERB
cana-6215	70	19	for	for	ADP
cana-6215	70	20	efficient	efficient	ADJ
cana-6215	70	21	computation	computation	NOUN
cana-6215	70	22	and	and	CCONJ
cana-6215	70	23	deployment	deployment	NOUN
cana-6215	70	24	on	on	ADP
cana-6215	70	25	limited	limited	ADJ
cana-6215	70	26	-	-	PUNCT
cana-6215	70	27	resource	resource	NOUN
cana-6215	70	28	devices	device	NOUN
cana-6215	70	29	.	.	PUNCT
cana-6215	71	1	it	it	PRON
cana-6215	71	2	uses	use	VERB
cana-6215	71	3	depthwise	depthwise	VERB
cana-6215	71	4	separable	separable	ADJ
cana-6215	71	5	convolutions	convolution	NOUN
cana-6215	71	6	and	and	CCONJ
cana-6215	71	7	inverted	inverted	ADJ
cana-6215	71	8	residuals	residual	NOUN
cana-6215	71	9	with	with	ADP
cana-6215	71	10	linear	linear	ADJ
cana-6215	71	11	bottlenecks	bottleneck	NOUN
cana-6215	71	12	,	,	PUNCT
cana-6215	71	13	significantly	significantly	ADV
cana-6215	71	14	reducing	reduce	VERB
cana-6215	71	15	computational	computational	ADJ
cana-6215	71	16	cost	cost	NOUN
cana-6215	71	17	while	while	SCONJ
cana-6215	71	18	maintaining	maintain	VERB
cana-6215	71	19	high	high	ADJ
cana-6215	71	20	accuracy	accuracy	NOUN
cana-6215	71	21	.	.	PUNCT
cana-6215	72	1	in	in	ADP
cana-6215	72	2	this	this	DET
cana-6215	72	3	research	research	NOUN
cana-6215	72	4	,	,	PUNCT
cana-6215	72	5	mobilenetv2	mobilenetv2	PROPN
cana-6215	72	6	was	be	AUX
cana-6215	72	7	applied	apply	VERB
cana-6215	72	8	to	to	PART
cana-6215	72	9	liver	liver	VERB
cana-6215	72	10	ct	ct	NUM
cana-6215	72	11	images	image	NOUN
cana-6215	72	12	for	for	ADP
cana-6215	72	13	classifying	classify	VERB
cana-6215	72	14	liver	liver	NOUN
cana-6215	72	15	tumors	tumor	NOUN
cana-6215	72	16	,	,	PUNCT
cana-6215	72	17	achieving	achieve	VERB
cana-6215	72	18	excellent	excellent	ADJ
cana-6215	72	19	accuracy	accuracy	NOUN
cana-6215	72	20	with	with	ADP
cana-6215	72	21	minimal	minimal	ADJ
cana-6215	72	22	training	training	NOUN
cana-6215	72	23	time	time	NOUN
cana-6215	72	24	.	.	PUNCT
cana-6215	73	1	its	its	PRON
cana-6215	73	2	compact	compact	ADJ
cana-6215	73	3	design	design	NOUN
cana-6215	73	4	made	make	VERB
cana-6215	73	5	it	it	PRON
cana-6215	73	6	ideal	ideal	ADJ
cana-6215	73	7	for	for	ADP
cana-6215	73	8	handling	handle	VERB
cana-6215	73	9	smaller	small	ADJ
cana-6215	73	10	medical	medical	ADJ
cana-6215	73	11	datasets	dataset	NOUN
cana-6215	73	12	,	,	PUNCT
cana-6215	73	13	offering	offer	VERB
cana-6215	73	14	a	a	DET
cana-6215	73	15	good	good	ADJ
cana-6215	73	16	balance	balance	NOUN
cana-6215	73	17	between	between	ADP
cana-6215	73	18	efficiency	efficiency	NOUN
cana-6215	73	19	,	,	PUNCT
cana-6215	73	20	accuracy	accuracy	NOUN
cana-6215	73	21	,	,	PUNCT
cana-6215	73	22	and	and	CCONJ
cana-6215	73	23	resource	resource	NOUN
cana-6215	73	24	utilization	utilization	NOUN
cana-6215	73	25	(	(	PUNCT
cana-6215	73	26	figure	figure	NOUN
cana-6215	73	27	4	4	NUM
cana-6215	73	28	)	)	PUNCT
cana-6215	73	29	.	.	PUNCT
cana-6215	74	1	figure	figure	VERB
cana-6215	74	2	4	4	NUM
cana-6215	74	3	:	:	PUNCT
cana-6215	74	4	architecture	architecture	NOUN
cana-6215	74	5	of	of	ADP
cana-6215	74	6	the	the	DET
cana-6215	74	7	mobilenetv2	mobilenetv2	PROPN
cana-6215	74	8	model	model	NOUN
cana-6215	74	9	employing	employ	VERB
cana-6215	74	10	depthwise	depthwise	NOUN
cana-6215	74	11	separable	separable	ADJ
cana-6215	74	12	convolutions	convolution	NOUN
cana-6215	74	13	for	for	ADP
cana-6215	74	14	efficient	efficient	ADJ
cana-6215	74	15	computation	computation	NOUN
cana-6215	74	16	•	•	NOUN
cana-6215	74	17	inceptionv3	inceptionv3	NOUN
cana-6215	75	1	inceptionv3	inceptionv3	NOUN
cana-6215	75	2	is	be	AUX
cana-6215	75	3	a	a	DET
cana-6215	75	4	deep	deep	ADJ
cana-6215	75	5	convolutional	convolutional	ADJ
cana-6215	75	6	model	model	NOUN
cana-6215	75	7	that	that	PRON
cana-6215	75	8	improves	improve	VERB
cana-6215	75	9	computational	computational	ADJ
cana-6215	75	10	efficiency	efficiency	NOUN
cana-6215	75	11	by	by	ADP
cana-6215	75	12	factorizing	factorize	VERB
cana-6215	75	13	convolutions	convolution	NOUN
cana-6215	75	14	and	and	CCONJ
cana-6215	75	15	incorporating	incorporate	VERB
cana-6215	75	16	multiple	multiple	ADJ
cana-6215	75	17	kernel	kernel	NOUN
cana-6215	75	18	sizes	size	NOUN
cana-6215	75	19	within	within	ADP
cana-6215	75	20	the	the	DET
cana-6215	75	21	same	same	ADJ
cana-6215	75	22	layer	layer	NOUN
cana-6215	75	23	to	to	PART
cana-6215	75	24	capture	capture	VERB
cana-6215	75	25	both	both	DET
cana-6215	75	26	local	local	ADJ
cana-6215	75	27	and	and	CCONJ
cana-6215	75	28	global	global	ADJ
cana-6215	75	29	image	image	NOUN
cana-6215	75	30	features	feature	NOUN
cana-6215	75	31	.	.	PUNCT
cana-6215	76	1	it	it	PRON
cana-6215	76	2	employs	employ	VERB
cana-6215	76	3	techniques	technique	NOUN
cana-6215	76	4	such	such	ADJ
cana-6215	76	5	as	as	ADP
cana-6215	76	6	batch	batch	NOUN
cana-6215	76	7	normalization	normalization	NOUN
cana-6215	76	8	,	,	PUNCT
cana-6215	76	9	label	label	NOUN
cana-6215	76	10	smoothing	smoothing	NOUN
cana-6215	76	11	,	,	PUNCT
cana-6215	76	12	and	and	CCONJ
cana-6215	76	13	auxiliary	auxiliary	ADJ
cana-6215	76	14	classifiers	classifier	NOUN
cana-6215	76	15	to	to	PART
cana-6215	76	16	enhance	enhance	VERB
cana-6215	76	17	performance	performance	NOUN
cana-6215	76	18	and	and	CCONJ
cana-6215	76	19	prevent	prevent	VERB
cana-6215	76	20	overfitting	overfitting	NOUN
cana-6215	76	21	.	.	PUNCT
cana-6215	77	1	in	in	ADP
cana-6215	77	2	the	the	DET
cana-6215	77	3	proposed	propose	VERB
cana-6215	77	4	study	study	NOUN
cana-6215	77	5	,	,	PUNCT
cana-6215	77	6	inceptionv3	inceptionv3	PROPN
cana-6215	77	7	was	be	AUX
cana-6215	77	8	used	use	VERB
cana-6215	77	9	to	to	PART
cana-6215	77	10	extract	extract	VERB
cana-6215	77	11	rich	rich	ADJ
cana-6215	77	12	,	,	PUNCT
cana-6215	77	13	multi	multi	ADJ
cana-6215	77	14	-	-	ADJ
cana-6215	77	15	scale	scale	ADJ
cana-6215	77	16	feature	feature	NOUN
cana-6215	77	17	representations	representation	NOUN
cana-6215	77	18	from	from	ADP
cana-6215	77	19	liver	liver	NOUN
cana-6215	77	20	ct	ct	PROPN
cana-6215	77	21	scans	scan	NOUN
cana-6215	77	22	.	.	PUNCT
cana-6215	78	1	its	its	PRON
cana-6215	78	2	ability	ability	NOUN
cana-6215	78	3	to	to	PART
cana-6215	78	4	integrate	integrate	VERB
cana-6215	78	5	diverse	diverse	ADJ
cana-6215	78	6	feature	feature	NOUN
cana-6215	78	7	maps	map	NOUN
cana-6215	78	8	helped	helped	AUX
cana-6215	78	9	improve	improve	VERB
cana-6215	78	10	classification	classification	NOUN
cana-6215	78	11	precision	precision	NOUN
cana-6215	78	12	,	,	PUNCT
cana-6215	78	13	especially	especially	ADV
cana-6215	78	14	in	in	ADP
cana-6215	78	15	distinguishing	distinguish	VERB
cana-6215	78	16	subtle	subtle	ADJ
cana-6215	78	17	variations	variation	NOUN
cana-6215	78	18	among	among	ADP
cana-6215	78	19	liver	liver	NOUN
cana-6215	78	20	tissue	tissue	NOUN
cana-6215	78	21	classes	class	NOUN
cana-6215	78	22	(	(	PUNCT
cana-6215	78	23	figure	figure	NOUN
cana-6215	78	24	5	5	NUM
cana-6215	78	25	)	)	PUNCT
cana-6215	78	26	.	.	PUNCT
cana-6215	79	1	figure	figure	VERB
cana-6215	79	2	5	5	NUM
cana-6215	79	3	:	:	PUNCT
cana-6215	79	4	architecture	architecture	NOUN
cana-6215	79	5	of	of	ADP
cana-6215	79	6	the	the	DET
cana-6215	79	7	inceptionv3	inceptionv3	NOUN
cana-6215	79	8	model	model	NOUN
cana-6215	79	9	utilizing	utilize	VERB
cana-6215	79	10	multi	multi	ADJ
cana-6215	79	11	-	-	ADJ
cana-6215	79	12	scale	scale	ADJ
cana-6215	79	13	convolutional	convolutional	ADJ
cana-6215	79	14	modules	module	NOUN
cana-6215	79	15	for	for	ADP
cana-6215	79	16	rich	rich	ADJ
cana-6215	79	17	feature	feature	NOUN
cana-6215	79	18	representation	representation	NOUN
cana-6215	79	19	•	•	PROPN
cana-6215	79	20	densenet121	densenet121	PROPN
cana-6215	79	21	densenet121	densenet121	PROPN
cana-6215	79	22	is	be	AUX
cana-6215	79	23	a	a	DET
cana-6215	79	24	densely	densely	ADV
cana-6215	79	25	connected	connect	VERB
cana-6215	79	26	convolutional	convolutional	ADJ
cana-6215	79	27	neural	neural	ADJ
cana-6215	79	28	network	network	NOUN
cana-6215	79	29	where	where	SCONJ
cana-6215	79	30	each	each	DET
cana-6215	79	31	layer	layer	NOUN
cana-6215	79	32	receives	receive	VERB
cana-6215	79	33	inputs	input	NOUN
cana-6215	79	34	from	from	ADP
cana-6215	79	35	all	all	DET
cana-6215	79	36	preceding	precede	VERB
cana-6215	79	37	layers	layer	NOUN
cana-6215	79	38	,	,	PUNCT
cana-6215	79	39	promoting	promote	VERB
cana-6215	79	40	feature	feature	NOUN
cana-6215	79	41	reuse	reuse	NOUN
cana-6215	79	42	and	and	CCONJ
cana-6215	79	43	reducing	reduce	VERB
cana-6215	79	44	the	the	DET
cana-6215	79	45	number	number	NOUN
cana-6215	79	46	of	of	ADP
cana-6215	79	47	parameters	parameter	NOUN
cana-6215	79	48	.	.	PUNCT
cana-6215	80	1	this	this	DET
cana-6215	80	2	architecture	architecture	NOUN
cana-6215	80	3	ensures	ensure	VERB
cana-6215	80	4	efficient	efficient	ADJ
cana-6215	80	5	gradient	gradient	NOUN
cana-6215	80	6	flow	flow	NOUN
cana-6215	80	7	,	,	PUNCT
cana-6215	80	8	improved	improve	VERB
cana-6215	80	9	learning	learning	NOUN
cana-6215	80	10	efficiency	efficiency	NOUN
cana-6215	80	11	,	,	PUNCT
cana-6215	80	12	and	and	CCONJ
cana-6215	80	13	stronger	strong	ADJ
cana-6215	80	14	feature	feature	NOUN
cana-6215	80	15	propagation	propagation	NOUN
cana-6215	80	16	.	.	PUNCT
cana-6215	81	1	in	in	ADP
cana-6215	81	2	this	this	DET
cana-6215	81	3	work	work	NOUN
cana-6215	81	4	,	,	PUNCT
cana-6215	81	5	densenet121	densenet121	PROPN
cana-6215	81	6	was	be	AUX
cana-6215	81	7	implemented	implement	VERB
cana-6215	81	8	to	to	PART
cana-6215	81	9	classify	classify	VERB
cana-6215	81	10	liver	liver	NOUN
cana-6215	81	11	tumors	tumor	NOUN
cana-6215	81	12	by	by	ADP
cana-6215	81	13	leveraging	leverage	VERB
cana-6215	81	14	its	its	PRON
cana-6215	81	15	dense	dense	ADJ
cana-6215	81	16	connectivity	connectivity	NOUN
cana-6215	81	17	to	to	PART
cana-6215	81	18	extract	extract	VERB
cana-6215	81	19	deep	deep	ADJ
cana-6215	81	20	and	and	CCONJ
cana-6215	81	21	discriminative	discriminative	VERB
cana-6215	81	22	image	image	NOUN
cana-6215	81	23	features	feature	NOUN
cana-6215	81	24	.	.	PUNCT
cana-6215	82	1	the	the	DET
cana-6215	82	2	model	model	NOUN
cana-6215	82	3	achieved	achieve	VERB
cana-6215	82	4	stable	stable	ADJ
cana-6215	82	5	and	and	CCONJ
cana-6215	82	6	high	high	ADJ
cana-6215	82	7	-	-	PUNCT
cana-6215	82	8	performance	performance	NOUN
cana-6215	82	9	results	result	NOUN
cana-6215	82	10	,	,	PUNCT
cana-6215	82	11	demonstrating	demonstrate	VERB
cana-6215	82	12	superior	superior	ADJ
cana-6215	82	13	learning	learning	NOUN
cana-6215	82	14	communications	communication	NOUN
cana-6215	82	15	on	on	ADP
cana-6215	82	16	applied	apply	VERB
cana-6215	82	17	nonlinear	nonlinear	ADJ
cana-6215	82	18	analysis	analysis	NOUN
cana-6215	82	19	issn	issn	NOUN
cana-6215	82	20	:	:	PUNCT
cana-6215	82	21	1074	1074	NUM
cana-6215	82	22	-	-	PUNCT
cana-6215	82	23	133x	133x	NUM
cana-6215	82	24	vol	vol	NOUN
cana-6215	82	25	32	32	NUM
cana-6215	82	26	no	no	NOUN
cana-6215	82	27	.	.	PUNCT
cana-6215	83	1	4s	4s	NUM
cana-6215	83	2	(	(	PUNCT
cana-6215	83	3	2025	2025	NUM
cana-6215	83	4	)	)	PUNCT
cana-6215	83	5	778	778	NUM
cana-6215	83	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-6215	83	7	capability	capability	NOUN
cana-6215	83	8	even	even	ADV
cana-6215	83	9	with	with	ADP
cana-6215	83	10	a	a	DET
cana-6215	83	11	relatively	relatively	ADV
cana-6215	83	12	small	small	ADJ
cana-6215	83	13	dataset	dataset	NOUN
cana-6215	83	14	,	,	PUNCT
cana-6215	83	15	making	make	VERB
cana-6215	83	16	it	it	PRON
cana-6215	83	17	suitable	suitable	ADJ
cana-6215	83	18	for	for	ADP
cana-6215	83	19	complex	complex	ADJ
cana-6215	83	20	medical	medical	ADJ
cana-6215	83	21	imaging	imaging	NOUN
cana-6215	83	22	applications	application	NOUN
cana-6215	83	23	(	(	PUNCT
cana-6215	83	24	figure	figure	NOUN
cana-6215	83	25	6	6	NUM
cana-6215	83	26	)	)	PUNCT
cana-6215	83	27	.	.	PUNCT
cana-6215	84	1	figure	figure	VERB
cana-6215	84	2	6	6	NUM
cana-6215	84	3	:	:	PUNCT
cana-6215	84	4	architecture	architecture	NOUN
cana-6215	84	5	of	of	ADP
cana-6215	84	6	the	the	DET
cana-6215	84	7	densenet121	densenet121	PROPN
cana-6215	84	8	model	model	NOUN
cana-6215	84	9	featuring	feature	VERB
cana-6215	84	10	densely	densely	ADV
cana-6215	84	11	connected	connect	VERB
cana-6215	84	12	layers	layer	NOUN
cana-6215	84	13	for	for	ADP
cana-6215	84	14	improved	improved	ADJ
cana-6215	84	15	gradient	gradient	NOUN
cana-6215	84	16	flow	flow	NOUN
cana-6215	84	17	and	and	CCONJ
cana-6215	84	18	feature	feature	NOUN
cana-6215	84	19	reuse	reuse	NOUN
cana-6215	84	20	4	4	NUM
cana-6215	84	21	.	.	PUNCT
cana-6215	84	22	proposed	propose	VERB
cana-6215	84	23	system	system	NOUN
cana-6215	84	24	framework	framework	NOUN
cana-6215	84	25	the	the	DET
cana-6215	84	26	proposed	propose	VERB
cana-6215	84	27	study	study	NOUN
cana-6215	84	28	aims	aim	VERB
cana-6215	84	29	to	to	PART
cana-6215	84	30	develop	develop	VERB
cana-6215	84	31	an	an	DET
cana-6215	84	32	efficient	efficient	ADJ
cana-6215	84	33	deep	deep	ADJ
cana-6215	84	34	transfer	transfer	NOUN
cana-6215	84	35	learning	learn	VERB
cana-6215	84	36	framework	framework	NOUN
cana-6215	84	37	for	for	ADP
cana-6215	84	38	accurate	accurate	ADJ
cana-6215	84	39	classification	classification	NOUN
cana-6215	84	40	of	of	ADP
cana-6215	84	41	liver	liver	NOUN
cana-6215	84	42	tumors	tumor	NOUN
cana-6215	84	43	using	use	VERB
cana-6215	84	44	computed	computed	ADJ
cana-6215	84	45	tomography	tomography	NOUN
cana-6215	84	46	(	(	PUNCT
cana-6215	84	47	ct	ct	NOUN
cana-6215	84	48	)	)	PUNCT
cana-6215	84	49	images	image	NOUN
cana-6215	84	50	(	(	PUNCT
cana-6215	84	51	figure	figure	NOUN
cana-6215	84	52	7	7	NUM
cana-6215	84	53	)	)	PUNCT
cana-6215	84	54	.	.	PUNCT
cana-6215	85	1	the	the	DET
