id	sid	tid	token	lemma	pos
app01-9407	1	1	acta	acta	PROPN
app01-9407	1	2	polytechnica	polytechnica	PROPN
app01-9407	1	3	ctu	ctu	NOUN
app01-9407	1	4	proceedings	proceeding	NOUN
app01-9407	1	5	https://doi.org/10.14311/app.2023.42.0087	https://doi.org/10.14311/app.2023.42.0087	PROPN
app01-9407	1	6	acta	acta	PROPN
app01-9407	1	7	polytechnica	polytechnica	PROPN
app01-9407	1	8	ctu	ctu	NOUN
app01-9407	1	9	proceedings	proceeding	NOUN
app01-9407	1	10	42:87–93	42:87–93	PROPN
app01-9407	1	11	,	,	PUNCT
app01-9407	1	12	2023	2023	NUM
app01-9407	1	13	©	©	ADP
app01-9407	1	14	2023	2023	NUM
app01-9407	1	15	the	the	DET
app01-9407	1	16	author(s	author(s	NOUN
app01-9407	1	17	)	)	PUNCT
app01-9407	1	18	.	.	PUNCT
app01-9407	2	1	licensed	license	VERB
app01-9407	2	2	under	under	ADP
app01-9407	2	3	a	a	DET
app01-9407	2	4	cc	cc	NOUN
app01-9407	2	5	-	-	PUNCT
app01-9407	2	6	by	by	ADP
app01-9407	2	7	4.0	4.0	NUM
app01-9407	2	8	licence	licence	NOUN
app01-9407	2	9	published	publish	VERB
app01-9407	2	10	by	by	ADP
app01-9407	2	11	the	the	DET
app01-9407	2	12	czech	czech	PROPN
app01-9407	2	13	technical	technical	PROPN
app01-9407	2	14	university	university	PROPN
app01-9407	2	15	in	in	ADP
app01-9407	2	16	prague	prague	PROPN
app01-9407	2	17	segmentation	segmentation	NOUN
app01-9407	2	18	of	of	ADP
app01-9407	2	19	pores	pore	NOUN
app01-9407	2	20	in	in	ADP
app01-9407	2	21	carbon	carbon	NOUN
app01-9407	2	22	fiber	fiber	NOUN
app01-9407	2	23	reinforced	reinforce	VERB
app01-9407	2	24	polymers	polymer	NOUN
app01-9407	2	25	using	use	VERB
app01-9407	2	26	the	the	DET
app01-9407	2	27	u	u	ADJ
app01-9407	2	28	-	-	ADJ
app01-9407	2	29	net	net	ADJ
app01-9407	2	30	convolutional	convolutional	ADJ
app01-9407	2	31	neural	neural	ADJ
app01-9407	2	32	network	network	NOUN
app01-9407	2	33	miroslav	miroslav	PROPN
app01-9407	2	34	yosifova	yosifova	PROPN
app01-9407	2	35	,	,	PUNCT
app01-9407	2	36	b,∗	b,∗	PROPN
app01-9407	2	37	,	,	PUNCT
app01-9407	2	38	patrick	patrick	PROPN
app01-9407	2	39	weinbergerb	weinbergerb	PROPN
app01-9407	2	40	,	,	PUNCT
app01-9407	2	41	bernhard	bernhard	PROPN
app01-9407	2	42	plankb	plankb	PROPN
app01-9407	2	43	,	,	PUNCT
app01-9407	2	44	bernhard	bernhard	PROPN
app01-9407	2	45	fröhlerb	fröhlerb	ADJ
app01-9407	2	46	,	,	PUNCT
app01-9407	2	47	markus	markus	NOUN
app01-9407	2	48	hoeglingerb	hoeglingerb	NOUN
app01-9407	2	49	,	,	PUNCT
app01-9407	2	50	johann	johann	PROPN
app01-9407	2	51	kastnerb	kastnerb	PROPN
app01-9407	2	52	,	,	PUNCT
app01-9407	2	53	christoph	christoph	PROPN
app01-9407	2	54	heinzlc	heinzlc	VERB
app01-9407	2	55	a	a	DET
app01-9407	2	56	university	university	PROPN
app01-9407	2	57	of	of	ADP
app01-9407	2	58	antwerp	antwerp	PROPN
app01-9407	2	59	,	,	PUNCT
app01-9407	2	60	imec	imec	PROPN
app01-9407	2	61	-	-	PUNCT
app01-9407	2	62	visionlab	visionlab	NOUN
app01-9407	2	63	,	,	PUNCT
app01-9407	2	64	department	department	NOUN
app01-9407	2	65	of	of	ADP
app01-9407	2	66	physics	physics	PROPN
app01-9407	2	67	,	,	PUNCT
app01-9407	2	68	universiteitsplein	universiteitsplein	ADJ
app01-9407	2	69	1	1	NUM
app01-9407	2	70	,	,	PUNCT
app01-9407	2	71	2610	2610	NUM
app01-9407	2	72	antwerpen	antwerpen	PROPN
app01-9407	2	73	,	,	PUNCT
app01-9407	2	74	belgium	belgium	PROPN
app01-9407	2	75	b	b	PROPN
app01-9407	2	76	university	university	PROPN
app01-9407	2	77	of	of	ADP
app01-9407	2	78	applied	apply	VERB
app01-9407	2	79	sciences	sciences	PROPN
app01-9407	2	80	upper	upper	ADJ
app01-9407	2	81	austria	austria	PROPN
app01-9407	2	82	,	,	PUNCT
app01-9407	2	83	research	research	NOUN
app01-9407	2	84	group	group	NOUN
app01-9407	2	85	x	x	NOUN
app01-9407	2	86	-	-	NOUN
app01-9407	2	87	ray	ray	NOUN
app01-9407	2	88	computed	compute	VERB
app01-9407	2	89	tomography	tomography	NOUN
app01-9407	2	90	,	,	PUNCT
app01-9407	2	91	stelzhamerstraße	stelzhamerstraße	NOUN
app01-9407	2	92	23	23	NUM
app01-9407	2	93	,	,	PUNCT
app01-9407	2	94	4600	4600	NUM
app01-9407	2	95	wels	wel	NOUN
app01-9407	2	96	,	,	PUNCT
app01-9407	2	97	austria	austria	PROPN
app01-9407	2	98	c	c	PROPN
app01-9407	2	99	university	university	PROPN
app01-9407	2	100	of	of	ADP
app01-9407	2	101	passau	passau	PROPN
app01-9407	2	102	,	,	PUNCT
app01-9407	2	103	faculty	faculty	NOUN
app01-9407	2	104	of	of	ADP
app01-9407	2	105	computer	computer	NOUN
app01-9407	2	106	science	science	NOUN
app01-9407	2	107	and	and	CCONJ
app01-9407	2	108	mathematics	mathematic	NOUN
app01-9407	2	109	,	,	PUNCT
app01-9407	2	110	innstraße	innstraße	NOUN
app01-9407	2	111	43	43	NUM
app01-9407	2	112	,	,	PUNCT
app01-9407	2	113	94032	94032	NUM
app01-9407	2	114	passau	passau	PROPN
app01-9407	2	115	,	,	PUNCT
app01-9407	2	116	germany	germany	PROPN
app01-9407	2	117	∗	∗	NOUN
app01-9407	2	118	corresponding	correspond	VERB
app01-9407	2	119	author	author	NOUN
app01-9407	2	120	:	:	PUNCT
app01-9407	2	121	miroslav.yosifov@fh-wels.at	miroslav.yosifov@fh-wels.at	PROPN
app01-9407	3	1	abstract	abstract	ADJ
app01-9407	3	2	.	.	PUNCT
app01-9407	4	1	this	this	DET
app01-9407	4	2	study	study	NOUN
app01-9407	4	3	demonstrates	demonstrate	VERB
app01-9407	4	4	the	the	DET
app01-9407	4	5	utilization	utilization	NOUN
app01-9407	4	6	of	of	ADP
app01-9407	4	7	deep	deep	ADJ