cana-6215	85	2	methodology	methodology	NOUN
cana-6215	85	3	is	be	AUX
cana-6215	85	4	designed	design	VERB
cana-6215	85	5	to	to	PART
cana-6215	85	6	ensure	ensure	VERB
cana-6215	85	7	precise	precise	ADJ
cana-6215	85	8	tumor	tumor	NOUN
cana-6215	85	9	detection	detection	NOUN
cana-6215	85	10	and	and	CCONJ
cana-6215	85	11	differentiation	differentiation	NOUN
cana-6215	85	12	between	between	ADP
cana-6215	85	13	malignant	malignant	ADJ
cana-6215	85	14	,	,	PUNCT
cana-6215	85	15	benign	benign	ADJ
cana-6215	85	16	,	,	PUNCT
cana-6215	85	17	and	and	CCONJ
cana-6215	85	18	normal	normal	ADJ
cana-6215	85	19	liver	liver	NOUN
cana-6215	85	20	tissues	tissue	NOUN
cana-6215	85	21	.	.	PUNCT
cana-6215	86	1	the	the	DET
cana-6215	86	2	following	follow	VERB
cana-6215	86	3	subsections	subsection	NOUN
cana-6215	86	4	describe	describe	VERB
cana-6215	86	5	the	the	DET
cana-6215	86	6	materials	material	NOUN
cana-6215	86	7	,	,	PUNCT
cana-6215	86	8	dataset	dataset	ADJ
cana-6215	86	9	preparation	preparation	NOUN
cana-6215	86	10	,	,	PUNCT
cana-6215	86	11	preprocessing	preprocesse	VERB
cana-6215	86	12	techniques	technique	NOUN
cana-6215	86	13	,	,	PUNCT
cana-6215	86	14	model	model	NOUN
cana-6215	86	15	architectures	architecture	NOUN
cana-6215	86	16	,	,	PUNCT
cana-6215	86	17	training	training	NOUN
cana-6215	86	18	parameters	parameter	NOUN
cana-6215	86	19	,	,	PUNCT
cana-6215	86	20	and	and	CCONJ
cana-6215	86	21	evaluation	evaluation	NOUN
cana-6215	86	22	metrics	metric	NOUN
cana-6215	86	23	in	in	ADP
cana-6215	86	24	detail	detail	NOUN
cana-6215	86	25	(	(	PUNCT
cana-6215	86	26	table	table	NOUN
cana-6215	86	27	2	2	NUM
cana-6215	86	28	)	)	PUNCT
cana-6215	86	29	.	.	PUNCT
cana-6215	87	1	table	table	NOUN
cana-6215	87	2	2	2	NUM
cana-6215	87	3	:	:	PUNCT
cana-6215	87	4	summary	summary	NOUN
cana-6215	87	5	of	of	ADP
cana-6215	87	6	materials	material	NOUN
cana-6215	87	7	and	and	CCONJ
cana-6215	87	8	methods	method	NOUN
cana-6215	87	9	used	use	VERB
cana-6215	87	10	in	in	ADP
cana-6215	87	11	the	the	DET
cana-6215	87	12	proposed	propose	VERB
cana-6215	87	13	study	study	NOUN
cana-6215	87	14	category	category	NOUN
cana-6215	87	15	description	description	NOUN
cana-6215	87	16	dataset	dataset	NOUN
cana-6215	87	17	source	source	NOUN
cana-6215	87	18	computed	compute	VERB
cana-6215	87	19	tomography	tomography	NOUN
cana-6215	87	20	(	(	PUNCT
cana-6215	87	21	ct	ct	PROPN
cana-6215	87	22	)	)	PUNCT
cana-6215	87	23	scan	scan	ADJ
cana-6215	87	24	images	image	NOUN
cana-6215	87	25	collected	collect	VERB
cana-6215	87	26	from	from	ADP
cana-6215	87	27	the	the	DET
cana-6215	87	28	radiology	radiology	NOUN
cana-6215	87	29	institute	institute	NOUN
cana-6215	87	30	,	,	PUNCT
cana-6215	87	31	baghdad	baghdad	PROPN
cana-6215	87	32	medical	medical	ADJ
cana-6215	87	33	city	city	PROPN
cana-6215	87	34	,	,	PUNCT
cana-6215	87	35	iraq	iraq	PROPN
cana-6215	87	36	.	.	PUNCT
cana-6215	88	1	dataset	dataset	NOUN
cana-6215	88	2	composition	composition	NOUN
cana-6215	88	3	images	image	NOUN
cana-6215	88	4	categorized	categorize	VERB
cana-6215	88	5	into	into	ADP
cana-6215	88	6	three	three	NUM
cana-6215	88	7	classes	class	NOUN
cana-6215	88	8	:	:	PUNCT
cana-6215	88	9	malignant	malignant	ADJ
cana-6215	88	10	tumor	tumor	NOUN
cana-6215	88	11	,	,	PUNCT
cana-6215	88	12	benign	benign	ADJ
cana-6215	88	13	tumor	tumor	NOUN
cana-6215	88	14	,	,	PUNCT
cana-6215	88	15	and	and	CCONJ
cana-6215	88	16	normal	normal	ADJ
cana-6215	88	17	liver	liver	NOUN
cana-6215	88	18	.	.	PUNCT
cana-6215	89	1	data	datum	NOUN
cana-6215	89	2	preprocessing	preprocesse	VERB
cana-6215	89	3	image	image	NOUN
cana-6215	89	4	resizing	resize	VERB
cana-6215	89	5	to	to	ADP
cana-6215	89	6	a	a	DET
cana-6215	89	7	uniform	uniform	ADJ
cana-6215	89	8	dimension	dimension	NOUN
cana-6215	89	9	,	,	PUNCT
cana-6215	89	10	noise	noise	NOUN
cana-6215	89	11	removal	removal	NOUN
cana-6215	89	12	using	use	VERB
cana-6215	89	13	gaussian	gaussian	ADJ
cana-6215	89	14	filtering	filtering	NOUN
cana-6215	89	15	,	,	PUNCT
cana-6215	89	16	intensity	intensity	NOUN
cana-6215	89	17	normalization	normalization	NOUN
cana-6215	89	18	,	,	PUNCT
cana-6215	89	19	and	and	CCONJ
cana-6215	89	20	data	datum	NOUN
cana-6215	89	21	augmentation	augmentation	NOUN
cana-6215	89	22	(	(	PUNCT
cana-6215	89	23	rotation	rotation	NOUN
cana-6215	89	24	,	,	PUNCT
cana-6215	89	25	flipping	flipping	NOUN
cana-6215	89	26	,	,	PUNCT
cana-6215	89	27	zooming	zooming	NOUN
cana-6215	89	28	)	)	PUNCT
cana-6215	89	29	to	to	PART
cana-6215	89	30	enhance	enhance	VERB
cana-6215	89	31	dataset	dataset	ADJ
cana-6215	89	32	diversity	diversity	NOUN
cana-6215	89	33	.	.	PUNCT
cana-6215	90	1	feature	feature	NOUN
cana-6215	90	2	extraction	extraction	NOUN
cana-6215	90	3	high	high	ADJ
cana-6215	90	4	-	-	PUNCT
cana-6215	90	5	level	level	NOUN
cana-6215	90	6	features	feature	NOUN
cana-6215	90	7	extracted	extract	VERB
cana-6215	90	8	using	use	VERB
cana-6215	90	9	pre	pre	ADJ
cana-6215	90	10	-	-	ADJ
cana-6215	90	11	trained	train	VERB
cana-6215	90	12	deep	deep	ADJ
cana-6215	90	13	convolutional	convolutional	ADJ
cana-6215	90	14	neural	neural	ADJ
cana-6215	90	15	networks	network	NOUN
cana-6215	90	16	(	(	PUNCT
cana-6215	90	17	cnns	cnns	PROPN
cana-6215	90	18	)	)	PUNCT
cana-6215	90	19	such	such	ADJ
cana-6215	90	20	as	as	ADP
cana-6215	90	21	vgg-16	vgg-16	NOUN
cana-6215	90	22	,	,	PUNCT
cana-6215	90	23	resnet-50	resnet-50	PROPN
cana-6215	90	24	,	,	PUNCT
cana-6215	90	25	and	and	CCONJ
cana-6215	90	26	mobilenetv2	mobilenetv2	NOUN
cana-6215	90	27	through	through	ADP
cana-6215	90	28	transfer	transfer	NOUN
cana-6215	90	29	learning	learning	NOUN
cana-6215	90	30	.	.	PUNCT
cana-6215	91	1	transfer	transfer	NOUN
cana-6215	91	2	learning	learn	VERB
cana-6215	91	3	approach	approach	NOUN
cana-6215	91	4	utilized	utilize	VERB
cana-6215	91	5	pre	pre	ADJ
cana-6215	91	6	-	-	ADJ
cana-6215	91	7	trained	train	VERB
cana-6215	91	8	models	model	NOUN
cana-6215	91	9	fine	fine	ADV
cana-6215	91	10	-	-	PUNCT
cana-6215	91	11	tuned	tune	VERB
cana-6215	91	12	on	on	ADP
cana-6215	91	13	the	the	DET
cana-6215	91	14	liver	liver	NOUN
cana-6215	91	15	ct	ct	PROPN
cana-6215	91	16	dataset	dataset	VERB
cana-6215	91	17	to	to	ADP
cana-6215	91	18	leverage	leverage	NOUN
cana-6215	91	19	learned	learn	VERB
cana-6215	91	20	weights	weight	NOUN
cana-6215	91	21	from	from	ADP
cana-6215	91	22	imagenet	imagenet	NOUN
cana-6215	91	23	for	for	ADP
cana-6215	91	24	improved	improved	ADJ
cana-6215	91	25	accuracy	accuracy	NOUN
cana-6215	91	26	with	with	ADP
cana-6215	91	27	limited	limited	ADJ
cana-6215	91	28	medical	medical	ADJ
cana-6215	91	29	data	datum	NOUN
cana-6215	91	30	.	.	PUNCT
cana-6215	92	1	training	training	NOUN
cana-6215	92	2	and	and	CCONJ
cana-6215	92	3	validation	validation	NOUN
cana-6215	92	4	split	split	NOUN
cana-6215	92	5	dataset	dataset	NOUN
cana-6215	92	6	divided	divide	VERB
cana-6215	92	7	into	into	ADP
cana-6215	92	8	80	80	NUM
cana-6215	92	9	%	%	NOUN
cana-6215	92	10	training	training	NOUN
cana-6215	92	11	and	and	CCONJ
cana-6215	92	12	20	20	NUM
cana-6215	92	13	%	%	NOUN
cana-6215	92	14	testing	testing	NOUN
cana-6215	92	15	sets	set	NOUN
cana-6215	92	16	;	;	PUNCT
cana-6215	92	17	model	model	NOUN
cana-6215	92	18	performance	performance	NOUN
cana-6215	92	19	validated	validate	VERB
cana-6215	92	20	using	use	VERB
cana-6215	92	21	cross	cross	ADJ
cana-6215	92	22	-	-	ADJ
cana-6215	92	23	validation	validation	ADJ
cana-6215	92	24	techniques	technique	NOUN
cana-6215	92	25	.	.	PUNCT
cana-6215	93	1	optimization	optimization	NOUN
cana-6215	93	2	technique	technique	NOUN
cana-6215	93	3	adam	adam	PROPN
cana-6215	93	4	optimizer	optimizer	NOUN
cana-6215	93	5	used	use	VERB
cana-6215	93	6	with	with	ADP
cana-6215	93	7	a	a	DET
cana-6215	93	8	learning	learn	VERB
cana-6215	93	9	rate	rate	NOUN
cana-6215	93	10	of	of	ADP
cana-6215	93	11	0.0001	0.0001	NUM
cana-6215	93	12	;	;	PUNCT
cana-6215	93	13	categorical	categorical	ADJ
cana-6215	93	14	cross	cross	NOUN
cana-6215	93	15	-	-	ADJ
cana-6215	93	16	entropy	entropy	NOUN
cana-6215	93	17	employed	employ	VERB
cana-6215	93	18	as	as	ADP
cana-6215	93	19	the	the	DET
cana-6215	93	20	loss	loss	NOUN
cana-6215	93	21	function	function	NOUN
cana-6215	93	22	.	.	PUNCT
cana-6215	94	1	performance	performance	NOUN
cana-6215	94	2	metrics	metric	NOUN
cana-6215	94	3	model	model	NOUN
cana-6215	94	4	performance	performance	NOUN
cana-6215	94	5	evaluated	evaluate	VERB
cana-6215	94	6	using	use	VERB
cana-6215	94	7	accuracy	accuracy	NOUN
cana-6215	94	8	,	,	PUNCT
cana-6215	94	9	sensitivity	sensitivity	NOUN
cana-6215	94	10	,	,	PUNCT
cana-6215	94	11	specificity	specificity	NOUN
cana-6215	94	12	,	,	PUNCT
cana-6215	94	13	precision	precision	NOUN
cana-6215	94	14	,	,	PUNCT
cana-6215	94	15	and	and	CCONJ
cana-6215	94	16	f1	f1	NOUN
cana-6215	94	17	-	-	PUNCT
cana-6215	94	18	score	score	NOUN
cana-6215	94	19	.	.	PUNCT
cana-6215	95	1	communications	communication	NOUN
cana-6215	95	2	on	on	ADP
cana-6215	95	3	applied	apply	VERB
cana-6215	95	4	nonlinear	nonlinear	ADJ
cana-6215	95	5	analysis	analysis	NOUN
cana-6215	95	6	issn	issn	NOUN
cana-6215	95	7	:	:	PUNCT
cana-6215	95	8	1074	1074	NUM
cana-6215	95	9	-	-	PUNCT
cana-6215	95	10	133x	133x	NUM
cana-6215	95	11	vol	vol	NOUN
cana-6215	95	12	32	32	NUM
cana-6215	95	13	no	no	NOUN
cana-6215	95	14	.	.	PUNCT
cana-6215	96	1	4s	4s	NUM
cana-6215	96	2	(	(	PUNCT
cana-6215	96	3	2025	2025	NUM
cana-6215	96	4	)	)	PUNCT
cana-6215	96	5	779	779	NUM
cana-6215	96	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-6215	96	7	•	•	NUM
cana-6215	96	8	dataset	dataset	NOUN
cana-6215	96	9	description	description	NOUN
cana-6215	96	10	the	the	DET
cana-6215	96	11	dataset	dataset	NOUN
cana-6215	96	12	used	use	VERB
cana-6215	96	13	in	in	ADP
cana-6215	96	14	this	this	DET
cana-6215	96	15	study	study	NOUN
cana-6215	96	16	was	be	AUX
cana-6215	96	17	obtained	obtain	VERB
cana-6215	96	18	from	from	ADP
cana-6215	96	19	the	the	DET
cana-6215	96	20	radiology	radiology	NOUN
cana-6215	96	21	institute	institute	NOUN
cana-6215	96	22	,	,	PUNCT
cana-6215	96	23	baghdad	baghdad	PROPN
cana-6215	96	24	medical	medical	ADJ
cana-6215	96	25	city	city	PROPN
cana-6215	96	26	,	,	PUNCT
cana-6215	96	27	iraq	iraq	PROPN
cana-6215	96	28	.	.	PUNCT
cana-6215	97	1	it	it	PRON
cana-6215	97	2	comprises	comprise	VERB
cana-6215	97	3	ct	ct	NOUN
cana-6215	97	4	scan	scan	ADJ
cana-6215	97	5	images	image	NOUN
cana-6215	97	6	of	of	ADP
cana-6215	97	7	patients	patient	NOUN
cana-6215	97	8	with	with	ADP
cana-6215	97	9	confirmed	confirm	VERB
cana-6215	97	10	liver	liver	NOUN
cana-6215	97	11	conditions	condition	NOUN
cana-6215	97	12	,	,	PUNCT
cana-6215	97	13	including	include	VERB
cana-6215	97	14	malignant	malignant	ADJ
cana-6215	97	15	tumors	tumor	NOUN
cana-6215	97	16	,	,	PUNCT
cana-6215	97	17	benign	benign	ADJ
cana-6215	97	18	tumors	tumor	NOUN
cana-6215	97	19	,	,	PUNCT
cana-6215	97	20	and	and	CCONJ
cana-6215	97	21	normal	normal	ADJ
cana-6215	97	22	livers	liver	NOUN
cana-6215	97	23	.	.	PUNCT
cana-6215	98	1	the	the	DET
cana-6215	98	2	images	image	NOUN
cana-6215	98	3	were	be	AUX
cana-6215	98	4	collected	collect	VERB
cana-6215	98	5	under	under	ADP
cana-6215	98	6	standardized	standardized	ADJ
cana-6215	98	7	imaging	imaging	NOUN
cana-6215	98	8	conditions	condition	NOUN
cana-6215	98	9	and	and	CCONJ
cana-6215	98	10	were	be	AUX
cana-6215	98	11	pre	pre	VERB
cana-6215	98	12	-	-	VERB
cana-6215	98	13	labeled	label	VERB
cana-6215	98	14	by	by	ADP
cana-6215	98	15	medical	medical	ADJ
cana-6215	98	16	experts	expert	NOUN
cana-6215	98	17	.	.	PUNCT
cana-6215	99	1	the	the	DET
cana-6215	99	2	dataset	dataset	NOUN
cana-6215	99	3	was	be	AUX
cana-6215	99	4	divided	divide	VERB
cana-6215	99	5	into	into	ADP
cana-6215	99	6	three	three	NUM
cana-6215	99	7	classes	class	NOUN
cana-6215	99	8	for	for	ADP
cana-6215	99	9	classification	classification	NOUN
cana-6215	99	10	purposes	purpose	NOUN
cana-6215	99	11	—	—	PUNCT
cana-6215	99	12	malignant	malignant	ADJ
cana-6215	99	13	,	,	PUNCT
cana-6215	99	14	benign	benign	ADJ
cana-6215	99	15	,	,	PUNCT
cana-6215	99	16	and	and	CCONJ
cana-6215	99	17	normal	normal	ADJ
cana-6215	99	18	—	—	PUNCT
cana-6215	99	19	to	to	PART
cana-6215	99	20	facilitate	facilitate	VERB
cana-6215	99	21	accurate	accurate	ADJ
cana-6215	99	22	categorization	categorization	NOUN
cana-6215	99	23	of	of	ADP
cana-6215	99	24	liver	liver	NOUN
cana-6215	99	25	tissue	tissue	NOUN
cana-6215	99	26	abnormalities	abnormality	NOUN
cana-6215	99	27	.	.	PUNCT
cana-6215	100	1	•	•	NUM
cana-6215	100	2	data	datum	NOUN
cana-6215	100	3	preprocessing	preprocessing	NOUN
cana-6215	100	4	before	before	ADP
cana-6215	100	5	model	model	NOUN
cana-6215	100	6	training	training	NOUN
cana-6215	100	7	,	,	PUNCT
cana-6215	100	8	the	the	DET
cana-6215	100	9	dataset	dataset	NOUN
cana-6215	100	10	underwent	undergo	VERB
cana-6215	100	11	several	several	ADJ
cana-6215	100	12	preprocessing	preprocessing	NOUN
cana-6215	100	13	steps	step	NOUN
cana-6215	100	14	to	to	PART
cana-6215	100	15	ensure	ensure	VERB
cana-6215	100	16	data	datum	NOUN
cana-6215	100	17	quality	quality	NOUN
cana-6215	100	18	and	and	CCONJ
cana-6215	100	19	consistency	consistency	NOUN
cana-6215	100	20	.	.	PUNCT
cana-6215	101	1	all	all	DET
cana-6215	101	2	images	image	NOUN
cana-6215	101	3	were	be	AUX
cana-6215	101	4	resized	resize	VERB
cana-6215	101	5	to	to	ADP
cana-6215	101	6	a	a	DET
cana-6215	101	7	fixed	fix	VERB
cana-6215	101	8	dimension	dimension	NOUN
cana-6215	101	9	(	(	PUNCT
cana-6215	101	10	224	224	NUM
cana-6215	101	11	×	×	NOUN
cana-6215	101	12	224	224	NUM
cana-6215	101	13	pixels	pixel	NOUN
cana-6215	101	14	)	)	PUNCT
cana-6215	101	15	to	to	PART
cana-6215	101	16	match	match	VERB
cana-6215	101	17	the	the	DET
cana-6215	101	18	input	input	NOUN
cana-6215	101	19	requirements	requirement	NOUN
cana-6215	101	20	of	of	ADP
cana-6215	101	21	deep	deep	ADJ
cana-6215	101	22	learning	learning	NOUN
cana-6215	101	23	architectures	architecture	NOUN
cana-6215	101	24	such	such	ADJ
cana-6215	101	25	as	as	ADP
cana-6215	101	26	vgg-16	vgg-16	NOUN
cana-6215	101	27	,	,	PUNCT
cana-6215	101	28	resnet-50	resnet-50	PROPN
cana-6215	101	29	,	,	PUNCT
cana-6215	101	30	and	and	CCONJ
cana-6215	101	31	mobilenetv2	mobilenetv2	PROPN
cana-6215	101	32	.	.	PUNCT
cana-6215	102	1	image	image	NOUN
cana-6215	102	2	enhancement	enhancement	NOUN
cana-6215	102	3	techniques	technique	NOUN
cana-6215	102	4	were	be	AUX
cana-6215	102	5	applied	apply	VERB
cana-6215	102	6	to	to	PART
cana-6215	102	7	reduce	reduce	VERB
cana-6215	102	8	noise	noise	NOUN
cana-6215	102	9	and	and	CCONJ
cana-6215	102	10	improve	improve	VERB
cana-6215	102	11	contrast	contrast	NOUN
cana-6215	102	12	,	,	PUNCT
cana-6215	102	13	thereby	thereby	ADV
cana-6215	102	14	highlighting	highlight	VERB
cana-6215	102	15	important	important	ADJ
cana-6215	102	16	liver	liver	NOUN
cana-6215	102	17	features	feature	NOUN
cana-6215	102	18	.	.	PUNCT
cana-6215	103	1	normalization	normalization	NOUN
cana-6215	103	2	was	be	AUX
cana-6215	103	3	performed	perform	VERB
cana-6215	103	4	to	to	PART
cana-6215	103	5	scale	scale	VERB
cana-6215	103	6	pixel	pixel	NOUN
cana-6215	103	7	intensity	intensity	NOUN
cana-6215	103	8	values	value	NOUN
cana-6215	103	9	between	between	ADP
cana-6215	103	10	0	0	NUM
cana-6215	103	11	and	and	CCONJ
cana-6215	103	12	1	1	NUM
cana-6215	103	13	.	.	PUNCT
cana-6215	104	1	additionally	additionally	ADV
cana-6215	104	2	,	,	PUNCT
cana-6215	104	3	data	datum	NOUN
cana-6215	104	4	augmentation	augmentation	NOUN
cana-6215	104	5	methods	method	NOUN
cana-6215	104	6	such	such	ADJ
cana-6215	104	7	as	as	ADP
cana-6215	104	8	rotation	rotation	NOUN
cana-6215	104	9	,	,	PUNCT
cana-6215	104	10	flipping	flipping	NOUN
cana-6215	104	11	,	,	PUNCT
cana-6215	104	12	zooming	zooming	NOUN
cana-6215	104	13	,	,	PUNCT
cana-6215	104	14	and	and	CCONJ
cana-6215	104	15	translation	translation	NOUN
cana-6215	104	16	were	be	AUX
cana-6215	104	17	implemented	implement	VERB
cana-6215	104	18	to	to	PART
cana-6215	104	19	artificially	artificially	ADV
cana-6215	104	20	increase	increase	VERB
cana-6215	104	21	the	the	DET
cana-6215	104	22	dataset	dataset	NOUN
cana-6215	104	23	size	size	NOUN
cana-6215	104	24	and	and	CCONJ
cana-6215	104	25	reduce	reduce	VERB
cana-6215	104	26	overfitting	overfitting	NOUN
cana-6215	104	27	during	during	ADP
cana-6215	104	28	model	model	NOUN
cana-6215	104	29	training	training	NOUN
cana-6215	104	30	.	.	PUNCT
cana-6215	105	1	•	•	NUM
cana-6215	105	2	feature	feature	NOUN
cana-6215	105	3	extraction	extraction	NOUN
cana-6215	105	4	using	use	VERB
cana-6215	105	5	transfer	transfer	NOUN
cana-6215	105	6	learning	learn	VERB
cana-6215	105	7	to	to	PART
cana-6215	105	8	overcome	overcome	VERB
cana-6215	105	9	the	the	DET
cana-6215	105	10	challenge	challenge	NOUN
cana-6215	105	11	of	of	ADP
cana-6215	105	12	limited	limited	ADJ
cana-6215	105	13	annotated	annotate	VERB
cana-6215	105	14	medical	medical	ADJ
cana-6215	105	15	data	datum	NOUN
cana-6215	105	16	,	,	PUNCT
cana-6215	105	17	transfer	transfer	NOUN
cana-6215	105	18	learning	learning	NOUN
cana-6215	105	19	(	(	PUNCT
cana-6215	105	20	tl	tl	PROPN
cana-6215	105	21	)	)	PUNCT
cana-6215	105	22	was	be	AUX
cana-6215	105	23	adopted	adopt	VERB
cana-6215	105	24	.	.	PUNCT
cana-6215	106	1	tl	tl	PROPN
cana-6215	106	2	utilizes	utilize	VERB
cana-6215	106	3	pre	pre	ADJ
cana-6215	106	4	-	-	ADJ
cana-6215	106	5	trained	train	VERB
cana-6215	106	6	deep	deep	ADJ
cana-6215	106	7	convolutional	convolutional	ADJ
cana-6215	106	8	neural	neural	ADJ
cana-6215	106	9	networks	network	NOUN
cana-6215	106	10	(	(	PUNCT
cana-6215	106	11	cnns	cnns	PROPN
cana-6215	106	12	)	)	PUNCT
cana-6215	106	13	that	that	PRON
cana-6215	106	14	have	have	AUX
cana-6215	106	15	already	already	ADV
cana-6215	106	16	learned	learn	VERB
cana-6215	106	17	robust	robust	ADJ
cana-6215	106	18	feature	feature	NOUN
cana-6215	106	19	representations	representation	NOUN
cana-6215	106	20	from	from	ADP
cana-6215	106	21	large	large	ADJ
cana-6215	106	22	datasets	dataset	NOUN
cana-6215	106	23	such	such	ADJ
cana-6215	106	24	as	as	ADP
cana-6215	106	25	imagenet	imagenet	NOUN
cana-6215	106	26	.	.	PUNCT
cana-6215	107	1	in	in	ADP
cana-6215	107	2	this	this	DET
cana-6215	107	3	study	study	NOUN
cana-6215	107	4	,	,	PUNCT
cana-6215	107	5	three	three	NUM
cana-6215	107	6	pre	pre	ADJ
cana-6215	107	7	-	-	ADJ
cana-6215	107	8	trained	train	VERB
cana-6215	107	9	models	model	NOUN
cana-6215	107	10	—	—	PUNCT
cana-6215	107	11	vgg-16	vgg-16	NOUN
cana-6215	107	12	,	,	PUNCT
cana-6215	107	13	resnet-50	resnet-50	PROPN
cana-6215	107	14	,	,	PUNCT
cana-6215	107	15	and	and	CCONJ
cana-6215	107	16	mobilenetv2	mobilenetv2	PROPN
cana-6215	107	17	—	—	PUNCT
cana-6215	107	18	were	be	AUX
cana-6215	107	19	fine	fine	ADV
cana-6215	107	20	-	-	PUNCT
cana-6215	107	21	tuned	tune	VERB
cana-6215	107	22	on	on	ADP
cana-6215	107	23	the	the	DET
cana-6215	107	24	liver	liver	NOUN
cana-6215	107	25	ct	ct	NUM
cana-6215	107	26	image	image	NOUN
cana-6215	107	27	dataset	dataset	NOUN
cana-6215	107	28	.	.	PUNCT
cana-6215	108	1	the	the	DET
cana-6215	108	2	convolutional	convolutional	ADJ
cana-6215	108	3	layers	layer	NOUN
cana-6215	108	4	of	of	ADP
cana-6215	108	5	these	these	DET
cana-6215	108	6	models	model	NOUN
cana-6215	108	7	were	be	AUX
cana-6215	108	8	used	use	VERB
cana-6215	108	9	for	for	ADP
cana-6215	108	10	automatic	automatic	ADJ
cana-6215	108	11	feature	feature	NOUN
cana-6215	108	12	extraction	extraction	NOUN
cana-6215	108	13	,	,	PUNCT
cana-6215	108	14	while	while	SCONJ
cana-6215	108	15	the	the	DET
cana-6215	108	16	fully	fully	ADV
cana-6215	108	17	connected	connected	ADJ
cana-6215	108	18	layers	layer	NOUN
cana-6215	108	19	were	be	AUX
cana-6215	108	20	modified	modify	VERB
cana-6215	108	21	to	to	PART
cana-6215	108	22	match	match	VERB
cana-6215	108	23	the	the	DET
cana-6215	108	24	number	number	NOUN
cana-6215	108	25	of	of	ADP
cana-6215	108	26	target	target	NOUN
cana-6215	108	27	classes	class	NOUN
cana-6215	108	28	(	(	PUNCT
cana-6215	108	29	three	three	NUM
cana-6215	108	30	categories	category	NOUN
cana-6215	108	31	)	)	PUNCT
cana-6215	108	32	.	.	PUNCT
cana-6215	109	1	this	this	DET
cana-6215	109	2	approach	approach	NOUN
cana-6215	109	3	enabled	enable	VERB
cana-6215	109	4	efficient	efficient	ADJ
cana-6215	109	5	learning	learning	NOUN
cana-6215	109	6	from	from	ADP
cana-6215	109	7	limited	limited	ADJ
cana-6215	109	8	data	datum	NOUN
cana-6215	109	9	and	and	CCONJ
cana-6215	109	10	improved	improved	ADJ
cana-6215	109	11	model	model	NOUN
cana-6215	109	12	generalization	generalization	NOUN
cana-6215	109	13	.	.	PUNCT
cana-6215	110	1	figure	figure	VERB
cana-6215	110	2	6	6	NUM
cana-6215	110	3	:	:	PUNCT
cana-6215	110	4	flow	flow	NOUN
cana-6215	110	5	diagram	diagram	NOUN
cana-6215	110	6	for	for	ADP
cana-6215	110	7	proposed	propose	VERB
cana-6215	110	8	liver	liver	NOUN
cana-6215	110	9	disease	disease	NOUN
cana-6215	110	10	classification	classification	NOUN
cana-6215	110	11	framework	framework	NOUN
cana-6215	110	12	communications	communication	NOUN
cana-6215	110	13	on	on	ADP
cana-6215	110	14	applied	apply	VERB
cana-6215	110	15	nonlinear	nonlinear	ADJ
cana-6215	110	16	analysis	analysis	NOUN
cana-6215	110	17	issn	issn	NOUN
cana-6215	110	18	:	:	PUNCT
cana-6215	110	19	1074	1074	NUM
cana-6215	110	20	-	-	PUNCT
cana-6215	110	21	133x	133x	NUM
cana-6215	110	22	vol	vol	NOUN
cana-6215	110	23	32	32	NUM
cana-6215	110	24	no	no	NOUN
cana-6215	110	25	.	.	PUNCT
cana-6215	111	1	4s	4s	NUM
cana-6215	111	2	(	(	PUNCT
cana-6215	111	3	2025	2025	NUM
cana-6215	111	4	)	)	PUNCT
cana-6215	111	5	780	780	NUM
cana-6215	111	6	https://internationalpubls.com	https://internationalpubls.com	SYM
cana-6215	111	7	•	•	NUM
cana-6215	111	8	model	model	NOUN
cana-6215	111	9	training	training	NOUN
cana-6215	111	10	and	and	CCONJ
cana-6215	111	11	optimization	optimization	NOUN
cana-6215	111	12	the	the	DET
cana-6215	111	13	dataset	dataset	NOUN
cana-6215	111	14	was	be	AUX
cana-6215	111	15	divided	divide	VERB
cana-6215	111	16	into	into	ADP
cana-6215	111	17	80	80	NUM
cana-6215	111	18	%	%	NOUN
cana-6215	111	19	training	training	NOUN
cana-6215	111	20	and	and	CCONJ
cana-6215	111	21	20	20	NUM
cana-6215	111	22	%	%	NOUN
cana-6215	111	23	testing	testing	NOUN
cana-6215	111	24	subsets	subset	NOUN
cana-6215	111	25	.	.	PUNCT
cana-6215	112	1	the	the	DET
cana-6215	112	2	models	model	NOUN
cana-6215	112	3	were	be	AUX
cana-6215	112	4	trained	train	VERB
cana-6215	112	5	using	use	VERB
cana-6215	112	6	the	the	DET
cana-6215	112	7	adam	adam	PROPN
cana-6215	112	8	optimizer	optimizer	NOUN
cana-6215	112	9	with	with	ADP
cana-6215	112	10	a	a	DET
cana-6215	112	11	learning	learn	VERB
cana-6215	112	12	rate	rate	NOUN
cana-6215	112	13	of	of	ADP
cana-6215	112	14	0.0001	0.0001	NUM
cana-6215	112	15	,	,	PUNCT
cana-6215	112	16	and	and	CCONJ
cana-6215	112	17	categorical	categorical	ADJ
cana-6215	112	18	cross	cross	NOUN
cana-6215	112	19	-	-	NOUN
cana-6215	112	20	entropy	entropy	NOUN
cana-6215	112	21	was	be	AUX
cana-6215	112	22	used	use	VERB
cana-6215	112	23	as	as	ADP
cana-6215	112	24	the	the	DET
cana-6215	112	25	loss	loss	NOUN
cana-6215	112	26	function	function	NOUN
cana-6215	112	27	.	.	PUNCT
cana-6215	113	1	early	early	ADJ
cana-6215	113	2	stopping	stop	VERB
cana-6215	113	3	and	and	CCONJ
cana-6215	113	4	learning	learn	VERB
cana-6215	113	5	rate	rate	NOUN
cana-6215	113	6	scheduling	scheduling	NOUN
cana-6215	113	7	techniques	technique	NOUN
cana-6215	113	8	were	be	AUX
cana-6215	113	9	employed	employ	VERB
cana-6215	113	10	to	to	PART
cana-6215	113	11	prevent	prevent	VERB
cana-6215	113	12	overfitting	overfitting	NOUN
cana-6215	113	13	and	and	CCONJ
cana-6215	113	14	optimize	optimize	VERB
cana-6215	113	15	convergence	convergence	NOUN
cana-6215	113	16	speed	speed	NOUN
cana-6215	113	17	.	.	PUNCT
cana-6215	114	1	training	training	NOUN
cana-6215	114	2	was	be	AUX
cana-6215	114	3	carried	carry	VERB
cana-6215	114	4	out	out	ADP
cana-6215	114	5	for	for	ADP
cana-6215	114	6	50	50	NUM
cana-6215	114	7	epochs	epoch	NOUN
cana-6215	114	8	with	with	ADP
cana-6215	114	9	a	a	DET
cana-6215	114	10	batch	batch	NOUN
cana-6215	114	11	size	size	NOUN
cana-6215	114	12	of	of	ADP
cana-6215	114	13	32	32	NUM
cana-6215	114	14	.	.	PUNCT
cana-6215	115	1	all	all	DET
cana-6215	115	2	experiments	experiment	NOUN
cana-6215	115	3	were	be	AUX
cana-6215	115	4	conducted	conduct	VERB
cana-6215	115	5	on	on	ADP
cana-6215	115	6	a	a	DET
cana-6215	115	7	high	high	ADJ
cana-6215	115	8	-	-	PUNCT
cana-6215	115	9	performance	performance	NOUN
cana-6215	115	10	workstation	workstation	NOUN
cana-6215	115	11	equipped	equip	VERB
cana-6215	115	12	with	with	ADP
cana-6215	115	13	an	an	DET
cana-6215	115	14	intel	intel	PROPN
cana-6215	115	15	core	core	NOUN
cana-6215	115	16	i7	i7	NOUN
cana-6215	115	17	processor	processor	NOUN
cana-6215	115	18	,	,	PUNCT
cana-6215	115	19	16	16	NUM
cana-6215	115	20	gb	gb	NOUN
cana-6215	115	21	ram	ram	NOUN
cana-6215	115	22	,	,	PUNCT
cana-6215	115	23	and	and	CCONJ
cana-6215	115	24	an	an	DET
cana-6215	115	25	nvidia	nvidia	PROPN
cana-6215	115	26	gpu	gpu	NOUN
cana-6215	115	27	using	use	VERB
cana-6215	115	28	tensorflow	tensorflow	NOUN
cana-6215	115	29	and	and	CCONJ
cana-6215	115	30	keras	keras	PROPN
cana-6215	115	31	deep	deep	ADJ
cana-6215	115	32	learning	learning	NOUN
cana-6215	115	33	frameworks	framework	NOUN
cana-6215	115	34	.	.	PUNCT
cana-6215	116	1	5	5	X
cana-6215	116	2	.	.	X
cana-6215	116	3	performance	performance	NOUN
cana-6215	116	4	evaluation	evaluation	NOUN
cana-6215	116	5	to	to	PART
cana-6215	116	6	assess	assess	VERB
cana-6215	116	7	the	the	DET
cana-6215	116	8	classification	classification	NOUN
cana-6215	116	9	performance	performance	NOUN
cana-6215	116	10	of	of	ADP
cana-6215	116	11	the	the	DET
cana-6215	116	12	proposed	propose	VERB
cana-6215	116	13	models	model	NOUN
cana-6215	116	14	,	,	PUNCT
cana-6215	116	15	several	several	ADJ
cana-6215	116	16	evaluation	evaluation	NOUN
cana-6215	116	17	metrics	metric	NOUN
cana-6215	116	18	were	be	AUX
cana-6215	116	19	utilized	utilize	VERB
cana-6215	116	20	,	,	PUNCT
cana-6215	116	21	including	include	VERB
cana-6215	116	22	accuracy	accuracy	NOUN
cana-6215	116	23	,	,	PUNCT
cana-6215	116	24	precision	precision	NOUN
cana-6215	116	25	,	,	PUNCT
cana-6215	116	26	recall	recall	NOUN
cana-6215	116	27	(	(	PUNCT
cana-6215	116	28	sensitivity	sensitivity	NOUN
cana-6215	116	29	)	)	PUNCT
cana-6215	116	30	,	,	PUNCT
cana-6215	116	31	specificity	specificity	NOUN
cana-6215	116	32	,	,	PUNCT
cana-6215	116	33	and	and	CCONJ
cana-6215	116	34	f1	f1	NOUN
cana-6215	116	35	-	-	PUNCT
cana-6215	116	36	score	score	NOUN
cana-6215	116	37	.	.	PUNCT
cana-6215	117	1	these	these	DET
cana-6215	117	2	metrics	metric	NOUN
cana-6215	117	3	provided	provide	VERB
cana-6215	117	4	a	a	DET
cana-6215	117	5	comprehensive	comprehensive	ADJ
cana-6215	117	6	understanding	understanding	NOUN
cana-6215	117	7	of	of	ADP
cana-6215	117	8	model	model	NOUN
cana-6215	117	9	performance	performance	NOUN
cana-6215	117	10	across	across	ADP
cana-6215	117	11	different	different	ADJ
cana-6215	117	12	aspects	aspect	NOUN
cana-6215	117	13	of	of	ADP
cana-6215	117	14	classification	classification	NOUN
cana-6215	117	15	.	.	PUNCT
cana-6215	118	1	•	•	NUM
cana-6215	118	2	accuracy	accuracy	NOUN
cana-6215	118	3	accuracy	accuracy	NOUN
cana-6215	118	4	measures	measure	VERB
cana-6215	118	5	the	the	DET