app01-9407	4	8	learning	learning	NOUN
app01-9407	4	9	techniques	technique	NOUN
app01-9407	4	10	for	for	ADP
app01-9407	4	11	binary	binary	ADJ
app01-9407	4	12	semantic	semantic	ADJ
app01-9407	4	13	segmentation	segmentation	NOUN
app01-9407	4	14	of	of	ADP
app01-9407	4	15	pores	pore	NOUN
app01-9407	4	16	in	in	ADP
app01-9407	4	17	carbon	carbon	NOUN
app01-9407	4	18	fiber	fiber	NOUN
app01-9407	4	19	reinforced	reinforce	VERB
app01-9407	4	20	polymers	polymer	NOUN
app01-9407	4	21	(	(	PUNCT
app01-9407	4	22	cfrp	cfrp	NOUN
app01-9407	4	23	)	)	PUNCT
app01-9407	4	24	using	use	VERB
app01-9407	4	25	x	x	NOUN
app01-9407	4	26	-	-	NOUN
app01-9407	4	27	ray	ray	NOUN
app01-9407	4	28	computed	compute	VERB
app01-9407	4	29	tomography	tomography	NOUN
app01-9407	4	30	(	(	PUNCT
app01-9407	4	31	xct	xct	PROPN
app01-9407	4	32	)	)	PUNCT
app01-9407	4	33	datasets	dataset	NOUN
app01-9407	4	34	.	.	PUNCT
app01-9407	5	1	the	the	DET
app01-9407	5	2	proposed	propose	VERB
app01-9407	5	3	workflow	workflow	NOUN
app01-9407	5	4	is	be	AUX
app01-9407	5	5	designed	design	VERB
app01-9407	5	6	to	to	PART
app01-9407	5	7	generate	generate	VERB
app01-9407	5	8	efficient	efficient	ADJ
app01-9407	5	9	segmentation	segmentation	NOUN
app01-9407	5	10	models	model	NOUN
app01-9407	5	11	with	with	ADP
app01-9407	5	12	reasonable	reasonable	ADJ
app01-9407	5	13	execution	execution	NOUN
app01-9407	5	14	time	time	NOUN
app01-9407	5	15	,	,	PUNCT
app01-9407	5	16	applicable	applicable	ADJ
app01-9407	5	17	even	even	ADV
app01-9407	5	18	for	for	ADP
app01-9407	5	19	users	user	NOUN
app01-9407	5	20	using	use	VERB
app01-9407	5	21	consumer	consumer	NOUN
app01-9407	5	22	-	-	PUNCT
app01-9407	5	23	grade	grade	NOUN
app01-9407	5	24	gpu	gpu	NOUN
app01-9407	5	25	systems	system	NOUN
app01-9407	5	26	.	.	PUNCT
app01-9407	6	1	first	first	ADV
app01-9407	6	2	,	,	PUNCT
app01-9407	6	3	u	u	NOUN
app01-9407	6	4	-	-	NOUN
app01-9407	6	5	net	net	ADJ
app01-9407	6	6	,	,	PUNCT
app01-9407	6	7	a	a	DET
app01-9407	6	8	convolutional	convolutional	ADJ
app01-9407	6	9	neural	neural	ADJ
app01-9407	6	10	network	network	NOUN
app01-9407	6	11	,	,	PUNCT
app01-9407	6	12	is	be	AUX
app01-9407	6	13	modified	modify	VERB
app01-9407	6	14	to	to	PART
app01-9407	6	15	handle	handle	VERB
app01-9407	6	16	the	the	DET
app01-9407	6	17	segmentation	segmentation	NOUN
app01-9407	6	18	of	of	ADP
app01-9407	6	19	xct	xct	PROPN
app01-9407	6	20	datasets	dataset	NOUN
app01-9407	6	21	.	.	PUNCT
app01-9407	7	1	in	in	ADP
app01-9407	7	2	the	the	DET
app01-9407	7	3	second	second	ADJ
app01-9407	7	4	step	step	NOUN
app01-9407	7	5	,	,	PUNCT
app01-9407	7	6	suitable	suitable	ADJ
app01-9407	7	7	hyperparameters	hyperparameter	NOUN
app01-9407	7	8	are	be	AUX
app01-9407	7	9	determined	determine	VERB
app01-9407	7	10	through	through	ADP
app01-9407	7	11	a	a	DET
app01-9407	7	12	parameter	parameter	NOUN
app01-9407	7	13	analysis	analysis	NOUN
app01-9407	7	14	(	(	PUNCT
app01-9407	7	15	hyperparameter	hyperparameter	NOUN
app01-9407	7	16	tuning	tuning	NOUN
app01-9407	7	17	)	)	PUNCT
app01-9407	7	18	,	,	PUNCT
app01-9407	7	19	and	and	CCONJ
app01-9407	7	20	the	the	DET
app01-9407	7	21	parameter	parameter	NOUN
app01-9407	7	22	set	set	VERB
app01-9407	7	23	with	with	ADP
app01-9407	7	24	the	the	DET
app01-9407	7	25	best	good	ADJ
app01-9407	7	26	result	result	NOUN
app01-9407	7	27	was	be	AUX
app01-9407	7	28	used	use	VERB
app01-9407	7	29	for	for	ADP
app01-9407	7	30	the	the	DET
app01-9407	7	31	final	final	ADJ
app01-9407	7	32	training	training	NOUN
app01-9407	7	33	.	.	PUNCT
app01-9407	8	1	in	in	ADP
app01-9407	8	2	the	the	DET
app01-9407	8	3	final	final	ADJ
app01-9407	8	4	step	step	NOUN
app01-9407	8	5	,	,	PUNCT
app01-9407	8	6	we	we	PRON
app01-9407	8	7	report	report	VERB
app01-9407	8	8	on	on	ADP
app01-9407	8	9	our	our	PRON
app01-9407	8	10	efforts	effort	NOUN
app01-9407	8	11	of	of	ADP
app01-9407	8	12	implementing	implement	VERB
app01-9407	8	13	the	the	DET
app01-9407	8	14	testing	testing	NOUN
app01-9407	8	15	stage	stage	NOUN
app01-9407	8	16	in	in	ADP
app01-9407	8	17	open_ia	open_ia	PROPN
app01-9407	8	18	,	,	PUNCT
app01-9407	8	19	which	which	PRON
app01-9407	8	20	allows	allow	VERB
app01-9407	8	21	users	user	NOUN
app01-9407	8	22	to	to	PART
app01-9407	8	23	segment	segment	VERB
app01-9407	8	24	datasets	dataset	NOUN
app01-9407	8	25	with	with	ADP
app01-9407	8	26	the	the	DET
app01-9407	8	27	fully	fully	ADV
app01-9407	8	28	trained	train	VERB
app01-9407	8	29	model	model	NOUN
app01-9407	8	30	within	within	ADP
app01-9407	8	31	reasonable	reasonable	ADJ
app01-9407	8	32	time	time	NOUN
app01-9407	8	33	.	.	PUNCT
app01-9407	9	1	the	the	DET
app01-9407	9	2	model	model	NOUN
app01-9407	9	3	performs	perform	VERB
app01-9407	9	4	well	well	ADV
app01-9407	9	5	on	on	ADP
app01-9407	9	6	datasets	dataset	NOUN
app01-9407	9	7	with	with	ADP
app01-9407	9	8	both	both	CCONJ
app01-9407	9	9	high	high	ADJ
app01-9407	9	10	and	and	CCONJ
app01-9407	9	11	low	low	ADJ
app01-9407	9	12	resolution	resolution	NOUN
app01-9407	9	13	,	,	PUNCT
app01-9407	9	14	and	and	CCONJ
app01-9407	9	15	even	even	ADV
app01-9407	9	16	works	work	VERB
app01-9407	9	17	reasonably	reasonably	ADV
app01-9407	9	18	for	for	ADP