cana-6215	118	6	overall	overall	ADJ
cana-6215	118	7	correctness	correctness	NOUN
cana-6215	118	8	of	of	ADP
cana-6215	118	9	the	the	DET
cana-6215	118	10	model	model	NOUN
cana-6215	118	11	’s	’s	PART
cana-6215	118	12	predictions	prediction	NOUN
cana-6215	118	13	by	by	ADP
cana-6215	118	14	calculating	calculate	VERB
cana-6215	118	15	the	the	DET
cana-6215	118	16	ratio	ratio	NOUN
cana-6215	118	17	of	of	ADP
cana-6215	118	18	correctly	correctly	ADV
cana-6215	118	19	classified	classify	VERB
cana-6215	118	20	samples	sample	NOUN
cana-6215	118	21	(	(	PUNCT
cana-6215	118	22	both	both	CCONJ
cana-6215	118	23	true	true	ADJ
cana-6215	118	24	positives	positive	NOUN
cana-6215	118	25	and	and	CCONJ
cana-6215	118	26	true	true	ADJ
cana-6215	118	27	negatives	negative	NOUN
cana-6215	118	28	)	)	PUNCT
cana-6215	118	29	to	to	ADP
cana-6215	118	30	the	the	DET
cana-6215	118	31	total	total	ADJ
cana-6215	118	32	number	number	NOUN
cana-6215	118	33	of	of	ADP
cana-6215	118	34	samples	sample	NOUN
cana-6215	118	35	.	.	PUNCT
cana-6215	119	1	it	it	PRON
cana-6215	119	2	is	be	AUX
cana-6215	119	3	a	a	DET
cana-6215	119	4	general	general	ADJ
cana-6215	119	5	indicator	indicator	NOUN
cana-6215	119	6	of	of	ADP
cana-6215	119	7	model	model	NOUN
cana-6215	119	8	performance	performance	NOUN
cana-6215	119	9	but	but	CCONJ
cana-6215	119	10	can	can	AUX
cana-6215	119	11	be	be	AUX
cana-6215	119	12	misleading	misleading	ADJ
cana-6215	119	13	when	when	SCONJ
cana-6215	119	14	dealing	deal	VERB
cana-6215	119	15	with	with	ADP
cana-6215	119	16	imbalanced	imbalanced	ADJ
cana-6215	119	17	datasets	dataset	NOUN
cana-6215	119	18	.	.	PUNCT
cana-6215	120	1	accuracy	accuracy	NOUN
cana-6215	121	1	=	=	SYM
cana-6215	121	2	(	(	PUNCT
cana-6215	121	3	tp	tp	X
cana-6215	121	4	+	+	CCONJ
cana-6215	121	5	tn	tn	NOUN
cana-6215	121	6	)	)	PUNCT
cana-6215	121	7	/	/	PUNCT
cana-6215	121	8	(	(	PUNCT
cana-6215	121	9	tp	tp	ADP
cana-6215	121	10	+	+	CCONJ
cana-6215	121	11	tn	tn	NOUN
cana-6215	121	12	+	+	CCONJ
cana-6215	121	13	fp	fp	PROPN
cana-6215	121	14	+	+	NUM
cana-6215	121	15	fn	fn	NOUN
cana-6215	121	16	)	)	PUNCT
cana-6215	121	17	•	•	NUM
cana-6215	121	18	precision	precision	NOUN
cana-6215	121	19	precision	precision	NOUN
cana-6215	121	20	indicates	indicate	VERB
cana-6215	121	21	how	how	SCONJ
cana-6215	121	22	many	many	ADJ
cana-6215	121	23	of	of	ADP
cana-6215	121	24	the	the	DET
cana-6215	121	25	samples	sample	NOUN
cana-6215	121	26	predicted	predict	VERB
cana-6215	121	27	as	as	ADP
cana-6215	121	28	positive	positive	ADJ
cana-6215	121	29	are	be	AUX
cana-6215	121	30	actually	actually	ADV
cana-6215	121	31	positive	positive	ADJ
cana-6215	121	32	.	.	PUNCT
cana-6215	122	1	it	it	PRON
cana-6215	122	2	is	be	AUX
cana-6215	122	3	an	an	DET
cana-6215	122	4	important	important	ADJ
cana-6215	122	5	metric	metric	NOUN
cana-6215	122	6	when	when	SCONJ
cana-6215	122	7	the	the	DET
cana-6215	122	8	cost	cost	NOUN
cana-6215	122	9	of	of	ADP
cana-6215	122	10	false	false	ADJ
cana-6215	122	11	positives	positive	NOUN
cana-6215	122	12	is	be	AUX
cana-6215	122	13	high	high	ADJ
cana-6215	122	14	—	—	PUNCT
cana-6215	122	15	in	in	ADP
cana-6215	122	16	this	this	DET
cana-6215	122	17	case	case	NOUN
cana-6215	122	18	,	,	PUNCT
cana-6215	122	19	misclassifying	misclassifye	VERB
cana-6215	122	20	a	a	DET
cana-6215	122	21	healthy	healthy	ADJ
cana-6215	122	22	liver	liver	NOUN
cana-6215	122	23	as	as	ADP
cana-6215	122	24	diseased	disease	VERB
cana-6215	122	25	could	could	AUX
cana-6215	122	26	lead	lead	VERB
cana-6215	122	27	to	to	ADP
cana-6215	122	28	unnecessary	unnecessary	ADJ
cana-6215	122	29	medical	medical	ADJ
cana-6215	122	30	intervention	intervention	NOUN
cana-6215	122	31	.	.	PUNCT
cana-6215	123	1	precision	precision	NOUN
cana-6215	123	2	=	=	PUNCT
cana-6215	123	3	tp	tp	NOUN
cana-6215	123	4	/	/	PUNCT
cana-6215	123	5	(	(	PUNCT
cana-6215	123	6	tp	tp	ADP
cana-6215	123	7	+	+	CCONJ
cana-6215	123	8	fp	fp	X
cana-6215	123	9	)	)	PUNCT
cana-6215	123	10	•	•	NUM
cana-6215	123	11	recall	recall	NOUN
cana-6215	123	12	(	(	PUNCT
cana-6215	123	13	sensitivity	sensitivity	NOUN
cana-6215	123	14	)	)	PUNCT
cana-6215	123	15	recall	recall	NOUN
cana-6215	123	16	,	,	PUNCT
cana-6215	123	17	also	also	ADV
cana-6215	123	18	known	know	VERB
cana-6215	123	19	as	as	ADP
cana-6215	123	20	sensitivity	sensitivity	NOUN
cana-6215	123	21	or	or	CCONJ
cana-6215	123	22	true	true	ADJ
cana-6215	123	23	positive	positive	ADJ
cana-6215	123	24	rate	rate	NOUN
cana-6215	123	25	,	,	PUNCT
cana-6215	123	26	measures	measure	VERB
cana-6215	123	27	the	the	DET
cana-6215	123	28	model	model	NOUN
cana-6215	123	29	’s	’s	PART
cana-6215	123	30	ability	ability	NOUN
cana-6215	123	31	to	to	PART
cana-6215	123	32	correctly	correctly	ADV
cana-6215	123	33	identify	identify	VERB
cana-6215	123	34	all	all	DET
cana-6215	123	35	actual	actual	ADJ
cana-6215	123	36	positive	positive	ADJ
cana-6215	123	37	cases	case	NOUN
cana-6215	123	38	.	.	PUNCT
cana-6215	124	1	in	in	ADP
cana-6215	124	2	medical	medical	ADJ
cana-6215	124	3	diagnosis	diagnosis	NOUN
cana-6215	124	4	,	,	PUNCT
cana-6215	124	5	a	a	DET
cana-6215	124	6	high	high	ADJ
cana-6215	124	7	recall	recall	NOUN
cana-6215	124	8	ensures	ensure	VERB
cana-6215	124	9	that	that	SCONJ
cana-6215	124	10	most	most	ADJ
cana-6215	124	11	cancerous	cancerous	ADJ
cana-6215	124	12	cases	case	NOUN
cana-6215	124	13	are	be	AUX
cana-6215	124	14	detected	detect	VERB
cana-6215	124	15	,	,	PUNCT
cana-6215	124	16	minimizing	minimize	VERB
cana-6215	124	17	the	the	DET
cana-6215	124	18	risk	risk	NOUN
cana-6215	124	19	of	of	ADP
cana-6215	124	20	missing	miss	VERB
cana-6215	124	21	critical	critical	ADJ
cana-6215	124	22	diagnoses	diagnosis	NOUN
cana-6215	124	23	.	.	PUNCT
cana-6215	125	1	recall	recall	NOUN
cana-6215	125	2	=	=	PROPN
cana-6215	125	3	tp	tp	NOUN
cana-6215	125	4	/	/	PUNCT
cana-6215	125	5	(	(	PUNCT
cana-6215	125	6	tp	tp	X
cana-6215	125	7	+	+	CCONJ
cana-6215	125	8	fn	fn	NOUN
cana-6215	125	9	)	)	PUNCT
cana-6215	125	10	•	•	NOUN
cana-6215	125	11	f1	f1	NOUN
cana-6215	125	12	-	-	PUNCT
cana-6215	125	13	score	score	NOUN
cana-6215	125	14	the	the	DET
cana-6215	125	15	f1	f1	NOUN
cana-6215	125	16	-	-	PUNCT
cana-6215	125	17	score	score	NOUN
cana-6215	125	18	provides	provide	VERB
cana-6215	125	19	a	a	DET
cana-6215	125	20	balance	balance	NOUN
cana-6215	125	21	between	between	ADP
cana-6215	125	22	precision	precision	NOUN
cana-6215	125	23	and	and	CCONJ
cana-6215	125	24	recall	recall	NOUN
cana-6215	125	25	,	,	PUNCT
cana-6215	125	26	offering	offer	VERB
cana-6215	125	27	a	a	DET
cana-6215	125	28	single	single	ADJ
cana-6215	125	29	performance	performance	NOUN
cana-6215	125	30	measure	measure	NOUN
cana-6215	125	31	that	that	PRON
cana-6215	125	32	considers	consider	VERB
cana-6215	125	33	both	both	DET
cana-6215	125	34	false	false	ADJ
cana-6215	125	35	positives	positive	NOUN
cana-6215	125	36	and	and	CCONJ
cana-6215	125	37	false	false	ADJ
cana-6215	125	38	negatives	negative	NOUN
cana-6215	125	39	.	.	PUNCT
cana-6215	126	1	a	a	DET
cana-6215	126	2	high	high	ADJ
cana-6215	126	3	f1	f1	NOUN
cana-6215	126	4	-	-	PUNCT
cana-6215	126	5	score	score	NOUN
cana-6215	126	6	indicates	indicate	VERB
cana-6215	126	7	that	that	SCONJ
cana-6215	126	8	the	the	DET
cana-6215	126	9	model	model	NOUN
cana-6215	126	10	maintains	maintain	VERB
cana-6215	126	11	strong	strong	ADJ
cana-6215	126	12	performance	performance	NOUN
cana-6215	126	13	across	across	ADP
cana-6215	126	14	both	both	DET
cana-6215	126	15	metrics	metric	NOUN
cana-6215	126	16	,	,	PUNCT
cana-6215	126	17	making	make	VERB
cana-6215	126	18	it	it	PRON
cana-6215	126	19	especially	especially	ADV
cana-6215	126	20	useful	useful	ADJ
cana-6215	126	21	for	for	ADP
cana-6215	126	22	imbalanced	imbalanced	ADJ
cana-6215	126	23	datasets	dataset	NOUN
cana-6215	126	24	like	like	ADP
cana-6215	126	25	medical	medical	ADJ
cana-6215	126	26	imaging	imaging	NOUN
cana-6215	126	27	data	data	PROPN
cana-6215	126	28	.	.	PUNCT
cana-6215	127	1	f1	f1	NOUN
cana-6215	127	2	-	-	PUNCT
cana-6215	127	3	score	score	NOUN
cana-6215	127	4	=	=	SYM
cana-6215	127	5	2	2	NUM
cana-6215	127	6	×	×	NOUN
cana-6215	127	7	(	(	PUNCT
cana-6215	127	8	precision	precision	NOUN
cana-6215	127	9	×	×	PROPN
cana-6215	127	10	recall	recall	NOUN
cana-6215	127	11	)	)	PUNCT
cana-6215	127	12	/	/	PUNCT
cana-6215	127	13	(	(	PUNCT
cana-6215	127	14	precision	precision	NOUN
cana-6215	127	15	+	+	CCONJ
cana-6215	127	16	recall	recall	NOUN
cana-6215	127	17	)	)	PUNCT
cana-6215	127	18	6	6	NUM
cana-6215	127	19	.	.	X
cana-6215	127	20	result	result	NOUN
cana-6215	127	21	and	and	CCONJ
cana-6215	127	22	anslysis	anslysis	NOUN
cana-6215	127	23	the	the	DET
cana-6215	127	24	proposed	propose	VERB
cana-6215	127	25	transfer	transfer	NOUN
cana-6215	127	26	learning	learning	NOUN
cana-6215	127	27	(	(	PUNCT
cana-6215	127	28	tl	tl	PROPN
cana-6215	127	29	)	)	PUNCT
cana-6215	127	30	model	model	NOUN
cana-6215	127	31	for	for	ADP
cana-6215	127	32	liver	liver	NOUN
cana-6215	127	33	cancer	cancer	NOUN
cana-6215	127	34	diagnosis	diagnosis	NOUN
cana-6215	127	35	is	be	AUX
cana-6215	127	36	systematically	systematically	ADV
cana-6215	127	37	evaluated	evaluate	VERB
cana-6215	127	38	after	after	ADP
cana-6215	127	39	each	each	DET
cana-6215	127	40	training	train	VERB
cana-6215	127	41	epoch	epoch	NOUN
cana-6215	127	42	to	to	PART
cana-6215	127	43	ensure	ensure	VERB
cana-6215	127	44	consistent	consistent	ADJ
cana-6215	127	45	improvement	improvement	NOUN
cana-6215	127	46	in	in	ADP
cana-6215	127	47	performance	performance	NOUN
cana-6215	127	48	and	and	CCONJ
cana-6215	127	49	communications	communication	NOUN
cana-6215	127	50	on	on	ADP
cana-6215	127	51	applied	apply	VERB
cana-6215	127	52	nonlinear	nonlinear	ADJ
cana-6215	127	53	analysis	analysis	NOUN
cana-6215	127	54	issn	issn	NOUN
cana-6215	127	55	:	:	PUNCT
cana-6215	127	56	1074	1074	NUM
cana-6215	127	57	-	-	PUNCT
cana-6215	127	58	133x	133x	NUM
cana-6215	127	59	vol	vol	NOUN
cana-6215	127	60	32	32	NUM
cana-6215	127	61	no	no	NOUN
cana-6215	127	62	.	.	PUNCT
cana-6215	128	1	4s	4s	NUM
cana-6215	128	2	(	(	PUNCT
cana-6215	128	3	2025	2025	NUM
cana-6215	128	4	)	)	PUNCT
cana-6215	128	5	781	781	NUM
cana-6215	128	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-6215	128	7	convergence	convergence	NOUN
cana-6215	128	8	.	.	PUNCT
cana-6215	129	1	in	in	ADP
cana-6215	129	2	this	this	DET
cana-6215	129	3	study	study	NOUN
cana-6215	129	4	,	,	PUNCT
cana-6215	129	5	three	three	NUM
cana-6215	129	6	powerful	powerful	ADJ
cana-6215	129	7	pre	pre	ADJ
cana-6215	129	8	-	-	ADJ
cana-6215	129	9	trained	train	VERB
cana-6215	129	10	deep	deep	ADJ
cana-6215	129	11	learning	learning	NOUN
cana-6215	129	12	architectures	architecture	NOUN
cana-6215	129	13	—	—	PUNCT
cana-6215	129	14	vgg-16	vgg-16	NOUN
cana-6215	129	15	,	,	PUNCT
cana-6215	129	16	resnet-50	resnet-50	PROPN
cana-6215	129	17	,	,	PUNCT
cana-6215	129	18	and	and	CCONJ
cana-6215	129	19	mobilenetv2	mobilenetv2	PROPN
cana-6215	129	20	—	—	PUNCT
cana-6215	129	21	are	be	AUX
cana-6215	129	22	employed	employ	VERB
cana-6215	129	23	to	to	PART
cana-6215	129	24	extract	extract	VERB
cana-6215	129	25	and	and	CCONJ
cana-6215	129	26	classify	classify	VERB
cana-6215	129	27	discriminative	discriminative	NOUN
cana-6215	129	28	features	feature	NOUN
cana-6215	129	29	from	from	ADP
cana-6215	129	30	ct	ct	NUM
cana-6215	129	31	liver	liver	NOUN
cana-6215	129	32	images	image	NOUN
cana-6215	129	33	.	.	PUNCT
cana-6215	130	1	each	each	DET
cana-6215	130	2	model	model	NOUN
cana-6215	130	3	is	be	AUX
cana-6215	130	4	fine	fine	ADV
cana-6215	130	5	-	-	PUNCT
cana-6215	130	6	tuned	tune	VERB
cana-6215	130	7	to	to	PART
cana-6215	130	8	adapt	adapt	VERB
cana-6215	130	9	its	its	PRON
cana-6215	130	10	learned	learn	VERB
cana-6215	130	11	representations	representation	NOUN
cana-6215	130	12	from	from	ADP
cana-6215	130	13	largescale	largescale	ADJ
cana-6215	130	14	natural	natural	ADJ
cana-6215	130	15	image	image	NOUN
cana-6215	130	16	datasets	dataset	NOUN
cana-6215	130	17	to	to	ADP
cana-6215	130	18	the	the	DET
cana-6215	130	19	medical	medical	ADJ
cana-6215	130	20	imaging	imaging	NOUN
cana-6215	130	21	domain	domain	NOUN
cana-6215	130	22	.	.	PUNCT
cana-6215	131	1	this	this	DET
cana-6215	131	2	adaptation	adaptation	NOUN
cana-6215	131	3	allows	allow	VERB
cana-6215	131	4	the	the	DET
cana-6215	131	5	models	model	NOUN
cana-6215	131	6	to	to	PART
cana-6215	131	7	capture	capture	VERB
cana-6215	131	8	subtle	subtle	ADJ
cana-6215	131	9	textural	textural	ADJ
cana-6215	131	10	and	and	CCONJ
cana-6215	131	11	structural	structural	ADJ
cana-6215	131	12	variations	variation	NOUN
cana-6215	131	13	within	within	ADP
cana-6215	131	14	liver	liver	NOUN
cana-6215	131	15	tissues	tissue	NOUN
cana-6215	131	16	,	,	PUNCT
cana-6215	131	17	thereby	thereby	ADV
cana-6215	131	18	enhancing	enhance	VERB
cana-6215	131	19	their	their	PRON
cana-6215	131	20	ability	ability	NOUN
cana-6215	131	21	to	to	PART
cana-6215	131	22	distinguish	distinguish	VERB
cana-6215	131	23	between	between	ADP
cana-6215	131	24	normal	normal	ADJ
cana-6215	131	25	,	,	PUNCT
cana-6215	131	26	benign	benign	ADJ
cana-6215	131	27	,	,	PUNCT
cana-6215	131	28	and	and	CCONJ
cana-6215	131	29	malignant	malignant	ADJ
cana-6215	131	30	liver	liver	NOUN
cana-6215	131	31	cases	case	NOUN
cana-6215	131	32	.	.	PUNCT
cana-6215	132	1	by	by	ADP
cana-6215	132	2	freezing	freeze	VERB
cana-6215	132	3	the	the	DET
cana-6215	132	4	initial	initial	ADJ
cana-6215	132	5	layers	layer	NOUN
cana-6215	132	6	and	and	CCONJ
cana-6215	132	7	retraining	retrain	VERB
cana-6215	132	8	the	the	DET
cana-6215	132	9	deeper	deep	ADJ
cana-6215	132	10	layers	layer	NOUN
cana-6215	132	11	,	,	PUNCT
cana-6215	132	12	the	the	DET
cana-6215	132	13	tl	tl	PROPN
cana-6215	132	14	models	model	NOUN
cana-6215	132	15	efficiently	efficiently	ADV
cana-6215	132	16	reuse	reuse	VERB
cana-6215	132	17	previously	previously	ADV
cana-6215	132	18	learned	learn	VERB
cana-6215	132	19	features	feature	NOUN
cana-6215	132	20	while	while	SCONJ
cana-6215	132	21	specializing	specialize	VERB
cana-6215	132	22	in	in	ADP
cana-6215	132	23	detecting	detect	VERB
cana-6215	132	24	patterns	pattern	NOUN
cana-6215	132	25	specific	specific	ADJ
cana-6215	132	26	to	to	ADP
cana-6215	132	27	liver	liver	NOUN
cana-6215	132	28	tumors	tumor	NOUN
cana-6215	132	29	.	.	PUNCT
cana-6215	133	1	the	the	DET
cana-6215	133	2	experimental	experimental	ADJ
cana-6215	133	3	evaluation	evaluation	NOUN
cana-6215	133	4	demonstrates	demonstrate	VERB
cana-6215	133	5	that	that	SCONJ
cana-6215	133	6	the	the	DET
cana-6215	133	7	proposed	propose	VERB
cana-6215	133	8	tl	tl	PROPN
cana-6215	133	9	-	-	PUNCT
cana-6215	133	10	based	base	VERB
cana-6215	133	11	framework	framework	NOUN
cana-6215	133	12	achieves	achieve	VERB
cana-6215	133	13	remarkable	remarkable	ADJ
cana-6215	133	14	diagnostic	diagnostic	ADJ
cana-6215	133	15	performance	performance	NOUN
cana-6215	133	16	across	across	ADP
cana-6215	133	17	all	all	DET
cana-6215	133	18	tested	test	VERB
cana-6215	133	19	models	model	NOUN
cana-6215	133	20	.	.	PUNCT
cana-6215	134	1	specifically	specifically	ADV
cana-6215	134	2	,	,	PUNCT
cana-6215	134	3	the	the	DET
cana-6215	134	4	vgg-16	vgg-16	NOUN
cana-6215	134	5	and	and	CCONJ
cana-6215	134	6	mobilenetv2	mobilenetv2	PROPN
cana-6215	134	7	architectures	architecture	NOUN
cana-6215	134	8	yield	yield	VERB
cana-6215	134	9	a	a	DET
cana-6215	134	10	high	high	ADJ
cana-6215	134	11	classification	classification	NOUN
cana-6215	134	12	accuracy	accuracy	NOUN
cana-6215	134	13	of	of	ADP
cana-6215	134	14	99.10	99.10	NUM
cana-6215	134	15	%	%	NOUN
cana-6215	134	16	,	,	PUNCT
cana-6215	134	17	indicating	indicate	VERB
cana-6215	134	18	their	their	PRON
cana-6215	134	19	effectiveness	effectiveness	NOUN
cana-6215	134	20	in	in	ADP
cana-6215	134	21	identifying	identify	VERB
cana-6215	134	22	liver	liver	NOUN
cana-6215	134	23	abnormalities	abnormality	NOUN
cana-6215	134	24	with	with	ADP
cana-6215	134	25	minimal	minimal	ADJ
cana-6215	134	26	error	error	NOUN
cana-6215	134	27	.	.	PUNCT
cana-6215	135	1	this	this	DET
cana-6215	135	2	superior	superior	ADJ
cana-6215	135	3	performance	performance	NOUN
cana-6215	135	4	underscores	underscore	VERB
cana-6215	135	5	the	the	DET
cana-6215	135	6	strength	strength	NOUN
cana-6215	135	7	of	of	ADP
cana-6215	135	8	tl	tl	PROPN
cana-6215	135	9	in	in	ADP
cana-6215	135	10	medical	medical	ADJ
cana-6215	135	11	image	image	NOUN
cana-6215	135	12	classification	classification	NOUN
cana-6215	135	13	,	,	PUNCT
cana-6215	135	14	particularly	particularly	ADV
cana-6215	135	15	when	when	SCONJ
cana-6215	135	16	working	work	VERB
cana-6215	135	17	with	with	ADP
cana-6215	135	18	limited	limited	ADJ
cana-6215	135	19	datasets	dataset	NOUN
cana-6215	135	20	.	.	PUNCT
cana-6215	136	1	the	the	DET
cana-6215	136	2	ability	ability	NOUN
cana-6215	136	3	of	of	ADP
cana-6215	136	4	these	these	DET
cana-6215	136	5	models	model	NOUN
cana-6215	136	6	to	to	PART
cana-6215	136	7	generalize	generalize	VERB
cana-6215	136	8	well	well	ADV
cana-6215	136	9	,	,	PUNCT
cana-6215	136	10	despite	despite	SCONJ
cana-6215	136	11	the	the	DET
cana-6215	136	12	small	small	ADJ
cana-6215	136	13	dataset	dataset	NOUN
cana-6215	136	14	size	size	NOUN
cana-6215	136	15	,	,	PUNCT
cana-6215	136	16	highlights	highlight	VERB
cana-6215	136	17	their	their	PRON
cana-6215	136	18	robustness	robustness	NOUN
cana-6215	136	19	and	and	CCONJ
cana-6215	136	20	adaptability	adaptability	NOUN
cana-6215	136	21	(	(	PUNCT
cana-6215	136	22	table	table	NOUN
cana-6215	136	23	3	3	NUM
cana-6215	136	24	)	)	PUNCT
cana-6215	136	25	.	.	PUNCT
cana-6215	137	1	table	table	NOUN
cana-6215	137	2	3	3	NUM
cana-6215	137	3	:	:	PUNCT
cana-6215	137	4	performance	performance	NOUN
cana-6215	137	5	evaluation	evaluation	NOUN
cana-6215	137	6	of	of	ADP
cana-6215	137	7	proposed	propose	VERB
cana-6215	137	8	model	model	NOUN
cana-6215	137	9	with	with	ADP
cana-6215	137	10	existing	exist	VERB
cana-6215	137	11	models	model	NOUN
cana-6215	137	12	model	model	NOUN
cana-6215	137	13	accuracy	accuracy	PROPN
cana-6215	137	14	recall	recall	PROPN
cana-6215	137	15	precision	precision	NOUN
cana-6215	137	16	f1	f1	NOUN
cana-6215	137	17	-	-	PUNCT
cana-6215	137	18	score	score	NOUN
cana-6215	137	19	proposed	propose	VERB
cana-6215	137	20	model	model	NOUN
cana-6215	137	21	99.33	99.33	NUM
cana-6215	137	22	%	%	NOUN
cana-6215	137	23	99.25	99.25	NUM
cana-6215	137	24	%	%	NOUN
cana-6215	137	25	99.25	99.25	NUM
cana-6215	137	26	%	%	NOUN
cana-6215	137	27	99.50	99.50	NUM
cana-6215	137	28	%	%	NOUN
cana-6215	137	29	vgg16	vgg16	NOUN
cana-6215	137	30	91.23	91.23	NUM
cana-6215	137	31	%	%	NOUN
cana-6215	137	32	91.25	91.25	NUM
cana-6215	137	33	%	%	NOUN
cana-6215	137	34	92.0	92.0	NUM
cana-6215	137	35	%	%	NOUN
cana-6215	137	36	91.50	91.50	NUM
cana-6215	137	37	%	%	NOUN
cana-6215	137	38	inceptionv3	inceptionv3	NOUN
cana-6215	138	1	86.74	86.74	NUM
cana-6215	138	2	%	%	NOUN
cana-6215	138	3	86.75	86.75	NUM
cana-6215	138	4	%	%	NOUN
cana-6215	138	5	88.0	88.0	NUM
cana-6215	138	6	%	%	NOUN
cana-6215	138	7	87.0	87.0	NUM
cana-6215	138	8	%	%	NOUN
cana-6215	138	9	densenet121	densenet121	PROPN
cana-6215	138	10	93.69	93.69	NUM
cana-6215	138	11	%	%	NOUN
cana-6215	138	12	93.5	93.5	NUM
cana-6215	138	13	%	%	NOUN
cana-6215	138	14	94.0	94.0	NUM
cana-6215	138	15	%	%	NOUN
cana-6215	138	16	93.75	93.75	NUM
cana-6215	138	17	%	%	NOUN
cana-6215	138	18	resnet50	resnet50	NOUN
cana-6215	138	19	91.55	91.55	NUM
cana-6215	138	20	%	%	NOUN
cana-6215	138	21	91.5	91.5	NUM
cana-6215	138	22	%	%	NOUN
cana-6215	138	23	91.75	91.75	NUM
cana-6215	138	24	%	%	NOUN
cana-6215	138	25	91.50	91.50	NUM
cana-6215	138	26	%	%	NOUN
cana-6215	138	27	cnn	cnn	PROPN
cana-6215	138	28	model	model	NOUN
cana-6215	138	29	94.15	94.15	NUM
cana-6215	138	30	%	%	NOUN
cana-6215	138	31	94.52	94.52	NUM
cana-6215	138	32	%	%	NOUN
cana-6215	138	33	94.75	94.75	NUM
cana-6215	138	34	%	%	NOUN
cana-6215	138	35	94.50	94.50	NUM
cana-6215	138	36	%	%	NOUN
cana-6215	138	37	the	the	DET
cana-6215	138	38	comparative	comparative	ADJ
cana-6215	138	39	performance	performance	NOUN
cana-6215	138	40	evaluation	evaluation	NOUN
cana-6215	138	41	of	of	ADP
cana-6215	138	42	various	various	ADJ
cana-6215	138	43	deep	deep	ADJ
cana-6215	138	44	learning	learning	NOUN
cana-6215	138	45	models	model	NOUN
cana-6215	138	46	demonstrates	demonstrate	VERB
cana-6215	138	47	that	that	SCONJ
cana-6215	138	48	the	the	DET
cana-6215	138	49	proposed	propose	VERB
cana-6215	138	50	model	model	NOUN
cana-6215	138	51	significantly	significantly	ADV
cana-6215	138	52	outperforms	outperform	VERB
cana-6215	138	53	other	other	ADJ
cana-6215	138	54	architectures	architecture	NOUN
cana-6215	138	55	in	in	ADP
cana-6215	138	56	the	the	DET
cana-6215	138	57	classification	classification	NOUN
cana-6215	138	58	of	of	ADP
cana-6215	138	59	liver	liver	NOUN
cana-6215	138	60	cancer	cancer	NOUN
cana-6215	138	61	.	.	PUNCT
cana-6215	139	1	with	with	ADP
cana-6215	139	2	an	an	DET
cana-6215	139	3	impressive	impressive	ADJ
cana-6215	139	4	accuracy	accuracy	NOUN
cana-6215	139	5	of	of	ADP
cana-6215	139	6	99.33	99.33	NUM
cana-6215	139	7	%	%	NOUN
cana-6215	139	8	,	,	PUNCT
cana-6215	139	9	recall	recall	NOUN
cana-6215	139	10	of	of	ADP
cana-6215	139	11	99.25	99.25	NUM
cana-6215	139	12	%	%	NOUN
cana-6215	139	13	,	,	PUNCT
cana-6215	139	14	precision	precision	NOUN
cana-6215	139	15	of	of	ADP
cana-6215	139	16	99.25	99.25	NUM
cana-6215	139	17	%	%	NOUN
cana-6215	139	18	,	,	PUNCT
cana-6215	139	19	and	and	CCONJ
cana-6215	139	20	an	an	DET
cana-6215	139	21	f1	f1	NOUN
cana-6215	139	22	-	-	PUNCT
cana-6215	139	23	score	score	NOUN
cana-6215	139	24	of	of	ADP
cana-6215	139	25	99.50	99.50	NUM
cana-6215	139	26	%	%	NOUN
cana-6215	139	27	,	,	PUNCT
cana-6215	139	28	the	the	DET
cana-6215	139	29	proposed	propose	VERB
cana-6215	139	30	model	model	NOUN
cana-6215	139	31	exhibits	exhibit	VERB
cana-6215	139	32	superior	superior	ADJ
cana-6215	139	33	predictive	predictive	ADJ
cana-6215	139	34	capability	capability	NOUN
cana-6215	139	35	and	and	CCONJ
cana-6215	139	36	robustness	robustness	NOUN
cana-6215	139	37	.	.	PUNCT
cana-6215	140	1	this	this	DET
cana-6215	140	2	exceptional	exceptional	ADJ
cana-6215	140	3	performance	performance	NOUN
cana-6215	140	4	highlights	highlight	VERB
cana-6215	140	5	the	the	DET
cana-6215	140	6	effectiveness	effectiveness	NOUN
cana-6215	140	7	of	of	ADP
cana-6215	140	8	the	the	DET
cana-6215	140	9	proposed	propose	VERB
cana-6215	140	10	transfer	transfer	NOUN
cana-6215	140	11	learning	learn	VERB
cana-6215	140	12	framework	framework	NOUN
cana-6215	140	13	and	and	CCONJ
cana-6215	140	14	its	its	PRON
cana-6215	140	15	fine	fine	ADV
cana-6215	140	16	-	-	PUNCT
cana-6215	140	17	tuning	tune	VERB
cana-6215	140	18	process	process	NOUN
cana-6215	140	19	in	in	ADP
cana-6215	140	20	extracting	extract	VERB
cana-6215	140	21	high	high	ADJ
cana-6215	140	22	-	-	PUNCT
cana-6215	140	23	level	level	NOUN
cana-6215	140	24	,	,	PUNCT
cana-6215	140	25	discriminative	discriminative	NOUN
cana-6215	140	26	features	feature	NOUN
cana-6215	140	27	from	from	ADP
cana-6215	140	28	ct	ct	NUM
cana-6215	140	29	scan	scan	ADJ
cana-6215	140	30	images	image	NOUN
cana-6215	140	31	.	.	PUNCT
cana-6215	141	1	by	by	ADP
cana-6215	141	2	leveraging	leverage	VERB
cana-6215	141	3	optimized	optimize	VERB
cana-6215	141	4	feature	feature	NOUN
cana-6215	141	5	representations	representation	NOUN
cana-6215	141	6	and	and	CCONJ
cana-6215	141	7	advanced	advanced	ADJ
cana-6215	141	8	pre	pre	ADJ
cana-6215	141	9	-	-	ADJ
cana-6215	141	10	processing	processing	ADJ
cana-6215	141	11	techniques	technique	NOUN
cana-6215	141	12	,	,	PUNCT
cana-6215	141	13	the	the	DET
cana-6215	141	14	proposed	propose	VERB
cana-6215	141	15	model	model	NOUN
cana-6215	141	16	achieves	achieve	VERB
cana-6215	141	17	high	high	ADJ
cana-6215	141	18	generalization	generalization	NOUN
cana-6215	141	19	even	even	ADV
cana-6215	141	20	with	with	ADP
cana-6215	141	21	a	a	DET
cana-6215	141	22	limited	limited	ADJ
cana-6215	141	23	dataset	dataset	NOUN
cana-6215	141	24	,	,	PUNCT
cana-6215	141	25	reducing	reduce	VERB
cana-6215	141	26	both	both	DET
cana-6215	141	27	false	false	ADJ
cana-6215	141	28	positives	positive	NOUN
cana-6215	141	29	and	and	CCONJ
cana-6215	141	30	false	false	ADJ
cana-6215	141	31	negatives	negative	NOUN
cana-6215	141	32	.	.	PUNCT
cana-6215	142	1	such	such	ADJ
cana-6215	142	2	precision	precision	NOUN
cana-6215	142	3	is	be	AUX
cana-6215	142	4	particularly	particularly	ADV
cana-6215	142	5	crucial	crucial	ADJ
cana-6215	142	6	in	in	ADP
cana-6215	142	7	medical	medical	ADJ
cana-6215	142	8	image	image	NOUN
cana-6215	142	9	diagnosis	diagnosis	NOUN
cana-6215	142	10	,	,	PUNCT
cana-6215	142	11	where	where	SCONJ
cana-6215	142	12	even	even	ADV
cana-6215	142	13	a	a	DET
cana-6215	142	14	small	small	ADJ
cana-6215	142	15	error	error	NOUN
cana-6215	142	16	could	could	AUX
cana-6215	142	17	lead	lead	VERB
cana-6215	142	18	to	to	ADP
cana-6215	142	19	misdiagnosis	misdiagnosis	NOUN
cana-6215	142	20	and	and	CCONJ
cana-6215	142	21	improper	improper	ADJ
cana-6215	142	22	treatment	treatment	NOUN
cana-6215	142	23	(	(	PUNCT
cana-6215	142	24	table	table	NOUN
cana-6215	142	25	3	3	NUM
cana-6215	142	26	and	and	CCONJ
cana-6215	142	27	figure	figure	VERB
cana-6215	142	28	8)	8)	NUM
cana-6215	142	29	.	.	PUNCT
cana-6215	143	1	communications	communication	NOUN
cana-6215	143	2	on	on	ADP
cana-6215	143	3	applied	apply	VERB
cana-6215	143	4	nonlinear	nonlinear	ADJ
cana-6215	143	5	analysis	analysis	NOUN
cana-6215	143	6	issn	issn	NOUN
cana-6215	143	7	:	:	PUNCT
cana-6215	143	8	1074	1074	NUM
cana-6215	143	9	-	-	PUNCT
cana-6215	143	10	133x	133x	NUM
cana-6215	143	11	vol	vol	NOUN
cana-6215	143	12	32	32	NUM
cana-6215	143	13	no	no	NOUN
cana-6215	143	14	.	.	PUNCT
cana-6215	144	1	4s	4s	NUM
cana-6215	144	2	(	(	PUNCT
cana-6215	144	3	2025	2025	NUM
cana-6215	144	4	)	)	PUNCT
cana-6215	144	5	782	782	NUM
cana-6215	145	1	https://internationalpubls.com	https://internationalpubls.com	X
cana-6215	145	2	figure	figure	NOUN
cana-6215	145	3	8	8	NUM
cana-6215	145	4	:	:	PUNCT
cana-6215	145	5	performance	performance	NOUN
cana-6215	145	6	evaluation	evaluation	NOUN
cana-6215	145	7	of	of	ADP
cana-6215	145	8	proposed	propose	VERB
cana-6215	145	9	model	model	NOUN
cana-6215	145	10	with	with	ADP
cana-6215	145	11	existing	exist	VERB
cana-6215	145	12	models	model	NOUN
cana-6215	145	13	in	in	ADP
cana-6215	145	14	comparison	comparison	NOUN
cana-6215	145	15	,	,	PUNCT
cana-6215	145	16	other	other	ADJ
cana-6215	145	17	models	model	NOUN
cana-6215	145	18	such	such	ADJ
cana-6215	145	19	as	as	ADP
cana-6215	145	20	vgg16	vgg16	PROPN
cana-6215	145	21	,	,	PUNCT
cana-6215	145	22	resnet50	resnet50	NOUN
cana-6215	145	23	,	,	PUNCT
cana-6215	145	24	densenet121	densenet121	PROPN
cana-6215	145	25	,	,	PUNCT
cana-6215	145	26	and	and	CCONJ
cana-6215	145	27	inceptionv3	inceptionv3	NOUN
cana-6215	145	28	also	also	ADV
cana-6215	145	29	perform	perform	VERB
cana-6215	145	30	well	well	ADV
cana-6215	145	31	but	but	CCONJ
cana-6215	145	32	fall	fall	VERB
cana-6215	145	33	short	short	ADV
cana-6215	145	34	of	of	ADP
cana-6215	145	35	the	the	DET
cana-6215	145	36	proposed	propose	VERB
cana-6215	145	37	approach	approach	NOUN
cana-6215	145	38	.	.	PUNCT
cana-6215	146	1	for	for	ADP
cana-6215	146	2	instance	instance	NOUN