app01-9407	9	19	barely	barely	ADV
app01-9407	9	20	visible	visible	ADJ
app01-9407	9	21	pores	pore	NOUN
app01-9407	9	22	with	with	ADP
app01-9407	9	23	different	different	ADJ
app01-9407	9	24	shapes	shape	NOUN
app01-9407	9	25	and	and	CCONJ
app01-9407	9	26	size	size	NOUN
app01-9407	9	27	.	.	PUNCT
app01-9407	10	1	in	in	ADP
app01-9407	10	2	our	our	PRON
app01-9407	10	3	experiments	experiment	NOUN
app01-9407	10	4	,	,	PUNCT
app01-9407	10	5	we	we	PRON
app01-9407	10	6	could	could	AUX
app01-9407	10	7	show	show	VERB
app01-9407	10	8	that	that	SCONJ
app01-9407	10	9	u	u	NOUN
app01-9407	10	10	-	-	NOUN
app01-9407	10	11	net	net	ADJ
app01-9407	10	12	is	be	AUX
app01-9407	10	13	suitable	suitable	ADJ
app01-9407	10	14	for	for	ADP
app01-9407	10	15	pore	pore	ADJ
app01-9407	10	16	segmentation	segmentation	NOUN
app01-9407	10	17	.	.	PUNCT
app01-9407	11	1	despite	despite	SCONJ
app01-9407	11	2	being	be	AUX
app01-9407	11	3	trained	train	VERB
app01-9407	11	4	on	on	ADP
app01-9407	11	5	a	a	DET
app01-9407	11	6	limited	limited	ADJ
app01-9407	11	7	number	number	NOUN
app01-9407	11	8	of	of	ADP
app01-9407	11	9	datasets	dataset	NOUN
app01-9407	11	10	,	,	PUNCT
app01-9407	11	11	it	it	PRON
app01-9407	11	12	exhibits	exhibit	VERB
app01-9407	11	13	a	a	DET
app01-9407	11	14	satisfactory	satisfactory	ADJ
app01-9407	11	15	level	level	NOUN
app01-9407	11	16	of	of	ADP
app01-9407	11	17	prediction	prediction	NOUN
app01-9407	11	18	accuracy	accuracy	NOUN
app01-9407	11	19	.	.	PUNCT
app01-9407	12	1	keywords	keyword	NOUN
app01-9407	12	2	:	:	PUNCT
app01-9407	12	3	deep	deep	ADJ
app01-9407	12	4	learning	learning	NOUN
app01-9407	12	5	,	,	PUNCT
app01-9407	12	6	segmentation	segmentation	NOUN
app01-9407	12	7	,	,	PUNCT
app01-9407	12	8	u	u	NOUN
app01-9407	12	9	-	-	NOUN
app01-9407	12	10	net	net	ADJ
app01-9407	12	11	,	,	PUNCT
app01-9407	12	12	computed	compute	VERB
app01-9407	12	13	tomography	tomography	NOUN
app01-9407	12	14	,	,	PUNCT
app01-9407	12	15	pores	pore	NOUN
app01-9407	12	16	,	,	PUNCT
app01-9407	12	17	carbon	carbon	NOUN
app01-9407	12	18	fiber	fiber	NOUN
app01-9407	12	19	reinforced	reinforce	VERB
app01-9407	12	20	polymers	polymer	NOUN
app01-9407	12	21	.	.	PUNCT
app01-9407	13	1	1	1	X
app01-9407	13	2	.	.	X
app01-9407	13	3	introduction	introduction	NOUN
app01-9407	13	4	fiber	fiber	NOUN
app01-9407	13	5	reinforced	reinforce	VERB
app01-9407	13	6	polymers	polymer	NOUN
app01-9407	13	7	are	be	AUX
app01-9407	13	8	getting	get	VERB
app01-9407	13	9	increasingly	increasingly	ADV
app01-9407	13	10	important	important	ADJ
app01-9407	13	11	in	in	ADP
app01-9407	13	12	our	our	PRON
app01-9407	13	13	daily	daily	ADJ
app01-9407	13	14	lives	life	NOUN
app01-9407	13	15	.	.	PUNCT
app01-9407	14	1	as	as	SCONJ
app01-9407	14	2	they	they	PRON
app01-9407	14	3	are	be	AUX
app01-9407	14	4	used	use	VERB
app01-9407	14	5	in	in	ADP
app01-9407	14	6	safety	safety	NOUN
app01-9407	14	7	-	-	PUNCT
app01-9407	14	8	critical	critical	ADJ
app01-9407	14	9	areas	area	NOUN
app01-9407	14	10	such	such	ADJ
app01-9407	14	11	as	as	ADP
app01-9407	14	12	aerospace	aerospace	NOUN
app01-9407	14	13	,	,	PUNCT
app01-9407	14	14	fast	fast	ADJ
app01-9407	14	15	and	and	CCONJ
app01-9407	14	16	reliable	reliable	ADJ
app01-9407	14	17	analysis	analysis	NOUN
app01-9407	14	18	of	of	ADP
app01-9407	14	19	these	these	DET
app01-9407	14	20	materials	material	NOUN
app01-9407	14	21	is	be	AUX
app01-9407	14	22	necessary	necessary	ADJ
app01-9407	14	23	.	.	PUNCT
app01-9407	15	1	industrial	industrial	ADJ
app01-9407	15	2	x	x	NOUN
app01-9407	15	3	-	-	NOUN
app01-9407	15	4	ray	ray	NOUN
app01-9407	15	5	computed	compute	VERB
app01-9407	15	6	tomography	tomography	NOUN
app01-9407	15	7	is	be	AUX
app01-9407	15	8	a	a	DET
app01-9407	15	9	powerful	powerful	ADJ
app01-9407	15	10	tool	tool	NOUN
app01-9407	15	11	to	to	PART
app01-9407	15	12	facilitate	facilitate	VERB
app01-9407	15	13	such	such	DET
app01-9407	15	14	an	an	DET
app01-9407	15	15	analysis	analysis	NOUN
app01-9407	15	16	,	,	PUNCT
app01-9407	15	17	but	but	CCONJ
app01-9407	15	18	it	it	PRON
app01-9407	15	19	is	be	AUX
app01-9407	15	20	still	still	ADV
app01-9407	15	21	a	a	DET
app01-9407	15	22	huge	huge	ADJ
app01-9407	15	23	effort	effort	NOUN
app01-9407	15	24	to	to	PART
app01-9407	15	25	analyze	analyze	VERB
app01-9407	15	26	,	,	PUNCT
app01-9407	15	27	visualize	visualize	VERB
app01-9407	15	28	and	and	CCONJ
app01-9407	15	29	quantify	quantify	VERB
app01-9407	15	30	the	the	DET
app01-9407	15	31	data	datum	NOUN
app01-9407	15	32	generated	generate	VERB
app01-9407	15	33	in	in	ADP
app01-9407	15	34	an	an	DET
app01-9407	15	35	xct	xct	PROPN
app01-9407	15	36	scan	scan	PROPN
app01-9407	16	1	[	[	X
app01-9407	16	2	1	1	NUM
app01-9407	16	3	]	]	PUNCT
app01-9407	16	4	.	.	PUNCT
app01-9407	17	1	deep	deep	ADJ
app01-9407	17	2	learning	learning	NOUN
app01-9407	17	3	has	have	AUX
app01-9407	17	4	been	be	AUX
app01-9407	17	5	successfully	successfully	ADV
app01-9407	17	6	employed	employ	VERB
app01-9407	17	7	in	in	ADP
app01-9407	17	8	fields	field	NOUN
app01-9407	17	9	like	like	ADP
app01-9407	17	10	computer	computer	NOUN
app01-9407	17	11	vision	vision	NOUN
app01-9407	18	1	[	[	X
app01-9407	18	2	2	2	NUM
app01-9407	18	3	]	]	PUNCT
app01-9407	18	4	,	,	PUNCT
app01-9407	18	5	object	object	NOUN
app01-9407	18	6	recognition	recognition	NOUN
app01-9407	19	1	[	[	X
app01-9407	19	2	3	3	NUM
app01-9407	19	3	]	]	PUNCT
app01-9407	19	4	,	,	PUNCT
app01-9407	19	5	medical	medical	ADJ
app01-9407	19	6	image	image	NOUN
app01-9407	19	7	analysis	analysis	NOUN
app01-9407	19	8	[	[	X
app01-9407	19	9	4	4	NUM
app01-9407	19	10	]	]	PUNCT
app01-9407	19	11	,	,	PUNCT
app01-9407	19	12	face	face	NOUN
app01-9407	19	13	-	-	PUNCT
app01-9407	19	14	recognition	recognition	NOUN
app01-9407	19	15	applications	application	NOUN
app01-9407	19	16	,	,	PUNCT
app01-9407	19	17	and	and	CCONJ
app01-9407	19	18	material	material	NOUN
app01-9407	19	19	inspection	inspection	NOUN
app01-9407	19	20	[	[	X
app01-9407	19	21	5–7	5–7	X
app01-9407	19	22	]	]	PUNCT
app01-9407	19	23	over	over	ADP
app01-9407	19	24	the	the	DET
app01-9407	19	25	past	past	ADJ
app01-9407	19	26	few	few	ADJ
app01-9407	19	27	decades	decade	NOUN
app01-9407	19	28	.	.	PUNCT
app01-9407	20	1	convolutional	convolutional	ADJ
app01-9407	20	2	neural	neural	ADJ
app01-9407	20	3	networks	network	NOUN
app01-9407	20	4	(	(	PUNCT
app01-9407	20	5	cnn	cnn	PROPN
app01-9407	20	6	)	)	PUNCT
app01-9407	20	7	are	be	AUX
app01-9407	20	8	one	one	NUM
app01-9407	20	9	of	of	ADP
app01-9407	20	10	the	the	DET
app01-9407	20	11	highly	highly	ADV
app01-9407	20	12	popular	popular	ADJ
app01-9407	20	13	classes	class	NOUN
app01-9407	20	14	of	of	ADP
app01-9407	20	15	deep	deep	ADJ
app01-9407	20	16	learning	learning	NOUN
app01-9407	20	17	methods	method	NOUN
app01-9407	20	18	,	,	PUNCT
app01-9407	20	19	which	which	PRON
app01-9407	20	20	has	have	AUX
app01-9407	20	21	showed	show	VERB
app01-9407	20	22	a	a	DET
app01-9407	20	23	remarkable	remarkable	ADJ
app01-9407	20	24	ability	ability	NOUN
app01-9407	20	25	to	to	PART
app01-9407	20	26	handle	handle	VERB
app01-9407	20	27	classification	classification	NOUN
app01-9407	20	28	and	and	CCONJ
app01-9407	20	29	segmentation	segmentation	NOUN
app01-9407	20	30	tasks	task	NOUN
app01-9407	20	31	for	for	ADP
app01-9407	20	32	image	image	NOUN
app01-9407	20	33	,	,	PUNCT
app01-9407	20	34	volume	volume	NOUN
app01-9407	20	35	,	,	PUNCT
app01-9407	20	36	and	and	CCONJ
app01-9407	20	37	video	video	NOUN
app01-9407	20	38	.	.	PUNCT
app01-9407	21	1	material	material	NOUN
app01-9407	21	2	science	science	NOUN
app01-9407	21	3	domain	domain	NOUN
app01-9407	21	4	experts	expert	NOUN
app01-9407	21	5	often	often	ADV
app01-9407	21	6	need	need	VERB
app01-9407	21	7	to	to	PART
app01-9407	21	8	segment	segment	VERB
app01-9407	21	9	pores	pore	NOUN
app01-9407	21	10	in	in	ADP
app01-9407	21	11	cfrp	cfrp	NOUN
app01-9407	21	12	to	to	PART
app01-9407	21	13	characterize	characterize	VERB
app01-9407	21	14	materials	material	NOUN
app01-9407	21	15	,	,	PUNCT
app01-9407	21	16	especially	especially	ADV
app01-9407	21	17	with	with	ADP
app01-9407	21	18	regards	regard	NOUN
app01-9407	21	19	to	to	ADP
app01-9407	21	20	pores	pore	NOUN
app01-9407	21	21	.	.	PUNCT
app01-9407	22	1	however	however	ADV
app01-9407	22	2	,	,	PUNCT
app01-9407	22	3	segmentation	segmentation	NOUN
app01-9407	22	4	of	of	ADP
app01-9407	22	5	images	image	NOUN
app01-9407	22	6	or	or	CCONJ
app01-9407	22	7	volumes	volume	NOUN
app01-9407	22	8	remains	remain	VERB
app01-9407	22	9	a	a	DET
app01-9407	22	10	challenging	challenging	ADJ
app01-9407	22	11	problem	problem	NOUN
app01-9407	22	12	.	.	PUNCT
app01-9407	23	1	despite	despite	SCONJ
app01-9407	23	2	the	the	DET
app01-9407	23	3	availability	availability	NOUN
app01-9407	23	4	of	of	ADP
app01-9407	23	5	various	various	ADJ
app01-9407	23	6	segmentation	segmentation	NOUN
app01-9407	23	7	and	and	CCONJ
app01-9407	23	8	classification	classification	NOUN
app01-9407	23	9	techniques	technique	NOUN
app01-9407	23	10	such	such	ADJ
app01-9407	23	11	as	as	ADP
app01-9407	23	12	k	k	NOUN
app01-9407	23	13	-	-	PUNCT
app01-9407	23	14	means	mean	VERB
app01-9407	23	15	[	[	X
app01-9407	23	16	8	8	NUM
app01-9407	23	17	,	,	PUNCT
app01-9407	23	18	9	9	NUM
app01-9407	23	19	]	]	PUNCT
app01-9407	23	20	,	,	PUNCT
app01-9407	23	21	watershed	watershe	VERB
app01-9407	23	22	[	[	X
app01-9407	23	23	10	10	NUM
app01-9407	23	24	]	]	PUNCT
app01-9407	23	25	and	and	CCONJ
app01-9407	23	26	thresholding	thresholde	VERB
app01-9407	23	27	techniques	technique	NOUN
app01-9407	23	28	,	,	PUNCT
app01-9407	23	29	there	there	PRON
app01-9407	23	30	are	be	VERB
app01-9407	23	31	still	still	ADV
app01-9407	23	32	several	several	ADJ
app01-9407	23	33	research	research	NOUN
app01-9407	23	34	efforts	effort	NOUN
app01-9407	23	35	to	to	PART
app01-9407	23	36	improve	improve	VERB
app01-9407	23	37	segmentation	segmentation	NOUN
app01-9407	23	38	methods	method	NOUN
app01-9407	23	39	[	[	X
app01-9407	23	40	11	11	NUM
app01-9407	23	41	,	,	PUNCT
app01-9407	23	42	12	12	NUM
app01-9407	23	43	]	]	PUNCT
app01-9407	23	44	.	.	PUNCT
app01-9407	24	1	accurate	accurate	ADJ
app01-9407	24	2	and	and	CCONJ
app01-9407	24	3	efficient	efficient	ADJ
app01-9407	24	4	segmentation	segmentation	NOUN
app01-9407	24	5	plays	play	VERB
app01-9407	24	6	a	a	DET
app01-9407	24	7	important	important	ADJ