cana-6215	146	3	,	,	PUNCT
cana-6215	146	4	densenet121	densenet121	PROPN
cana-6215	146	5	achieves	achieve	VERB
cana-6215	146	6	an	an	DET
cana-6215	146	7	accuracy	accuracy	NOUN
cana-6215	146	8	of	of	ADP
cana-6215	146	9	93.69	93.69	NUM
cana-6215	146	10	%	%	NOUN
cana-6215	146	11	,	,	PUNCT
cana-6215	146	12	while	while	SCONJ
cana-6215	146	13	vgg16	vgg16	NOUN
cana-6215	146	14	and	and	CCONJ
cana-6215	146	15	resnet50	resnet50	NOUN
cana-6215	146	16	show	show	VERB
cana-6215	146	17	similar	similar	ADJ
cana-6215	146	18	performances	performance	NOUN
cana-6215	146	19	around	around	ADP
cana-6215	146	20	91	91	NUM
cana-6215	146	21	%	%	NOUN
cana-6215	146	22	,	,	PUNCT
cana-6215	146	23	indicating	indicate	VERB
cana-6215	146	24	their	their	PRON
cana-6215	146	25	ability	ability	NOUN
cana-6215	146	26	to	to	PART
cana-6215	146	27	capture	capture	VERB
cana-6215	146	28	essential	essential	ADJ
cana-6215	146	29	image	image	NOUN
cana-6215	146	30	features	feature	NOUN
cana-6215	146	31	but	but	CCONJ
cana-6215	146	32	with	with	ADP
cana-6215	146	33	relatively	relatively	ADV
cana-6215	146	34	lower	low	ADJ
cana-6215	146	35	sensitivity	sensitivity	NOUN
cana-6215	146	36	.	.	PUNCT
cana-6215	147	1	the	the	DET
cana-6215	147	2	conventional	conventional	ADJ
cana-6215	147	3	cnn	cnn	PROPN
cana-6215	147	4	model	model	NOUN
cana-6215	147	5	demonstrates	demonstrate	VERB
cana-6215	147	6	moderate	moderate	ADJ
cana-6215	147	7	effectiveness	effectiveness	NOUN
cana-6215	147	8	with	with	ADP
cana-6215	147	9	94.15	94.15	NUM
cana-6215	147	10	%	%	NOUN
cana-6215	147	11	accuracy	accuracy	NOUN
cana-6215	147	12	,	,	PUNCT
cana-6215	147	13	validating	validate	VERB
cana-6215	147	14	its	its	PRON
cana-6215	147	15	potential	potential	NOUN
cana-6215	147	16	yet	yet	ADV
cana-6215	147	17	emphasizing	emphasize	VERB
cana-6215	147	18	the	the	DET
cana-6215	147	19	need	need	NOUN
cana-6215	147	20	for	for	ADP
cana-6215	147	21	deeper	deep	ADJ
cana-6215	147	22	architectures	architecture	NOUN
cana-6215	147	23	with	with	ADP
cana-6215	147	24	enhanced	enhanced	ADJ
cana-6215	147	25	transfer	transfer	NOUN
cana-6215	147	26	learning	learn	VERB
cana-6215	147	27	capabilities	capability	NOUN
cana-6215	147	28	.	.	PUNCT
cana-6215	148	1	7	7	X
cana-6215	148	2	.	.	X
cana-6215	148	3	conclusion	conclusion	NOUN
cana-6215	148	4	the	the	DET
cana-6215	148	5	proposed	propose	VERB
cana-6215	148	6	study	study	NOUN
cana-6215	148	7	presents	present	VERB
cana-6215	148	8	an	an	DET
cana-6215	148	9	efficient	efficient	ADJ
cana-6215	148	10	deep	deep	ADJ
cana-6215	148	11	transfer	transfer	NOUN
cana-6215	148	12	learning	learning	NOUN
cana-6215	148	13	-	-	PUNCT
cana-6215	148	14	based	base	VERB
cana-6215	148	15	framework	framework	NOUN
cana-6215	148	16	for	for	ADP
cana-6215	148	17	accurate	accurate	ADJ
cana-6215	148	18	liver	liver	NOUN
cana-6215	148	19	disease	disease	NOUN
cana-6215	148	20	prediction	prediction	NOUN
cana-6215	148	21	,	,	PUNCT
cana-6215	148	22	demonstrating	demonstrate	VERB
cana-6215	148	23	the	the	DET
cana-6215	148	24	potential	potential	NOUN
cana-6215	148	25	of	of	ADP
cana-6215	148	26	advanced	advanced	ADJ
cana-6215	148	27	artificial	artificial	ADJ
cana-6215	148	28	intelligence	intelligence	NOUN
cana-6215	148	29	techniques	technique	NOUN
cana-6215	148	30	in	in	ADP
cana-6215	148	31	transforming	transform	VERB
cana-6215	148	32	clinical	clinical	ADJ
cana-6215	148	33	diagnostics	diagnostic	NOUN
cana-6215	148	34	.	.	PUNCT
cana-6215	149	1	by	by	ADP
cana-6215	149	2	leveraging	leverage	VERB
cana-6215	149	3	state	state	NOUN
cana-6215	149	4	-	-	PUNCT
cana-6215	149	5	of	of	ADP
cana-6215	149	6	-	-	PUNCT
cana-6215	149	7	the	the	DET
cana-6215	149	8	-	-	PUNCT
cana-6215	149	9	art	art	NOUN
cana-6215	149	10	convolutional	convolutional	ADJ
cana-6215	149	11	neural	neural	ADJ
cana-6215	149	12	network	network	NOUN
cana-6215	149	13	architectures	architecture	NOUN
cana-6215	149	14	such	such	ADJ
cana-6215	149	15	as	as	ADP
cana-6215	149	16	efficientnetb2	efficientnetb2	PROPN
cana-6215	149	17	,	,	PUNCT
cana-6215	149	18	densenet121	densenet121	PROPN
cana-6215	149	19	,	,	PUNCT
cana-6215	149	20	inceptionv3	inceptionv3	NOUN
cana-6215	149	21	,	,	PUNCT
cana-6215	149	22	resnet50	resnet50	NOUN
cana-6215	149	23	,	,	PUNCT
cana-6215	149	24	and	and	CCONJ
cana-6215	149	25	vgg16	vgg16	PROPN
cana-6215	149	26	,	,	PUNCT
cana-6215	149	27	the	the	DET
cana-6215	149	28	framework	framework	NOUN
cana-6215	149	29	effectively	effectively	ADV
cana-6215	149	30	automates	automate	VERB
cana-6215	149	31	the	the	DET
cana-6215	149	32	classification	classification	NOUN
cana-6215	149	33	of	of	ADP
cana-6215	149	34	major	major	ADJ
cana-6215	149	35	liver	liver	NOUN
cana-6215	149	36	disease	disease	NOUN
cana-6215	149	37	categories	category	NOUN
cana-6215	149	38	—	—	PUNCT
cana-6215	149	39	ballooning	ballooning	NOUN
cana-6215	149	40	,	,	PUNCT
cana-6215	149	41	fibrosis	fibrosis	NOUN
cana-6215	149	42	,	,	PUNCT
cana-6215	149	43	inflammation	inflammation	NOUN
cana-6215	149	44	,	,	PUNCT
cana-6215	149	45	and	and	CCONJ
cana-6215	149	46	steatosis	steatosis	NOUN
cana-6215	149	47	—	—	PUNCT
cana-6215	149	48	using	use	VERB
cana-6215	149	49	histopathological	histopathological	ADJ
cana-6215	149	50	images	image	NOUN
cana-6215	149	51	.	.	PUNCT
cana-6215	150	1	through	through	ADP
cana-6215	150	2	rigorous	rigorous	ADJ
cana-6215	150	3	data	datum	NOUN
cana-6215	150	4	preprocessing	preprocessing	NOUN
cana-6215	150	5	,	,	PUNCT
cana-6215	150	6	transfer	transfer	NOUN
cana-6215	150	7	learning	learning	NOUN
cana-6215	150	8	,	,	PUNCT
cana-6215	150	9	and	and	CCONJ
cana-6215	150	10	fine	fine	ADV
cana-6215	150	11	-	-	PUNCT
cana-6215	150	12	tuning	tuning	NOUN
cana-6215	150	13	,	,	PUNCT
cana-6215	150	14	the	the	DET
cana-6215	150	15	developed	develop	VERB
cana-6215	150	16	model	model	NOUN
cana-6215	150	17	achieves	achieve	VERB
cana-6215	150	18	enhanced	enhance	VERB
cana-6215	150	19	accuracy	accuracy	NOUN
cana-6215	150	20	,	,	PUNCT
cana-6215	150	21	robustness	robustness	NOUN
cana-6215	150	22	,	,	PUNCT
cana-6215	150	23	and	and	CCONJ
cana-6215	150	24	generalization	generalization	NOUN
cana-6215	150	25	,	,	PUNCT
cana-6215	150	26	outperforming	outperform	VERB
cana-6215	150	27	traditional	traditional	ADJ
cana-6215	150	28	diagnostic	diagnostic	ADJ
cana-6215	150	29	methods	method	NOUN
cana-6215	150	30	that	that	PRON
cana-6215	150	31	are	be	AUX
cana-6215	150	32	often	often	ADV
cana-6215	150	33	subjective	subjective	ADJ
cana-6215	150	34	and	and	CCONJ
cana-6215	150	35	time	time	NOUN
cana-6215	150	36	-	-	PUNCT
cana-6215	150	37	intensive	intensive	ADJ
cana-6215	150	38	.	.	PUNCT
cana-6215	151	1	the	the	DET
cana-6215	151	2	results	result	NOUN
cana-6215	151	3	confirm	confirm	VERB
cana-6215	151	4	that	that	SCONJ
cana-6215	151	5	deep	deep	ADJ
cana-6215	151	6	transfer	transfer	NOUN
cana-6215	151	7	learning	learning	NOUN
cana-6215	151	8	models	model	NOUN
cana-6215	151	9	can	can	AUX
cana-6215	151	10	efficiently	efficiently	ADV
cana-6215	151	11	extract	extract	VERB
cana-6215	151	12	complex	complex	ADJ
cana-6215	151	13	tissue	tissue	NOUN
cana-6215	151	14	patterns	pattern	NOUN
cana-6215	151	15	,	,	PUNCT
cana-6215	151	16	reduce	reduce	VERB
cana-6215	151	17	diagnostic	diagnostic	ADJ
cana-6215	151	18	variability	variability	NOUN
cana-6215	151	19	,	,	PUNCT
cana-6215	151	20	and	and	CCONJ
cana-6215	151	21	assist	assist	VERB
cana-6215	151	22	pathologists	pathologist	NOUN
cana-6215	151	23	in	in	ADP
cana-6215	151	24	making	make	VERB
cana-6215	151	25	faster	fast	ADV
cana-6215	151	26	and	and	CCONJ
cana-6215	151	27	more	more	ADV
cana-6215	151	28	reliable	reliable	ADJ
cana-6215	151	29	decisions	decision	NOUN
cana-6215	151	30	.	.	PUNCT
cana-6215	152	1	moreover	moreover	ADV
cana-6215	152	2	,	,	PUNCT
cana-6215	152	3	the	the	DET
cana-6215	152	4	incorporation	incorporation	NOUN
cana-6215	152	5	of	of	ADP
cana-6215	152	6	explainable	explainable	ADJ
cana-6215	152	7	ai	ai	NOUN
cana-6215	152	8	techniques	technique	NOUN
cana-6215	152	9	such	such	ADJ
cana-6215	152	10	as	as	ADP
cana-6215	152	11	grad	grad	NOUN
cana-6215	152	12	-	-	PUNCT
cana-6215	152	13	cam	cam	NOUN
cana-6215	152	14	ensures	ensure	VERB
cana-6215	152	15	transparency	transparency	NOUN
cana-6215	152	16	and	and	CCONJ
cana-6215	152	17	interpretability	interpretability	NOUN
cana-6215	152	18	,	,	PUNCT
cana-6215	152	19	making	make	VERB
cana-6215	152	20	the	the	DET
cana-6215	152	21	system	system	NOUN
cana-6215	152	22	more	more	ADV
cana-6215	152	23	trustworthy	trustworthy	ADJ
cana-6215	152	24	in	in	ADP
cana-6215	152	25	clinical	clinical	ADJ
cana-6215	152	26	settings	setting	NOUN
cana-6215	152	27	.	.	PUNCT
cana-6215	153	1	this	this	DET
cana-6215	153	2	research	research	NOUN
cana-6215	153	3	not	not	PART
cana-6215	153	4	only	only	ADV
cana-6215	153	5	establishes	establish	VERB
cana-6215	153	6	a	a	DET
cana-6215	153	7	scalable	scalable	ADJ
cana-6215	153	8	and	and	CCONJ
cana-6215	153	9	reliable	reliable	ADJ
cana-6215	153	10	diagnostic	diagnostic	ADJ
cana-6215	153	11	framework	framework	NOUN
cana-6215	153	12	but	but	CCONJ
cana-6215	153	13	also	also	ADV
cana-6215	153	14	paves	pave	VERB
cana-6215	153	15	the	the	DET
cana-6215	153	16	way	way	NOUN
cana-6215	153	17	for	for	ADP
cana-6215	153	18	future	future	ADJ
cana-6215	153	19	advancements	advancement	NOUN
cana-6215	153	20	in	in	ADP
cana-6215	153	21	ai	ai	ADJ
cana-6215	153	22	-	-	PUNCT
cana-6215	153	23	driven	drive	VERB
cana-6215	153	24	medical	medical	ADJ
cana-6215	153	25	imaging	imaging	NOUN
cana-6215	153	26	,	,	PUNCT
cana-6215	153	27	supporting	support	VERB
cana-6215	153	28	early	early	ADJ
cana-6215	153	29	detection	detection	NOUN
cana-6215	153	30	,	,	PUNCT
cana-6215	153	31	improved	improve	VERB
cana-6215	153	32	treatment	treatment	NOUN
cana-6215	153	33	planning	planning	NOUN
cana-6215	153	34	,	,	PUNCT
cana-6215	153	35	and	and	CCONJ
cana-6215	153	36	ultimately	ultimately	ADV
cana-6215	153	37	better	well	ADJ
cana-6215	153	38	patient	patient	ADJ
cana-6215	153	39	outcomes	outcome	NOUN
cana-6215	153	40	in	in	ADP
cana-6215	153	41	liver	liver	NOUN
cana-6215	153	42	disease	disease	NOUN
cana-6215	153	43	management	management	NOUN
cana-6215	153	44	.	.	PUNCT
cana-6215	154	1	references	reference	NOUN
cana-6215	154	2	[	[	X
cana-6215	154	3	1	1	NUM
cana-6215	154	4	]	]	PUNCT
cana-6215	154	5	r.	r.	PROPN
cana-6215	154	6	sharma	sharma	PROPN
cana-6215	154	7	,	,	PUNCT
cana-6215	154	8	a.	a.	PROPN
cana-6215	154	9	verma	verma	PROPN
cana-6215	154	10	,	,	PUNCT
cana-6215	154	11	and	and	CCONJ
cana-6215	154	12	k.	k.	PROPN
cana-6215	154	13	gupta	gupta	PROPN
cana-6215	154	14	,	,	PUNCT
cana-6215	154	15	“	"	PUNCT
cana-6215	154	16	cnn	cnn	PROPN
cana-6215	154	17	-	-	PUNCT
cana-6215	154	18	based	base	VERB
cana-6215	154	19	liver	liver	NOUN
cana-6215	154	20	tumor	tumor	NOUN
cana-6215	154	21	segmentation	segmentation	NOUN
cana-6215	154	22	and	and	CCONJ
cana-6215	154	23	classification	classification	NOUN
cana-6215	154	24	from	from	ADP
cana-6215	154	25	ct	ct	NUM
cana-6215	154	26	images	image	NOUN
cana-6215	154	27	,	,	PUNCT
cana-6215	154	28	”	"	PUNCT
cana-6215	154	29	biomedical	biomedical	ADJ
cana-6215	154	30	signal	signal	NOUN
cana-6215	154	31	processing	processing	NOUN
cana-6215	154	32	and	and	CCONJ
cana-6215	154	33	control	control	NOUN
cana-6215	154	34	,	,	PUNCT
cana-6215	154	35	vol	vol	NOUN
cana-6215	154	36	.	.	PROPN
cana-6215	154	37	68	68	NUM
cana-6215	154	38	,	,	PUNCT
cana-6215	154	39	pp	pp	ADJ
cana-6215	154	40	.	.	PUNCT
cana-6215	155	1	102–118	102–118	NUM
cana-6215	155	2	,	,	PUNCT
cana-6215	155	3	2021	2021	NUM
cana-6215	155	4	.	.	PUNCT
cana-6215	156	1	[	[	X
cana-6215	156	2	2	2	NUM
cana-6215	156	3	]	]	PUNCT
cana-6215	156	4	f.	f.	PROPN
cana-6215	156	5	t.	t.	PROPN
cana-6215	156	6	al	al	PROPN
cana-6215	156	7	-	-	PUNCT
cana-6215	156	8	dhief	dhief	PROPN
cana-6215	156	9	,	,	PUNCT
cana-6215	156	10	m.	m.	NOUN
cana-6215	156	11	a.	a.	NOUN
cana-6215	156	12	mohd	mohd	PROPN
cana-6215	156	13	,	,	PUNCT
cana-6215	156	14	and	and	CCONJ
cana-6215	156	15	s.	s.	PROPN
cana-6215	156	16	a.	a.	PROPN
cana-6215	156	17	abbas	abbas	PROPN
cana-6215	156	18	,	,	PUNCT
cana-6215	156	19	“	"	PUNCT
cana-6215	156	20	transfer	transfer	VERB
cana-6215	156	21	learning	learning	NOUN
cana-6215	156	22	-	-	PUNCT
cana-6215	156	23	based	base	VERB
cana-6215	156	24	liver	liver	NOUN
cana-6215	156	25	cancer	cancer	NOUN
cana-6215	156	26	detection	detection	NOUN
cana-6215	156	27	using	use	VERB
cana-6215	156	28	deep	deep	ADJ
cana-6215	156	29	cnn	cnn	PROPN
cana-6215	156	30	models	model	NOUN
cana-6215	156	31	,	,	PUNCT
cana-6215	156	32	”	"	PUNCT
cana-6215	156	33	computers	computer	NOUN
cana-6215	156	34	in	in	ADP
cana-6215	156	35	biology	biology	NOUN
cana-6215	156	36	and	and	CCONJ
cana-6215	156	37	medicine	medicine	NOUN
cana-6215	156	38	,	,	PUNCT
cana-6215	156	39	vol	vol	NOUN
cana-6215	156	40	.	.	PROPN
cana-6215	156	41	145	145	NUM
cana-6215	156	42	,	,	PUNCT
cana-6215	156	43	p.	p.	NOUN
cana-6215	156	44	105423	105423	NUM
cana-6215	156	45	,	,	PUNCT
cana-6215	156	46	2022	2022	NUM
cana-6215	156	47	.	.	PUNCT
cana-6215	157	1	80.00	80.00	NUM
cana-6215	157	2	%	%	NOUN
cana-6215	157	3	85.00	85.00	NUM
cana-6215	157	4	%	%	NOUN
cana-6215	157	5	90.00	90.00	NUM
cana-6215	157	6	%	%	NOUN
cana-6215	157	7	95.00	95.00	NUM
cana-6215	157	8	%	%	NOUN
cana-6215	157	9	100.00	100.00	NUM
cana-6215	157	10	%	%	NOUN
cana-6215	157	11	105.00	105.00	NUM
cana-6215	157	12	%	%	NOUN
cana-6215	157	13	performance	performance	NOUN
cana-6215	157	14	evaluation	evaluation	NOUN
cana-6215	157	15	accuracy	accuracy	NOUN
cana-6215	157	16	recall	recall	NOUN
cana-6215	157	17	precision	precision	NOUN