app01-9407	24	8	role	role	NOUN
app01-9407	24	9	in	in	ADP
app01-9407	24	10	ensuring	ensure	VERB
app01-9407	24	11	the	the	DET
app01-9407	24	12	integrity	integrity	NOUN
app01-9407	24	13	of	of	ADP
app01-9407	24	14	safety	safety	NOUN
app01-9407	24	15	-	-	PUNCT
app01-9407	24	16	critical	critical	ADJ
app01-9407	24	17	materials	material	NOUN
app01-9407	24	18	.	.	PUNCT
app01-9407	25	1	to	to	PART
app01-9407	25	2	achieve	achieve	VERB
app01-9407	25	3	a	a	DET
app01-9407	25	4	robust	robust	ADJ
app01-9407	25	5	and	and	CCONJ
app01-9407	25	6	reliable	reliable	ADJ
app01-9407	25	7	segmentation	segmentation	NOUN
app01-9407	25	8	,	,	PUNCT
app01-9407	25	9	we	we	PRON
app01-9407	25	10	have	have	AUX
app01-9407	25	11	developed	develop	VERB
app01-9407	25	12	a	a	DET
app01-9407	25	13	modified	modify	VERB
app01-9407	25	14	version	version	NOUN
app01-9407	25	15	of	of	ADP
app01-9407	25	16	the	the	DET
app01-9407	25	17	u	u	ADJ
app01-9407	25	18	-	-	ADJ
app01-9407	25	19	net	net	ADJ
app01-9407	25	20	architecture	architecture	NOUN
app01-9407	25	21	.	.	PUNCT
app01-9407	26	1	our	our	PRON
app01-9407	26	2	contributions	contribution	NOUN
app01-9407	26	3	in	in	ADP
app01-9407	26	4	this	this	DET
app01-9407	26	5	work	work	NOUN
app01-9407	26	6	are	be	AUX
app01-9407	26	7	:	:	PUNCT
app01-9407	26	8	•	•	ADP
app01-9407	26	9	the	the	DET
app01-9407	26	10	modified	modify	VERB
app01-9407	26	11	3d	3d	NUM
app01-9407	26	12	u	u	NOUN
app01-9407	26	13	-	-	NOUN
app01-9407	26	14	net	net	NOUN
app01-9407	26	15	designed	design	VERB
app01-9407	26	16	for	for	ADP
app01-9407	26	17	efficient	efficient	ADJ
app01-9407	26	18	pore	pore	ADJ
app01-9407	26	19	segmentation	segmentation	NOUN
app01-9407	26	20	of	of	ADP
app01-9407	26	21	xct	xct	PROPN
app01-9407	26	22	scans	scan	NOUN
app01-9407	26	23	from	from	ADP
app01-9407	26	24	cfrp	cfrp	PROPN
app01-9407	26	25	.	.	PROPN
app01-9407	27	1	•	•	NUM
app01-9407	27	2	a	a	DET
app01-9407	27	3	network	network	NOUN
app01-9407	27	4	architecture	architecture	NOUN
app01-9407	27	5	and	and	CCONJ
app01-9407	27	6	testing	testing	NOUN
app01-9407	27	7	application	application	NOUN
app01-9407	27	8	in	in	ADP
app01-9407	27	9	open_ia	open_ia	NOUN
app01-9407	27	10	[	[	X
app01-9407	27	11	13	13	NUM
app01-9407	27	12	]	]	PUNCT
app01-9407	27	13	which	which	PRON
app01-9407	27	14	allows	allow	VERB
app01-9407	27	15	predictions	prediction	NOUN
app01-9407	27	16	on	on	ADP
app01-9407	27	17	consumergrade	consumergrade	ADJ
app01-9407	27	18	gpus	gpu	NOUN
app01-9407	27	19	.	.	PUNCT
app01-9407	28	1	•	•	NUM
app01-9407	28	2	evaluation	evaluation	NOUN
app01-9407	28	3	of	of	ADP
app01-9407	28	4	the	the	DET
app01-9407	28	5	neural	neural	ADJ
app01-9407	28	6	network	network	NOUN
app01-9407	28	7	’s	’s	PART
app01-9407	28	8	performance	performance	NOUN
app01-9407	28	9	on	on	ADP
app01-9407	28	10	unseen	unseen	ADJ
app01-9407	28	11	data	datum	NOUN
app01-9407	28	12	with	with	ADP
app01-9407	28	13	varying	vary	VERB
app01-9407	28	14	resolutions	resolution	NOUN
app01-9407	28	15	and	and	CCONJ
app01-9407	28	16	sizes	size	NOUN
app01-9407	28	17	.	.	PUNCT
app01-9407	29	1	•	•	NUM
app01-9407	29	2	visual	visual	ADJ
app01-9407	29	3	analysis	analysis	NOUN
app01-9407	29	4	and	and	CCONJ
app01-9407	29	5	quantitative	quantitative	ADJ
app01-9407	29	6	comparison	comparison	NOUN
app01-9407	29	7	of	of	ADP
app01-9407	29	8	the	the	DET
app01-9407	29	9	network	network	NOUN
app01-9407	29	10	’s	’s	PART
app01-9407	29	11	predictions	prediction	NOUN
app01-9407	29	12	with	with	ADP
app01-9407	29	13	the	the	DET
app01-9407	29	14	results	result	NOUN
app01-9407	29	15	(	(	PUNCT
app01-9407	29	16	labels	label	NOUN
app01-9407	29	17	)	)	PUNCT
app01-9407	29	18	generated	generate	VERB
app01-9407	29	19	from	from	ADP
app01-9407	29	20	the	the	DET
app01-9407	29	21	otsu	otsu	NOUN
app01-9407	29	22	thresholding	thresholding	NOUN
app01-9407	29	23	technique	technique	NOUN
app01-9407	29	24	.	.	PUNCT
app01-9407	30	1	2	2	X
app01-9407	30	2	.	.	X
app01-9407	30	3	background	background	NOUN
app01-9407	30	4	and	and	CCONJ
app01-9407	30	5	related	relate	VERB
app01-9407	30	6	work	work	NOUN
app01-9407	30	7	segmentation	segmentation	NOUN
app01-9407	30	8	is	be	AUX
app01-9407	30	9	defined	define	VERB
app01-9407	30	10	as	as	ADP
app01-9407	30	11	classification	classification	NOUN
app01-9407	30	12	of	of	ADP
app01-9407	30	13	an	an	DET
app01-9407	30	14	image	image	NOUN
app01-9407	30	15	into	into	ADP
app01-9407	30	16	regions	region	NOUN
app01-9407	30	17	which	which	PRON
app01-9407	30	18	contain	contain	VERB
app01-9407	30	19	similar	similar	ADJ
app01-9407	30	20	characteristics	characteristic	NOUN
app01-9407	30	21	(	(	PUNCT
app01-9407	30	22	e.g.	e.g.	ADV
app01-9407	30	23	,	,	PUNCT
app01-9407	30	24	similar	similar	ADJ
app01-9407	30	25	grey	grey	ADJ
app01-9407	30	26	values	value	NOUN
app01-9407	30	27	)	)	PUNCT
app01-9407	30	28	.	.	PUNCT
app01-9407	31	1	several	several	ADJ
app01-9407	31	2	segmentation	segmentation	NOUN
app01-9407	31	3	methods	method	NOUN
app01-9407	31	4	can	can	AUX
app01-9407	31	5	be	be	AUX
app01-9407	31	6	applied	apply	VERB
app01-9407	31	7	for	for	ADP
app01-9407	31	8	segmenting	segment	VERB
app01-9407	31	9	pores	pore	NOUN