cana-6215	157	18	f1	f1	NOUN
cana-6215	157	19	-	-	PUNCT
cana-6215	157	20	score	score	NOUN
cana-6215	157	21	communications	communication	NOUN
cana-6215	157	22	on	on	ADP
cana-6215	157	23	applied	apply	VERB
cana-6215	157	24	nonlinear	nonlinear	ADJ
cana-6215	157	25	analysis	analysis	NOUN
cana-6215	157	26	issn	issn	NOUN
cana-6215	157	27	:	:	PUNCT
cana-6215	157	28	1074	1074	NUM
cana-6215	157	29	-	-	PUNCT
cana-6215	157	30	133x	133x	NUM
cana-6215	157	31	vol	vol	NOUN
cana-6215	157	32	32	32	NUM
cana-6215	157	33	no	no	NOUN
cana-6215	157	34	.	.	PUNCT
cana-6215	158	1	4s	4s	NUM
cana-6215	158	2	(	(	PUNCT
cana-6215	158	3	2025	2025	NUM
cana-6215	158	4	)	)	PUNCT
cana-6215	158	5	783	783	NUM
cana-6215	158	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-6215	159	1	[	[	X
cana-6215	159	2	3	3	NUM
cana-6215	159	3	]	]	X
cana-6215	159	4	n.	n.	PROPN
cana-6215	159	5	kumar	kumar	PROPN
cana-6215	159	6	and	and	CCONJ
cana-6215	159	7	p.	p.	PROPN
cana-6215	159	8	singh	singh	PROPN
cana-6215	159	9	,	,	PUNCT
cana-6215	159	10	“	"	PUNCT
cana-6215	159	11	hybrid	hybrid	ADJ
cana-6215	159	12	cnn	cnn	PROPN
cana-6215	159	13	-	-	PUNCT
cana-6215	159	14	svm	svm	PROPN
cana-6215	159	15	model	model	NOUN
cana-6215	159	16	for	for	ADP
cana-6215	159	17	medical	medical	ADJ
cana-6215	159	18	image	image	NOUN
cana-6215	159	19	classification	classification	NOUN
cana-6215	159	20	,	,	PUNCT
cana-6215	159	21	”	"	PUNCT
cana-6215	159	22	journal	journal	NOUN
cana-6215	159	23	of	of	ADP
cana-6215	159	24	ambient	ambient	ADJ
cana-6215	159	25	intelligence	intelligence	NOUN
cana-6215	159	26	and	and	CCONJ
cana-6215	159	27	humanized	humanize	VERB
cana-6215	159	28	computing	computing	NOUN
cana-6215	159	29	,	,	PUNCT
cana-6215	159	30	vol	vol	NOUN
cana-6215	159	31	.	.	PROPN
cana-6215	160	1	11	11	NUM
cana-6215	160	2	,	,	PUNCT
cana-6215	160	3	no	no	INTJ
cana-6215	160	4	.	.	NOUN
cana-6215	160	5	8	8	NUM
cana-6215	160	6	,	,	PUNCT
cana-6215	160	7	pp	pp	ADJ
cana-6215	160	8	.	.	PUNCT
cana-6215	161	1	3327	3327	NUM
cana-6215	161	2	–	–	PUNCT
cana-6215	161	3	3336	3336	NUM
cana-6215	161	4	,	,	PUNCT
cana-6215	161	5	2020	2020	NUM
cana-6215	161	6	.	.	PUNCT
cana-6215	162	1	[	[	X
cana-6215	162	2	4	4	X
cana-6215	162	3	]	]	PUNCT
cana-6215	162	4	h.	h.	PROPN
cana-6215	162	5	ahmed	ahmed	PROPN
cana-6215	162	6	,	,	PUNCT
cana-6215	162	7	r.	r.	PROPN
cana-6215	162	8	al	al	PROPN
cana-6215	162	9	-	-	PUNCT
cana-6215	162	10	bayati	bayati	PROPN
cana-6215	162	11	,	,	PUNCT
cana-6215	162	12	and	and	CCONJ
cana-6215	162	13	l.	l.	PROPN
cana-6215	162	14	jasim	jasim	PROPN
cana-6215	162	15	,	,	PUNCT
cana-6215	162	16	“	"	PUNCT
cana-6215	162	17	lightweight	lightweight	ADJ
cana-6215	162	18	mobilenetv2	mobilenetv2	NOUN
cana-6215	162	19	approach	approach	NOUN
cana-6215	162	20	for	for	ADP
cana-6215	162	21	accurate	accurate	ADJ
cana-6215	162	22	liver	liver	NOUN
cana-6215	162	23	tumor	tumor	NOUN
cana-6215	162	24	detection	detection	NOUN
cana-6215	162	25	,	,	PUNCT
cana-6215	162	26	”	"	PUNCT
cana-6215	162	27	iraqi	iraqi	ADJ
cana-6215	162	28	journal	journal	NOUN
cana-6215	162	29	of	of	ADP
cana-6215	162	30	science	science	NOUN
cana-6215	162	31	,	,	PUNCT
cana-6215	162	32	vol	vol	NOUN
cana-6215	162	33	.	.	PROPN
cana-6215	162	34	64	64	NUM
cana-6215	162	35	,	,	PUNCT
cana-6215	162	36	no	no	INTJ
cana-6215	162	37	.	.	NOUN
cana-6215	162	38	4	4	NUM
cana-6215	162	39	,	,	PUNCT
cana-6215	162	40	pp	pp	ADJ
cana-6215	162	41	.	.	PUNCT
cana-6215	162	42	1021–1034	1021–1034	NUM
cana-6215	162	43	,	,	PUNCT
cana-6215	162	44	2023	2023	NUM
cana-6215	162	45	.	.	PUNCT
cana-6215	163	1	[	[	X
cana-6215	163	2	5	5	X
cana-6215	163	3	]	]	X
cana-6215	163	4	y.	y.	PROPN
cana-6215	163	5	wang	wang	PROPN
cana-6215	163	6	,	,	PUNCT
cana-6215	163	7	z.	z.	PROPN
cana-6215	163	8	zhou	zhou	PROPN
cana-6215	163	9	,	,	PUNCT
cana-6215	163	10	and	and	CCONJ
cana-6215	163	11	q.	q.	PROPN
cana-6215	163	12	zhang	zhang	PROPN
cana-6215	163	13	,	,	PUNCT
cana-6215	163	14	“	"	PUNCT
cana-6215	163	15	liver	liver	NOUN
cana-6215	163	16	lesion	lesion	NOUN
cana-6215	163	17	classification	classification	NOUN
cana-6215	163	18	using	use	VERB
cana-6215	163	19	deep	deep	ADJ
cana-6215	163	20	residual	residual	ADJ
cana-6215	163	21	networks	network	NOUN
cana-6215	163	22	,	,	PUNCT
cana-6215	163	23	”	"	PUNCT
cana-6215	163	24	ieee	ieee	NOUN
cana-6215	163	25	access	access	NOUN
cana-6215	163	26	,	,	PUNCT
cana-6215	163	27	vol	vol	NOUN
cana-6215	163	28	.	.	PROPN
cana-6215	163	29	7	7	NUM
cana-6215	163	30	,	,	PUNCT
cana-6215	163	31	pp	pp	ADJ
cana-6215	163	32	.	.	PUNCT
cana-6215	164	1	129321–129329	129321–129329	NUM
cana-6215	164	2	,	,	PUNCT
cana-6215	164	3	2019	2019	NUM
cana-6215	164	4	.	.	PUNCT
cana-6215	165	1	[	[	X
cana-6215	165	2	6	6	NUM
cana-6215	165	3	]	]	PUNCT
cana-6215	165	4	h.	h.	PROPN
cana-6215	165	5	chen	chen	PROPN
cana-6215	165	6	,	,	PUNCT
cana-6215	165	7	x.	x.	PROPN
cana-6215	165	8	li	li	PROPN
cana-6215	165	9	,	,	PUNCT
cana-6215	165	10	and	and	CCONJ
cana-6215	165	11	j.	j.	PROPN
cana-6215	165	12	xu	xu	PROPN
cana-6215	165	13	,	,	PUNCT
cana-6215	165	14	“	"	PUNCT
cana-6215	165	15	3d	3d	NUM
cana-6215	165	16	cnn	cnn	PROPN
cana-6215	165	17	-	-	PUNCT
cana-6215	165	18	based	base	VERB
cana-6215	165	19	liver	liver	NOUN
cana-6215	165	20	tumor	tumor	NOUN
cana-6215	165	21	classification	classification	NOUN
cana-6215	165	22	in	in	ADP
cana-6215	165	23	medical	medical	ADJ
cana-6215	165	24	imaging	imaging	NOUN
cana-6215	165	25	,	,	PUNCT
cana-6215	165	26	”	"	PUNCT
cana-6215	165	27	medical	medical	ADJ
cana-6215	165	28	image	image	NOUN
cana-6215	165	29	analysis	analysis	NOUN
cana-6215	165	30	,	,	PUNCT
cana-6215	165	31	vol	vol	NOUN
cana-6215	165	32	.	.	PROPN
cana-6215	165	33	74	74	NUM
cana-6215	165	34	,	,	PUNCT
cana-6215	165	35	p.	p.	NOUN
cana-6215	165	36	102241	102241	NUM
cana-6215	165	37	,	,	PUNCT
cana-6215	165	38	2021	2021	NUM
cana-6215	165	39	.	.	PUNCT
cana-6215	166	1	[	[	X
cana-6215	166	2	7	7	X
cana-6215	166	3	]	]	PUNCT
cana-6215	166	4	j.	j.	PROPN
cana-6215	166	5	li	li	PROPN
cana-6215	166	6	and	and	CCONJ
cana-6215	166	7	y.	y.	PROPN
cana-6215	166	8	zhao	zhao	PROPN
cana-6215	166	9	,	,	PUNCT
cana-6215	166	10	“	"	PUNCT
cana-6215	166	11	fine	fine	ADV
cana-6215	166	12	-	-	PUNCT
cana-6215	166	13	tuned	tune	VERB
cana-6215	166	14	vgg-16	vgg-16	NOUN
cana-6215	166	15	for	for	ADP
cana-6215	166	16	liver	liver	NOUN
cana-6215	166	17	cancer	cancer	NOUN
cana-6215	166	18	diagnosis	diagnosis	NOUN
cana-6215	166	19	using	use	VERB
cana-6215	166	20	ct	ct	NUM
cana-6215	166	21	scans	scan	NOUN
cana-6215	166	22	,	,	PUNCT
cana-6215	166	23	”	"	PUNCT
cana-6215	166	24	pattern	pattern	NOUN
cana-6215	166	25	recognition	recognition	NOUN
cana-6215	166	26	letters	letter	NOUN
cana-6215	166	27	,	,	PUNCT
cana-6215	166	28	vol	vol	NOUN
cana-6215	166	29	.	.	PROPN
cana-6215	166	30	135	135	NUM
cana-6215	166	31	,	,	PUNCT
cana-6215	166	32	pp	pp	ADJ
cana-6215	166	33	.	.	PUNCT
cana-6215	167	1	67–73	67–73	NUM
cana-6215	167	2	,	,	PUNCT
cana-6215	167	3	2020	2020	NUM
cana-6215	167	4	.	.	PUNCT
cana-6215	168	1	[	[	X
cana-6215	168	2	8	8	NUM
cana-6215	168	3	]	]	X
cana-6215	168	4	m.	m.	NOUN
cana-6215	168	5	patel	patel	PROPN
cana-6215	168	6	,	,	PUNCT
cana-6215	168	7	d.	d.	PROPN
cana-6215	168	8	shah	shah	PROPN
cana-6215	168	9	,	,	PUNCT
cana-6215	168	10	and	and	CCONJ
cana-6215	168	11	r.	r.	PROPN
cana-6215	168	12	jain	jain	PROPN
cana-6215	168	13	,	,	PUNCT
cana-6215	168	14	“	"	PUNCT
cana-6215	168	15	ensemble	ensemble	ADJ
cana-6215	168	16	deep	deep	ADJ
cana-6215	168	17	learning	learning	NOUN
cana-6215	168	18	model	model	NOUN
cana-6215	168	19	for	for	ADP
cana-6215	168	20	liver	liver	NOUN
cana-6215	168	21	disease	disease	NOUN
cana-6215	168	22	prediction	prediction	NOUN
cana-6215	168	23	,	,	PUNCT
cana-6215	168	24	”	"	PUNCT
cana-6215	168	25	expert	expert	NOUN
cana-6215	168	26	systems	system	NOUN
cana-6215	168	27	with	with	ADP
cana-6215	168	28	applications	application	NOUN
cana-6215	168	29	,	,	PUNCT
cana-6215	168	30	vol	vol	NOUN
cana-6215	168	31	.	.	PROPN
cana-6215	168	32	190	190	NUM
cana-6215	168	33	,	,	PUNCT
cana-6215	168	34	p.	p.	NOUN
cana-6215	168	35	116218	116218	NUM
cana-6215	168	36	,	,	PUNCT
cana-6215	168	37	2022	2022	NUM
cana-6215	168	38	.	.	PUNCT
cana-6215	169	1	[	[	X
cana-6215	169	2	9	9	NUM
cana-6215	169	3	]	]	X
cana-6215	169	4	h.	h.	PROPN
cana-6215	169	5	c.	c.	PROPN
cana-6215	169	6	shin	shin	PROPN
cana-6215	169	7	et	et	PROPN
cana-6215	169	8	al	al	PROPN
cana-6215	169	9	.	.	PROPN
cana-6215	169	10	,	,	PUNCT
cana-6215	169	11	“	"	PUNCT
cana-6215	169	12	deep	deep	ADJ
cana-6215	169	13	convolutional	convolutional	ADJ
cana-6215	169	14	neural	neural	ADJ
cana-6215	169	15	networks	network	NOUN
cana-6215	169	16	for	for	ADP
cana-6215	169	17	computer	computer	NOUN
cana-6215	169	18	-	-	PUNCT
cana-6215	169	19	aided	aid	VERB
cana-6215	169	20	detection	detection	NOUN
cana-6215	169	21	:	:	PUNCT
cana-6215	170	1	cnn	cnn	PROPN
cana-6215	170	2	architectures	architecture	NOUN
cana-6215	170	3	,	,	PUNCT
cana-6215	170	4	dataset	dataset	NOUN
cana-6215	170	5	characteristics	characteristic	NOUN
cana-6215	170	6	,	,	PUNCT
cana-6215	170	7	and	and	CCONJ
cana-6215	170	8	transfer	transfer	NOUN
cana-6215	170	9	learning	learning	NOUN
cana-6215	170	10	,	,	PUNCT
cana-6215	170	11	”	"	PUNCT
cana-6215	170	12	ieee	ieee	NOUN
cana-6215	170	13	trans	tran	NOUN
cana-6215	170	14	.	.	PROPN
cana-6215	171	1	med	med	PROPN
cana-6215	171	2	.	.	PUNCT
cana-6215	171	3	imaging	imaging	PROPN
cana-6215	171	4	,	,	PUNCT
cana-6215	171	5	vol	vol	NOUN
cana-6215	171	6	.	.	PROPN
cana-6215	171	7	35	35	NUM
cana-6215	171	8	,	,	PUNCT
cana-6215	171	9	no	no	INTJ
cana-6215	171	10	.	.	NOUN
cana-6215	171	11	5	5	NUM
cana-6215	171	12	,	,	PUNCT
cana-6215	171	13	pp	pp	ADJ
cana-6215	171	14	.	.	PUNCT
cana-6215	172	1	1285–1298	1285–1298	NUM
cana-6215	172	2	,	,	PUNCT
cana-6215	172	3	2016	2016	NUM
cana-6215	172	4	.	.	PUNCT
cana-6215	173	1	[	[	X
cana-6215	173	2	10	10	NUM
cana-6215	173	3	]	]	X
cana-6215	173	4	r.	r.	PROPN
cana-6215	173	5	r.	r.	PROPN
cana-6215	173	6	selvaraju	selvaraju	PROPN
cana-6215	173	7	et	et	PROPN
cana-6215	173	8	al	al	PROPN
cana-6215	173	9	.	.	PROPN
cana-6215	173	10	,	,	PUNCT
cana-6215	173	11	“	"	PUNCT
cana-6215	173	12	grad	grad	NOUN
cana-6215	173	13	-	-	PUNCT
cana-6215	173	14	cam	cam	NOUN
cana-6215	173	15	:	:	PUNCT
cana-6215	173	16	visual	visual	ADJ
cana-6215	173	17	explanations	explanation	NOUN
cana-6215	173	18	from	from	ADP
cana-6215	173	19	deep	deep	ADJ
cana-6215	173	20	networks	network	NOUN
cana-6215	173	21	via	via	ADP
cana-6215	173	22	gradient	gradient	NOUN
cana-6215	173	23	-	-	PUNCT
cana-6215	173	24	based	base	VERB
cana-6215	173	25	localization	localization	NOUN
cana-6215	173	26	,	,	PUNCT
cana-6215	173	27	”	"	PUNCT
cana-6215	173	28	in	in	ADP
cana-6215	173	29	proc	proc	NOUN
cana-6215	173	30	.	.	PUNCT
cana-6215	174	1	ieee	ieee	PROPN
cana-6215	174	2	int	int	PROPN
cana-6215	174	3	.	.	PUNCT
cana-6215	175	1	conf	conf	PROPN
cana-6215	175	2	.	.	PUNCT
cana-6215	176	1	comput	comput	NOUN
cana-6215	176	2	.	.	PUNCT
cana-6215	177	1	vis	vis	X
cana-6215	177	2	.	.	PUNCT
cana-6215	177	3	(	(	PUNCT
cana-6215	177	4	iccv	iccv	PROPN
cana-6215	177	5	)	)	PUNCT
cana-6215	177	6	,	,	PUNCT
cana-6215	177	7	2017	2017	NUM
cana-6215	177	8	,	,	PUNCT
cana-6215	177	9	pp	pp	ADJ
cana-6215	177	10	.	.	PUNCT
cana-6215	178	1	618–626	618–626	NUM
cana-6215	178	2	.	.	PUNCT
cana-6215	179	1	[	[	X
cana-6215	179	2	11	11	NUM
cana-6215	179	3	]	]	PUNCT
cana-6215	179	4	z.	z.	PROPN
cana-6215	179	5	obermeyer	obermeyer	PROPN
cana-6215	179	6	and	and	CCONJ
cana-6215	179	7	e.	e.	PROPN
cana-6215	179	8	j.	j.	PROPN
cana-6215	179	9	emanuel	emanuel	PROPN
cana-6215	179	10	,	,	PUNCT
cana-6215	179	11	“	"	PUNCT
cana-6215	179	12	predicting	predict	VERB
cana-6215	179	13	the	the	DET
cana-6215	179	14	future	future	NOUN
cana-6215	179	15	—	—	PUNCT
cana-6215	179	16	big	big	ADJ
cana-6215	179	17	data	datum	NOUN
cana-6215	179	18	,	,	PUNCT
cana-6215	179	19	machine	machine	NOUN
cana-6215	179	20	learning	learning	NOUN
cana-6215	179	21	,	,	PUNCT
cana-6215	179	22	and	and	CCONJ
cana-6215	179	23	clinical	clinical	ADJ
cana-6215	179	24	medicine	medicine	NOUN
cana-6215	179	25	,	,	PUNCT
cana-6215	179	26	”	"	PUNCT
cana-6215	179	27	n.	n.	PROPN
cana-6215	179	28	engl	engl	PROPN
cana-6215	179	29	.	.	PUNCT
cana-6215	180	1	j.	j.	PROPN
cana-6215	180	2	med	med	PROPN
cana-6215	180	3	.	.	PROPN
cana-6215	180	4	,	,	PUNCT
cana-6215	180	5	vol	vol	NOUN
cana-6215	180	6	.	.	PROPN
cana-6215	181	1	375	375	NUM
cana-6215	181	2	,	,	PUNCT
cana-6215	181	3	no	no	INTJ
cana-6215	181	4	.	.	NOUN
cana-6215	181	5	13	13	NUM
cana-6215	181	6	,	,	PUNCT
cana-6215	181	7	pp	pp	ADJ
cana-6215	181	8	.	.	PUNCT
cana-6215	182	1	1216–1219	1216–1219	NUM
cana-6215	182	2	,	,	PUNCT
cana-6215	182	3	2016	2016	NUM
cana-6215	182	4	.	.	PUNCT
cana-6215	183	1	[	[	X
cana-6215	183	2	12	12	NUM
cana-6215	183	3	]	]	X
cana-6215	183	4	g.	g.	NOUN
cana-6215	183	5	litjens	litjens	PROPN
cana-6215	183	6	,	,	PUNCT
cana-6215	183	7	t.	t.	PROPN
cana-6215	183	8	kooi	kooi	PROPN
cana-6215	183	9	,	,	PUNCT
cana-6215	183	10	b.	b.	PROPN
cana-6215	183	11	e.	e.	PROPN
cana-6215	183	12	bejnordi	bejnordi	PROPN
cana-6215	183	13	,	,	PUNCT
cana-6215	183	14	a.	a.	NOUN
cana-6215	183	15	a.	a.	NOUN
cana-6215	183	16	a.	a.	NOUN
cana-6215	183	17	setio	setio	PROPN
cana-6215	183	18	,	,	PUNCT
cana-6215	183	19	f.	f.	PROPN
cana-6215	183	20	ciompi	ciompi	PROPN
cana-6215	183	21	,	,	PUNCT
cana-6215	183	22	m.	m.	NOUN
cana-6215	183	23	ghafoorian	ghafoorian	PROPN
cana-6215	183	24	,	,	PUNCT
cana-6215	183	25	j.	j.	PROPN
cana-6215	183	26	a.	a.	PROPN
cana-6215	183	27	van	van	PROPN
cana-6215	183	28	der	der	PROPN
cana-6215	183	29	laak	laak	PROPN