app01-9407	31	10	in	in	ADP
app01-9407	31	11	xct	xct	PROPN
app01-9407	31	12	scans	scan	NOUN
app01-9407	31	13	of	of	ADP
app01-9407	31	14	fiber	fiber	NOUN
app01-9407	31	15	-	-	PUNCT
app01-9407	31	16	reinforced	reinforce	VERB
app01-9407	31	17	polymers	polymer	NOUN
app01-9407	31	18	.	.	PUNCT
app01-9407	32	1	here	here	ADV
app01-9407	32	2	a	a	DET
app01-9407	32	3	short	short	ADJ
app01-9407	32	4	overview	overview	NOUN
app01-9407	32	5	is	be	AUX
app01-9407	32	6	given	give	VERB
app01-9407	32	7	over	over	ADP
app01-9407	32	8	segmentation	segmentation	NOUN
app01-9407	32	9	methods	method	NOUN
app01-9407	32	10	employed	employ	VERB
app01-9407	32	11	in	in	ADP
app01-9407	32	12	this	this	DET
app01-9407	32	13	87	87	NUM
app01-9407	32	14	https://doi.org/10.14311/app.2023.42.0087	https://doi.org/10.14311/app.2023.42.0087	PROPN
app01-9407	32	15	https://creativecommons.org/licenses/by/4.0/	https://creativecommons.org/licenses/by/4.0/	PROPN
app01-9407	32	16	https://www.cvut.cz/en	https://www.cvut.cz/en	PROPN
app01-9407	32	17	m.	m.	NOUN
app01-9407	32	18	yosifov	yosifov	NOUN
app01-9407	32	19	,	,	PUNCT
app01-9407	32	20	p.	p.	NOUN
app01-9407	32	21	weinberger	weinberger	PROPN
app01-9407	32	22	,	,	PUNCT
app01-9407	32	23	b.	b.	PROPN
app01-9407	32	24	plank	plank	PROPN
app01-9407	32	25	et	et	PROPN
app01-9407	32	26	al	al	PROPN
app01-9407	32	27	.	.	PROPN
app01-9407	32	28	acta	acta	PROPN
app01-9407	32	29	polytechnica	polytechnica	PROPN
app01-9407	32	30	ctu	ctu	PROPN
app01-9407	32	31	proceedings	proceeding	NOUN
app01-9407	32	32	figure	figure	VERB
app01-9407	32	33	1	1	NUM
app01-9407	32	34	.	.	PUNCT
app01-9407	33	1	(	(	PUNCT
app01-9407	33	2	a	a	X
app01-9407	33	3	)	)	PUNCT
app01-9407	33	4	shows	show	VERB
app01-9407	33	5	3d	3d	NUM
app01-9407	33	6	rendering	rendering	NOUN
app01-9407	33	7	of	of	ADP
app01-9407	33	8	sample	sample	NOUN
app01-9407	33	9	1	1	NUM
app01-9407	33	10	,	,	PUNCT
app01-9407	33	11	sample	sample	NOUN
app01-9407	33	12	2	2	NUM
app01-9407	33	13	,	,	PUNCT
app01-9407	33	14	and	and	CCONJ
app01-9407	33	15	sample	sample	NOUN
app01-9407	33	16	3	3	NUM
app01-9407	33	17	.	.	PUNCT
app01-9407	34	1	(	(	PUNCT
app01-9407	34	2	b	b	NOUN
app01-9407	34	3	)	)	PUNCT
app01-9407	34	4	,	,	PUNCT
app01-9407	34	5	(	(	PUNCT
app01-9407	34	6	c	c	X
app01-9407	34	7	)	)	PUNCT
app01-9407	34	8	and	and	CCONJ
app01-9407	34	9	(	(	PUNCT
app01-9407	34	10	d	d	X
app01-9407	34	11	)	)	PUNCT
app01-9407	34	12	respectively	respectively	ADV
app01-9407	34	13	display	display	VERB
app01-9407	34	14	axis	axis	NOUN
app01-9407	34	15	-	-	PUNCT
app01-9407	34	16	aligned	align	VERB
app01-9407	34	17	2d	2d	NOUN
app01-9407	34	18	slices	slice	NOUN
app01-9407	34	19	of	of	ADP
app01-9407	34	20	sample	sample	NOUN
app01-9407	34	21	1	1	NUM
app01-9407	34	22	,	,	PUNCT
app01-9407	34	23	2	2	NUM
app01-9407	34	24	and	and	CCONJ
app01-9407	34	25	3	3	NUM
app01-9407	34	26	.	.	NOUN
app01-9407	34	27	area	area	NOUN
app01-9407	34	28	,	,	PUNCT
app01-9407	34	29	the	the	DET
app01-9407	34	30	methods	method	NOUN
app01-9407	34	31	that	that	PRON
app01-9407	34	32	we	we	PRON
app01-9407	34	33	used	use	VERB
app01-9407	34	34	in	in	ADP
app01-9407	34	35	our	our	PRON
app01-9407	34	36	work	work	NOUN
app01-9407	34	37	and	and	CCONJ
app01-9407	34	38	those	those	PRON
app01-9407	34	39	which	which	PRON
app01-9407	34	40	we	we	PRON
app01-9407	34	41	based	base	VERB
app01-9407	34	42	our	our	PRON
app01-9407	34	43	work	work	NOUN
app01-9407	34	44	on	on	ADP
app01-9407	34	45	.	.	PUNCT
app01-9407	35	1	2.1	2.1	NUM
app01-9407	35	2	.	.	PUNCT
app01-9407	35	3	thresholding	thresholding	NOUN
app01-9407	35	4	-	-	PUNCT
app01-9407	35	5	based	base	VERB
app01-9407	35	6	segmentation	segmentation	NOUN
app01-9407	35	7	global	global	ADJ
app01-9407	35	8	thresholding	thresholding	NOUN
app01-9407	35	9	techniques	technique	NOUN
app01-9407	35	10	are	be	AUX
app01-9407	35	11	used	use	VERB
app01-9407	35	12	in	in	ADP
app01-9407	35	13	a	a	DET
app01-9407	35	14	wide	wide	ADJ
app01-9407	35	15	range	range	NOUN
app01-9407	35	16	of	of	ADP
app01-9407	35	17	diverse	diverse	ADJ
app01-9407	35	18	applications	application	NOUN
app01-9407	35	19	.	.	PUNCT
app01-9407	36	1	a	a	DET
app01-9407	36	2	detailed	detailed	ADJ
app01-9407	36	3	summary	summary	NOUN
app01-9407	36	4	on	on	ADP
app01-9407	36	5	thresholding	thresholde	VERB
app01-9407	36	6	techniques	technique	NOUN
app01-9407	36	7	has	have	AUX
app01-9407	36	8	been	be	AUX
app01-9407	36	9	published	publish	VERB
app01-9407	36	10	by	by	ADP
app01-9407	36	11	sezgin	sezgin	NOUN
app01-9407	36	12	and	and	CCONJ
app01-9407	36	13	sankur	sankur	VERB
app01-9407	36	14	[	[	X
app01-9407	36	15	14	14	NUM
app01-9407	36	16	]	]	PUNCT
app01-9407	36	17	.	.	PUNCT
app01-9407	37	1	various	various	ADJ
app01-9407	37	2	kinds	kind	NOUN
app01-9407	37	3	of	of	ADP
app01-9407	37	4	threshold	threshold	NOUN
app01-9407	37	5	methods	method	NOUN
app01-9407	37	6	have	have	AUX
app01-9407	37	7	been	be	AUX
app01-9407	37	8	described	describe	VERB
app01-9407	37	9	and	and	CCONJ
app01-9407	37	10	compared	compare	VERB
app01-9407	37	11	on	on	ADP
app01-9407	37	12	carbon	carbon	NOUN
app01-9407	37	13	fiber	fiber	NOUN
app01-9407	37	14	polymer	polymer	NOUN
app01-9407	37	15	samples	sample	NOUN
app01-9407	37	16	by	by	ADP
app01-9407	37	17	rao	rao	PROPN
app01-9407	37	18	et	et	PROPN
app01-9407	37	19	.	.	PUNCT
app01-9407	38	1	al	al	PROPN
app01-9407	38	2	.	.	PUNCT
app01-9407	39	1	[	[	X
app01-9407	39	2	12	12	NUM
app01-9407	39	3	]	]	PUNCT
app01-9407	39	4	.	.	PUNCT
app01-9407	40	1	in	in	ADP
app01-9407	40	2	this	this	DET
app01-9407	40	3	paper	paper	NOUN
app01-9407	40	4	we	we	PRON
app01-9407	40	5	have	have	AUX
app01-9407	40	6	decided	decide	VERB
app01-9407	40	7	to	to	PART
app01-9407	40	8	use	use	VERB
app01-9407	40	9	otsu	otsu	NOUN
app01-9407	40	10	thresholding	thresholde	VERB
app01-9407	40	11	[	[	X
app01-9407	40	12	15	15	NUM
app01-9407	40	13	]	]	PUNCT
app01-9407	40	14	to	to	PART
app01-9407	40	15	create	create	VERB
app01-9407	40	16	label	label	NOUN
app01-9407	40	17	images	image	NOUN
app01-9407	40	18	for	for	ADP
app01-9407	40	19	the	the	DET
app01-9407	40	20	training	training	NOUN
app01-9407	40	21	,	,	PUNCT
app01-9407	40	22	testing	testing	NOUN
app01-9407	40	23	and	and	CCONJ
app01-9407	40	24	validation	validation	NOUN
app01-9407	40	25	as	as	SCONJ
app01-9407	40	26	shown	show	VERB
app01-9407	40	27	by	by	ADP
app01-9407	40	28	reh	reh	PROPN
app01-9407	40	29	et	et	PROPN
app01-9407	40	30	al	al	PROPN
app01-9407	40	31	.	.	PUNCT
app01-9407	41	1	segmenting	segment	VERB
app01-9407	41	2	pores	pore	NOUN
app01-9407	41	3	[	[	X
app01-9407	41	4	16	16	NUM
app01-9407	41	5	]	]	PUNCT
app01-9407	41	6	.	.	PUNCT
app01-9407	42	1	in	in	ADP
app01-9407	42	2	addition	addition	NOUN
app01-9407	42	3	,	,	PUNCT
app01-9407	42	4	it	it	PRON
app01-9407	42	5	is	be	AUX
app01-9407	42	6	a	a	DET
app01-9407	42	7	non	non	ADJ
app01-9407	42	8	-	-	ADJ
app01-9407	42	9	parametric	parametric	ADJ
app01-9407	42	10	threshold	threshold	NOUN
app01-9407	42	11	method	method	NOUN
app01-9407	42	12	.	.	PUNCT
app01-9407	43	1	therefore	therefore	ADV
app01-9407	43	2	,	,	PUNCT
app01-9407	43	3	no	no	DET
app01-9407	43	4	additional	additional	ADJ
app01-9407	43	5	user	user	NOUN
app01-9407	43	6	-	-	PUNCT
app01-9407	43	7	depended	depend	VERB
app01-9407	43	8	parameters	parameter	NOUN
app01-9407	43	9	have	have	VERB
app01-9407	43	10	to	to	PART
app01-9407	43	11	be	be	AUX
app01-9407	43	12	defined	define	VERB
app01-9407	43	13	.	.	PUNCT
app01-9407	44	1	2.2	2.2	NUM
app01-9407	44	2	.	.	PUNCT
app01-9407	45	1	convolutional	convolutional	ADJ
app01-9407	45	2	neural	neural	ADJ
app01-9407	45	3	networks	network	NOUN
app01-9407	45	4	convolutional	convolutional	ADJ
app01-9407	45	5	neural	neural	ADJ
app01-9407	45	6	networks	network	NOUN
app01-9407	45	7	now	now	ADV
app01-9407	45	8	are	be	AUX
app01-9407	45	9	the	the	DET
app01-9407	45	10	state	state	NOUN
app01-9407	45	11	-	-	PUNCT
app01-9407	45	12	ofthe	ofthe	NOUN
app01-9407	45	13	-	-	PUNCT
app01-9407	45	14	art	art	NOUN
app01-9407	45	15	technique	technique	NOUN
app01-9407	45	16	in	in	ADP
app01-9407	45	17	biomedical	biomedical	ADJ
app01-9407	45	18	image	image	NOUN
app01-9407	45	19	segmentation	segmentation	NOUN
app01-9407	45	20	[	[	X
app01-9407	45	21	17	17	NUM
app01-9407	45	22	]	]	PUNCT
app01-9407	45	23	.	.	PUNCT
app01-9407	46	1	one	one	NUM
app01-9407	46	2	of	of	ADP
app01-9407	46	3	these	these	DET
app01-9407	46	4	cnn	cnn	PROPN
app01-9407	46	5	architectures	architecture	NOUN
app01-9407	46	6	is	be	AUX
app01-9407	46	7	u	u	NOUN
app01-9407	46	8	-	-	NOUN
app01-9407	46	9	net	net	ADJ
app01-9407	46	10	.	.	PUNCT
app01-9407	47	1	u	u	NOUN
app01-9407	47	2	-	-	NOUN
app01-9407	47	3	net	net	ADJ
app01-9407	47	4	was	be	AUX
app01-9407	47	5	initially	initially	ADV
app01-9407	47	6	developed	develop	VERB
app01-9407	47	7	to	to	PART
app01-9407	47	8	address	address	VERB
app01-9407	47	9	the	the	DET
app01-9407	47	10	segmentation	segmentation	NOUN
app01-9407	47	11	task	task	NOUN
app01-9407	47	12	of	of	ADP
app01-9407	47	13	2d	2d	PROPN
app01-9407	47	14	biomedical	biomedical	ADJ
app01-9407	47	15	electron	electron	NOUN
app01-9407	47	16	microscopic	microscopic	NOUN
app01-9407	47	17	images	image	NOUN
app01-9407	47	18	and	and	CCONJ
app01-9407	47	19	demonstrated	demonstrate	VERB
app01-9407	47	20	remarkable	remarkable	ADJ
app01-9407	47	21	success	success	NOUN
app01-9407	47	22	during	during	ADP
app01-9407	47	23	the	the	DET
app01-9407	47	24	isbi	isbi	NOUN
app01-9407	47	25	cell	cell	NOUN
app01-9407	47	26	tracking	tracking	NOUN
app01-9407	47	27	challenge	challenge	NOUN
app01-9407	47	28	in	in	ADP
app01-9407	47	29	2015	2015	NUM
app01-9407	47	30	[	[	X
app01-9407	47	31	17	17	NUM
app01-9407	47	32	]	]	PUNCT
app01-9407	47	33	.	.	PUNCT
app01-9407	48	1	it	it	PRON
app01-9407	48	2	has	have	AUX
app01-9407	48	3	been	be	AUX
app01-9407	48	4	successfully	successfully	ADV
app01-9407	48	5	applied	apply	VERB
app01-9407	48	6	to	to	ADP
app01-9407	48	7	various	various	ADJ
app01-9407	48	8	segmentation	segmentation	NOUN
app01-9407	48	9	tasks	task	NOUN
app01-9407	48	10	,	,	PUNCT
app01-9407	48	11	including	include	VERB
app01-9407	48	12	cell	cell	NOUN
app01-9407	48	13	,	,	PUNCT
app01-9407	48	14	organ	organ	NOUN
app01-9407	48	15	segmentation	segmentation	NOUN
app01-9407	48	16	[	[	X
app01-9407	48	17	18	18	NUM
app01-9407	48	18	]	]	PUNCT