cana-6215	183	30	,	,	PUNCT
cana-6215	183	31	b.	b.	PROPN
cana-6215	183	32	van	van	PROPN
cana-6215	183	33	ginneken	ginneken	PROPN
cana-6215	183	34	,	,	PUNCT
cana-6215	183	35	and	and	CCONJ
cana-6215	183	36	c.	c.	PROPN
cana-6215	183	37	i.	i.	PROPN
cana-6215	183	38	sánchez	sánchez	PROPN
cana-6215	183	39	,	,	PUNCT
cana-6215	183	40	“	"	PUNCT
cana-6215	183	41	a	a	DET
cana-6215	183	42	survey	survey	NOUN
cana-6215	183	43	on	on	ADP
cana-6215	183	44	deep	deep	ADJ
cana-6215	183	45	learning	learning	NOUN
cana-6215	183	46	in	in	ADP
cana-6215	183	47	medical	medical	ADJ
cana-6215	183	48	image	image	NOUN
cana-6215	183	49	analysis	analysis	NOUN
cana-6215	183	50	,	,	PUNCT
cana-6215	183	51	”	"	PUNCT
cana-6215	183	52	med	me	VERB
cana-6215	183	53	.	.	PUNCT
cana-6215	183	54	image	image	PROPN
cana-6215	183	55	anal	anal	PROPN
cana-6215	183	56	.	.	PUNCT
cana-6215	183	57	,	,	PUNCT
cana-6215	183	58	vol	vol	NOUN
cana-6215	183	59	.	.	PROPN
cana-6215	184	1	42	42	NUM
cana-6215	184	2	,	,	PUNCT
cana-6215	184	3	pp	pp	ADJ
cana-6215	184	4	.	.	PUNCT
cana-6215	185	1	60–88	60–88	NUM
cana-6215	185	2	,	,	PUNCT
cana-6215	185	3	2017	2017	NUM
cana-6215	185	4	.	.	PUNCT
cana-6215	186	1	[	[	X
cana-6215	186	2	13	13	NUM
cana-6215	186	3	]	]	PUNCT
cana-6215	186	4	j.	j.	PROPN
cana-6215	186	5	chen	chen	PROPN
cana-6215	186	6	,	,	PUNCT
cana-6215	186	7	l.	l.	PROPN
cana-6215	186	8	song	song	PROPN
cana-6215	186	9	,	,	PUNCT
cana-6215	186	10	l.	l.	PROPN
cana-6215	186	11	wei	wei	PROPN
cana-6215	186	12	,	,	PUNCT
cana-6215	186	13	and	and	CCONJ
cana-6215	186	14	q.	q.	PROPN
cana-6215	186	15	zou	zou	PROPN
cana-6215	186	16	,	,	PUNCT
cana-6215	186	17	“	"	PUNCT
cana-6215	186	18	drug	drug	NOUN
cana-6215	186	19	–	–	PUNCT
cana-6215	186	20	disease	disease	NOUN
cana-6215	186	21	association	association	NOUN
cana-6215	186	22	prediction	prediction	NOUN
cana-6215	186	23	with	with	ADP
cana-6215	186	24	graphbased	graphbased	ADJ
cana-6215	186	25	bipartite	bipartite	PROPN
cana-6215	186	26	learning	learning	NOUN
cana-6215	186	27	,	,	PUNCT
cana-6215	186	28	”	"	PUNCT
cana-6215	186	29	bioinformatics	bioinformatics	NOUN
cana-6215	186	30	,	,	PUNCT
cana-6215	186	31	vol	vol	NOUN
cana-6215	186	32	.	.	PROPN
cana-6215	186	33	34	34	NUM
cana-6215	186	34	,	,	PUNCT
cana-6215	186	35	no	no	INTJ
cana-6215	186	36	.	.	NOUN
cana-6215	186	37	16	16	NUM
cana-6215	186	38	,	,	PUNCT
cana-6215	186	39	pp	pp	ADJ
cana-6215	186	40	.	.	PUNCT
cana-6215	187	1	i733	i733	NUM
cana-6215	187	2	–	–	PUNCT
cana-6215	187	3	i741	i741	PROPN
cana-6215	187	4	,	,	PUNCT
cana-6215	187	5	2018	2018	NUM
cana-6215	187	6	.	.	PUNCT
cana-6215	188	1	[	[	X
cana-6215	188	2	14	14	NUM
cana-6215	188	3	]	]	PUNCT
cana-6215	188	4	a.	a.	NOUN
cana-6215	188	5	a.	a.	PROPN
cana-6215	188	6	h.	h.	PROPN
cana-6215	188	7	r.	r.	PROPN
cana-6215	188	8	khan	khan	PROPN
cana-6215	188	9	,	,	PUNCT
cana-6215	188	10	m.	m.	NOUN
cana-6215	188	11	a.	a.	PROPN
cana-6215	188	12	hossain	hossain	PROPN
cana-6215	188	13	,	,	PUNCT
cana-6215	188	14	m.	m.	PROPN
cana-6215	188	15	s.	s.	PROPN
cana-6215	188	16	rahman	rahman	PROPN
cana-6215	188	17	,	,	PUNCT
cana-6215	188	18	and	and	CCONJ
cana-6215	188	19	m.	m.	PROPN
cana-6215	188	20	a.	a.	PROPN
cana-6215	188	21	hasan	hasan	PROPN
cana-6215	188	22	,	,	PUNCT
cana-6215	188	23	“	"	PUNCT
cana-6215	188	24	an	an	DET
cana-6215	188	25	ultrasoundbased	ultrasoundbased	ADJ
cana-6215	188	26	computer	computer	NOUN
cana-6215	188	27	-	-	PUNCT
cana-6215	188	28	aided	aid	VERB
cana-6215	188	29	diagnosis	diagnosis	NOUN
cana-6215	188	30	tool	tool	NOUN
cana-6215	188	31	for	for	ADP
cana-6215	188	32	steatosis	steatosis	NOUN
cana-6215	188	33	,	,	PUNCT
cana-6215	188	34	”	"	PUNCT
cana-6215	188	35	in	in	ADP
cana-6215	188	36	proc	proc	NOUN
cana-6215	188	37	.	.	PUNCT
cana-6215	189	1	ieee	ieee	PROPN
cana-6215	189	2	int	int	PROPN
cana-6215	189	3	.	.	PUNCT
cana-6215	190	1	conf	conf	PROPN
cana-6215	190	2	.	.	PUNCT
cana-6215	191	1	comput	comput	PROPN
cana-6215	191	2	.	.	PUNCT
cana-6215	192	1	inf	inf	PROPN
cana-6215	192	2	.	.	PUNCT
cana-6215	192	3	technol	technol	PROPN
cana-6215	192	4	.	.	PROPN
cana-6215	192	5	,	,	PUNCT
cana-6215	192	6	2013	2013	NUM
cana-6215	192	7	,	,	PUNCT
cana-6215	192	8	pp	pp	ADP
cana-6215	192	9	.	.	PUNCT
cana-6215	193	1	204–209	204–209	NUM
cana-6215	193	2	,	,	PUNCT
cana-6215	193	3	doi	doi	NOUN
cana-6215	193	4	:	:	PUNCT
cana-6215	193	5	10.1109	10.1109	NUM
cana-6215	193	6	/	/	SYM
cana-6215	193	7	iccitechnol.2013.6623104	iccitechnol.2013.6623104	X
cana-6215	193	8	.	.	PUNCT
cana-6215	194	1	[	[	X
cana-6215	194	2	15	15	NUM
cana-6215	194	3	]	]	X
cana-6215	194	4	m.	m.	NOUN
cana-6215	194	5	j.	j.	PROPN
cana-6215	194	6	house	house	PROPN
cana-6215	194	7	,	,	PUNCT
cana-6215	194	8	s.	s.	PROPN
cana-6215	194	9	j.	j.	PROPN
cana-6215	194	10	bangma	bangma	PROPN
cana-6215	194	11	,	,	PUNCT
cana-6215	194	12	m.	m.	NOUN
cana-6215	194	13	thomas	thomas	PROPN
cana-6215	194	14	et	et	PROPN
cana-6215	194	15	al	al	PROPN
cana-6215	194	16	.	.	PROPN
cana-6215	194	17	,	,	PUNCT
cana-6215	194	18	“	"	PUNCT
cana-6215	194	19	texture	texture	NOUN
cana-6215	194	20	-	-	PUNCT
cana-6215	194	21	based	base	VERB
cana-6215	194	22	classification	classification	NOUN
cana-6215	194	23	of	of	ADP
cana-6215	194	24	liver	liver	NOUN
cana-6215	194	25	fibrosis	fibrosis	NOUN
cana-6215	194	26	using	use	VERB
cana-6215	194	27	mri	mri	NOUN
cana-6215	194	28	,	,	PUNCT
cana-6215	194	29	”	"	PUNCT
cana-6215	194	30	ieee	ieee	PROPN
cana-6215	194	31	j.	j.	PROPN
cana-6215	194	32	magn	magn	PROPN
cana-6215	194	33	.	.	PUNCT
cana-6215	195	1	reson	reson	PROPN
cana-6215	195	2	.	.	PUNCT
cana-6215	196	1	imaging	imaging	NOUN
cana-6215	196	2	,	,	PUNCT
cana-6215	196	3	vol	vol	NOUN
cana-6215	196	4	.	.	PROPN
cana-6215	196	5	41	41	NUM
cana-6215	196	6	,	,	PUNCT
cana-6215	196	7	no	no	INTJ
cana-6215	196	8	.	.	NOUN
cana-6215	196	9	2	2	NUM
cana-6215	196	10	,	,	PUNCT
cana-6215	196	11	pp	pp	ADJ
cana-6215	196	12	.	.	PUNCT
cana-6215	197	1	322–328	322–328	NUM
cana-6215	197	2	,	,	PUNCT
cana-6215	197	3	feb	feb	PROPN
cana-6215	197	4	.	.	PROPN
cana-6215	197	5	2015	2015	NUM
cana-6215	197	6	.	.	PUNCT
cana-6215	198	1	[	[	X
cana-6215	198	2	16	16	NUM
cana-6215	198	3	]	]	PUNCT
cana-6215	198	4	k.	k.	NOUN
cana-6215	199	1	he	he	PRON
cana-6215	199	2	,	,	PUNCT
cana-6215	199	3	x.	x.	PROPN
cana-6215	199	4	zhang	zhang	PROPN
cana-6215	199	5	,	,	PUNCT
cana-6215	199	6	s.	s.	PROPN
cana-6215	199	7	ren	ren	PROPN
cana-6215	199	8	,	,	PUNCT
cana-6215	199	9	and	and	CCONJ
cana-6215	199	10	j.	j.	PROPN
cana-6215	199	11	sun	sun	PROPN
cana-6215	199	12	,	,	PUNCT
cana-6215	199	13	“	"	PUNCT
cana-6215	199	14	deep	deep	ADJ
cana-6215	199	15	residual	residual	ADJ
cana-6215	199	16	learning	learning	NOUN
cana-6215	199	17	for	for	ADP
cana-6215	199	18	image	image	NOUN
cana-6215	199	19	recognition	recognition	NOUN
cana-6215	199	20	,	,	PUNCT
cana-6215	199	21	”	"	PUNCT
cana-6215	199	22	in	in	ADP
cana-6215	199	23	proc	proc	NOUN
cana-6215	199	24	.	.	PUNCT
cana-6215	200	1	ieee	ieee	NOUN
cana-6215	200	2	conf	conf	NOUN
cana-6215	200	3	.	.	PUNCT
cana-6215	201	1	comput	comput	NOUN
cana-6215	201	2	.	.	PUNCT
cana-6215	202	1	vis	vis	X
cana-6215	202	2	.	.	X
cana-6215	202	3	pattern	pattern	NOUN
cana-6215	202	4	recognit	recognit	VERB
cana-6215	202	5	.	.	PUNCT
cana-6215	203	1	(	(	PUNCT
cana-6215	203	2	cvpr	cvpr	NOUN
cana-6215	203	3	)	)	PUNCT
cana-6215	203	4	,	,	PUNCT
cana-6215	203	5	2016	2016	NUM
cana-6215	203	6	,	,	PUNCT
cana-6215	203	7	pp	pp	ADJ
cana-6215	203	8	.	.	PUNCT
cana-6215	204	1	770–778	770–778	NUM
cana-6215	204	2	.	.	PUNCT
cana-6215	205	1	[	[	X
cana-6215	205	2	17	17	NUM
cana-6215	205	3	]	]	X
cana-6215	205	4	h.	h.	PROPN
cana-6215	205	5	c.	c.	PROPN
cana-6215	205	6	shin	shin	PROPN
cana-6215	205	7	,	,	PUNCT
cana-6215	205	8	h.	h.	PROPN
cana-6215	205	9	r.	r.	PROPN
cana-6215	205	10	roth	roth	PROPN
cana-6215	205	11	,	,	PUNCT
cana-6215	205	12	m.	m.	NOUN
cana-6215	205	13	gao	gao	PROPN
cana-6215	205	14	,	,	PUNCT
cana-6215	205	15	l.	l.	PROPN
cana-6215	205	16	lu	lu	PROPN
cana-6215	205	17	,	,	PUNCT
cana-6215	205	18	z.	z.	PROPN
cana-6215	205	19	xu	xu	PROPN
cana-6215	205	20	,	,	PUNCT
cana-6215	205	21	i.	i.	PROPN
cana-6215	205	22	nogues	nogues	PROPN
cana-6215	205	23	,	,	PUNCT
cana-6215	205	24	j.	j.	PROPN
cana-6215	205	25	yao	yao	PROPN
cana-6215	205	26	,	,	PUNCT
cana-6215	205	27	d.	d.	PROPN
cana-6215	205	28	mollura	mollura	PROPN
cana-6215	205	29	,	,	PUNCT
cana-6215	205	30	and	and	CCONJ
cana-6215	205	31	r.	r.	PROPN
cana-6215	205	32	m.	m.	PROPN
cana-6215	205	33	summers	summer	NOUN
cana-6215	205	34	,	,	PUNCT
cana-6215	205	35	“	"	PUNCT
cana-6215	205	36	deep	deep	ADJ
cana-6215	205	37	convolutional	convolutional	ADJ
cana-6215	205	38	neural	neural	ADJ
cana-6215	205	39	networks	network	NOUN
cana-6215	205	40	for	for	ADP
cana-6215	205	41	computer	computer	NOUN
cana-6215	205	42	-	-	PUNCT
cana-6215	205	43	aided	aid	VERB
cana-6215	205	44	detection	detection	NOUN
cana-6215	205	45	:	:	PUNCT
cana-6215	205	46	cnn	cnn	PROPN
cana-6215	205	47	architectures	architecture	NOUN
cana-6215	205	48	,	,	PUNCT
cana-6215	205	49	dataset	dataset	NOUN
cana-6215	205	50	characteristics	characteristic	NOUN
cana-6215	205	51	,	,	PUNCT
cana-6215	205	52	and	and	CCONJ
cana-6215	205	53	transfer	transfer	NOUN
cana-6215	205	54	learning	learning	NOUN
cana-6215	205	55	,	,	PUNCT
cana-6215	205	56	”	"	PUNCT
cana-6215	205	57	ieee	ieee	NOUN
cana-6215	205	58	trans	tran	NOUN
cana-6215	205	59	.	.	PROPN
cana-6215	206	1	med	med	PROPN
cana-6215	206	2	.	.	PUNCT
cana-6215	206	3	imaging	imaging	PROPN
cana-6215	206	4	,	,	PUNCT
cana-6215	206	5	vol	vol	NOUN
cana-6215	206	6	.	.	PROPN
cana-6215	206	7	35	35	NUM
cana-6215	206	8	,	,	PUNCT
cana-6215	206	9	no	no	INTJ
cana-6215	206	10	.	.	NOUN
cana-6215	206	11	5	5	NUM
cana-6215	206	12	,	,	PUNCT
cana-6215	206	13	pp	pp	ADJ
cana-6215	206	14	.	.	PUNCT
cana-6215	207	1	1285–1298	1285–1298	NUM
cana-6215	207	2	,	,	PUNCT
cana-6215	207	3	2016	2016	NUM
cana-6215	207	4	.	.	PUNCT
cana-6215	208	1	[	[	X
cana-6215	208	2	18	18	NUM
cana-6215	208	3	]	]	X
cana-6215	208	4	r.	r.	PROPN
cana-6215	208	5	r.	r.	PROPN
cana-6215	208	6	selvaraju	selvaraju	PROPN
cana-6215	208	7	,	,	PUNCT
cana-6215	208	8	m.	m.	NOUN
cana-6215	208	9	cogswell	cogswell	PROPN
cana-6215	208	10	,	,	PUNCT
cana-6215	208	11	a.	a.	PROPN
cana-6215	208	12	das	das	PROPN
cana-6215	208	13	,	,	PUNCT
cana-6215	208	14	r.	r.	PROPN
cana-6215	208	15	vedantam	vedantam	PROPN
cana-6215	208	16	,	,	PUNCT
cana-6215	208	17	d.	d.	PROPN
cana-6215	208	18	parikh	parikh	PROPN
cana-6215	208	19	,	,	PUNCT
cana-6215	208	20	and	and	CCONJ
cana-6215	208	21	d.	d.	PROPN
cana-6215	208	22	batra	batra	PROPN
cana-6215	208	23	,	,	PUNCT
cana-6215	208	24	“	"	PUNCT
cana-6215	208	25	gradcam	gradcam	NOUN
cana-6215	208	26	:	:	PUNCT
cana-6215	208	27	visual	visual	ADJ
cana-6215	208	28	explanations	explanation	NOUN
cana-6215	208	29	from	from	ADP
cana-6215	208	30	deep	deep	ADJ
cana-6215	208	31	networks	network	NOUN
cana-6215	208	32	via	via	ADP
cana-6215	208	33	gradient	gradient	NOUN
cana-6215	208	34	-	-	PUNCT
cana-6215	208	35	based	base	VERB
cana-6215	208	36	localization	localization	NOUN
cana-6215	208	37	,	,	PUNCT
cana-6215	208	38	”	"	PUNCT
cana-6215	208	39	in	in	ADP
cana-6215	208	40	proc	proc	NOUN
cana-6215	208	41	.	.	PUNCT
cana-6215	209	1	ieee	ieee	PROPN
cana-6215	209	2	int	int	PROPN
cana-6215	209	3	.	.	PUNCT
cana-6215	210	1	conf	conf	PROPN
cana-6215	210	2	.	.	PUNCT
cana-6215	211	1	comput	comput	NOUN
cana-6215	211	2	.	.	PUNCT
cana-6215	212	1	vis	vis	X
cana-6215	212	2	.	.	PUNCT
cana-6215	212	3	(	(	PUNCT
cana-6215	212	4	iccv	iccv	PROPN
cana-6215	212	5	)	)	PUNCT
cana-6215	212	6	,	,	PUNCT
cana-6215	212	7	2017	2017	NUM
cana-6215	212	8	,	,	PUNCT
cana-6215	212	9	pp	pp	ADJ
cana-6215	212	10	.	.	PUNCT
cana-6215	213	1	618–626	618–626	NUM
cana-6215	213	2	.	.	PUNCT
cana-6215	214	1	[	[	X
cana-6215	214	2	19	19	NUM
cana-6215	214	3	]	]	X
cana-6215	214	4	c.	c.	PROPN
cana-6215	214	5	szegedy	szegedy	PROPN
cana-6215	214	6	,	,	PUNCT
cana-6215	214	7	w.	w.	PROPN
cana-6215	214	8	liu	liu	PROPN
cana-6215	214	9	,	,	PUNCT
cana-6215	214	10	y.	y.	PROPN
cana-6215	214	11	jia	jia	PROPN
cana-6215	214	12	,	,	PUNCT
cana-6215	214	13	p.	p.	PROPN
cana-6215	214	14	sermanet	sermanet	NOUN
cana-6215	214	15	,	,	PUNCT
cana-6215	214	16	s.	s.	PROPN
cana-6215	214	17	reed	reed	PROPN
cana-6215	214	18	,	,	PUNCT
cana-6215	214	19	d.	d.	PROPN
cana-6215	214	20	anguelov	anguelov	PROPN
cana-6215	214	21	,	,	PUNCT
cana-6215	214	22	d.	d.	PROPN
cana-6215	214	23	erhan	erhan	PROPN
cana-6215	214	24	,	,	PUNCT
cana-6215	214	25	v.	v.	PROPN
cana-6215	214	26	vanhoucke	vanhoucke	PROPN
cana-6215	214	27	,	,	PUNCT
cana-6215	214	28	and	and	CCONJ
cana-6215	214	29	a.	a.	NOUN
cana-6215	214	30	rabinovich	rabinovich	NOUN
cana-6215	214	31	,	,	PUNCT
cana-6215	214	32	“	"	PUNCT
cana-6215	214	33	going	go	VERB
cana-6215	214	34	deeper	deeply	ADV
cana-6215	214	35	with	with	ADP
cana-6215	214	36	convolutions	convolution	NOUN
cana-6215	214	37	,	,	PUNCT
cana-6215	214	38	”	"	PUNCT
cana-6215	214	39	in	in	ADP
cana-6215	214	40	proc	proc	NOUN
cana-6215	214	41	.	.	PUNCT
cana-6215	215	1	ieee	ieee	NOUN
cana-6215	215	2	conf	conf	NOUN
cana-6215	215	3	.	.	PUNCT
cana-6215	216	1	comput	comput	NOUN
cana-6215	216	2	.	.	PUNCT
cana-6215	217	1	vis	vis	X
cana-6215	217	2	.	.	X
cana-6215	217	3	pattern	pattern	NOUN
cana-6215	217	4	recognit	recognit	VERB
cana-6215	217	5	.	.	PUNCT
cana-6215	218	1	(	(	PUNCT
cana-6215	218	2	cvpr	cvpr	NOUN
cana-6215	218	3	)	)	PUNCT
cana-6215	218	4	,	,	PUNCT
cana-6215	218	5	2015	2015	NUM
cana-6215	218	6	,	,	PUNCT
cana-6215	218	7	pp	pp	ADJ
cana-6215	218	8	.	.	PUNCT
cana-6215	219	1	1–9	1–9	NOUN
cana-6215	219	2	.	.	PUNCT