app01-9407	48	19	and	and	CCONJ
app01-9407	48	20	brain	brain	VERB
app01-9407	48	21	tumour	tumour	NOUN
app01-9407	48	22	detection	detection	NOUN
app01-9407	48	23	[	[	X
app01-9407	48	24	19	19	NUM
app01-9407	48	25	]	]	PUNCT
app01-9407	48	26	.	.	PUNCT
app01-9407	49	1	later	later	ADV
app01-9407	49	2	,	,	PUNCT
app01-9407	49	3	u	u	NOUN
app01-9407	49	4	-	-	NOUN
app01-9407	49	5	net	net	ADJ
app01-9407	49	6	was	be	AUX
app01-9407	49	7	further	far	ADV
app01-9407	49	8	developed	develop	VERB
app01-9407	49	9	to	to	PART
app01-9407	49	10	handle	handle	VERB
app01-9407	49	11	3d	3d	NUM
app01-9407	49	12	input	input	NOUN
app01-9407	49	13	shapes	shape	NOUN
app01-9407	49	14	for	for	ADP
app01-9407	49	15	segmenting	segment	VERB
app01-9407	49	16	3d	3d	NUM
app01-9407	49	17	microscopy	microscopy	NOUN
app01-9407	49	18	images	image	NOUN
app01-9407	49	19	[	[	X
app01-9407	49	20	18	18	NUM
app01-9407	49	21	]	]	PUNCT
app01-9407	49	22	.	.	PUNCT
app01-9407	50	1	3	3	X
app01-9407	50	2	.	.	X
app01-9407	50	3	data	datum	NOUN
app01-9407	50	4	characteristic	characteristic	ADJ
app01-9407	50	5	in	in	ADP
app01-9407	50	6	this	this	DET
app01-9407	50	7	paper	paper	NOUN
app01-9407	50	8	,	,	PUNCT
app01-9407	50	9	we	we	PRON
app01-9407	50	10	have	have	AUX
app01-9407	50	11	used	use	VERB
app01-9407	50	12	two	two	NUM
app01-9407	50	13	different	different	ADJ
app01-9407	50	14	samples	sample	NOUN
app01-9407	50	15	of	of	ADP
app01-9407	50	16	the	the	DET
app01-9407	50	17	same	same	ADJ
app01-9407	50	18	material	material	NOUN
app01-9407	50	19	which	which	PRON
app01-9407	50	20	have	have	AUX
app01-9407	50	21	been	be	AUX
app01-9407	50	22	obtained	obtain	VERB
app01-9407	50	23	from	from	ADP
app01-9407	50	24	fiber	fiber	NOUN
app01-9407	50	25	reinforced	reinforce	VERB
app01-9407	50	26	composites	composite	NOUN
app01-9407	50	27	.	.	PUNCT
app01-9407	51	1	10	10	NUM
app01-9407	51	2	×	×	NOUN
app01-9407	51	3	10	10	NUM
app01-9407	51	4	×	×	NOUN
app01-9407	51	5	2	2	NUM
app01-9407	51	6	mm3	mm3	NOUN
app01-9407	51	7	carbon	carbon	NOUN
app01-9407	51	8	fiber	fiber	NOUN
app01-9407	51	9	reinforced	reinforce	VERB
app01-9407	51	10	polymer	polymer	NOUN
app01-9407	51	11	samples	sample	NOUN
app01-9407	51	12	were	be	AUX
app01-9407	51	13	cut	cut	VERB
app01-9407	51	14	out	out	ADP
app01-9407	51	15	from	from	ADP
app01-9407	51	16	an	an	DET
app01-9407	51	17	360×510×2	360×510×2	NUM
app01-9407	51	18	mm3	mm3	NOUN
app01-9407	51	19	plate	plate	NOUN
app01-9407	51	20	.	.	PUNCT
app01-9407	52	1	the	the	DET
app01-9407	52	2	plates	plate	NOUN
app01-9407	52	3	for	for	ADP
app01-9407	52	4	sample	sample	NOUN
app01-9407	52	5	1	1	NUM
app01-9407	52	6	were	be	AUX
app01-9407	52	7	manufactured	manufacture	VERB
app01-9407	52	8	by	by	ADP
app01-9407	52	9	a	a	DET
app01-9407	52	10	wet	wet	NOUN
app01-9407	52	11	lay	lie	VERB
app01-9407	52	12	up	up	ADP
app01-9407	52	13	process	process	NOUN
app01-9407	52	14	with	with	ADP
app01-9407	52	15	a	a	DET
app01-9407	52	16	vacuum	vacuum	NOUN
app01-9407	52	17	bag	bag	NOUN
app01-9407	52	18	and	and	CCONJ
app01-9407	52	19	for	for	ADP
app01-9407	52	20	sample	sample	NOUN
app01-9407	52	21	2	2	NUM
app01-9407	52	22	a	a	DET
app01-9407	52	23	vacuum	vacuum	NOUN
app01-9407	52	24	assisted	assist	VERB
app01-9407	52	25	resin	resin	NOUN
app01-9407	52	26	infusion	infusion	NOUN
app01-9407	52	27	(	(	PUNCT
app01-9407	52	28	vari	vari	ADJ
app01-9407	52	29	)	)	PUNCT
app01-9407	52	30	process	process	NOUN
app01-9407	52	31	was	be	AUX
app01-9407	52	32	used	use	VERB
app01-9407	52	33	.	.	PUNCT
app01-9407	53	1	both	both	DET
app01-9407	53	2	sample	sample	NOUN
app01-9407	53	3	plates	plate	NOUN
app01-9407	53	4	were	be	AUX
app01-9407	53	5	manufactured	manufacture	VERB
app01-9407	53	6	of	of	ADP
app01-9407	53	7	six	six	NUM
app01-9407	53	8	layers	layer	NOUN
app01-9407	53	9	of	of	ADP
app01-9407	53	10	an	an	DET
app01-9407	53	11	2×2	2×2	NUM
app01-9407	53	12	twill	twill	NOUN
app01-9407	53	13	weave	weave	NOUN
app01-9407	53	14	pattern	pattern	NOUN
app01-9407	53	15	.	.	PUNCT
app01-9407	54	1	figure	figure	NOUN
app01-9407	54	2	1	1	NUM
app01-9407	54	3	shows	show	VERB
app01-9407	54	4	the	the	DET
app01-9407	54	5	training	training	NOUN
app01-9407	54	6	data	datum	NOUN
app01-9407	54	7	(	(	PUNCT
app01-9407	54	8	sample	sample	NOUN
app01-9407	54	9	1	1	NUM
app01-9407	54	10	)	)	PUNCT
app01-9407	54	11	,	,	PUNCT
app01-9407	54	12	the	the	DET
app01-9407	54	13	testing	testing	NOUN
app01-9407	54	14	data	datum	NOUN
app01-9407	54	15	(	(	PUNCT
app01-9407	54	16	sample	sample	NOUN
app01-9407	54	17	2	2	NUM
app01-9407	54	18	)	)	PUNCT
app01-9407	54	19	and	and	CCONJ
app01-9407	54	20	the	the	DET
app01-9407	54	21	testing	testing	NOUN
app01-9407	54	22	data	datum	NOUN
app01-9407	54	23	with	with	ADP
app01-9407	54	24	lower	low	ADJ
app01-9407	54	25	resolution	resolution	NOUN
app01-9407	54	26	10	10	NUM
app01-9407	54	27	µm3	µm3	PROPN
app01-9407	54	28	(	(	PUNCT
app01-9407	54	29	sample	sample	NOUN
app01-9407	54	30	3	3	NUM
app01-9407	54	31	)	)	PUNCT
app01-9407	54	32	.	.	PUNCT
app01-9407	55	1	volume	volume	NOUN
app01-9407	55	2	graphics	graphic	NOUN
app01-9407	55	3	was	be	AUX
app01-9407	55	4	used	use	VERB
app01-9407	55	5	for	for	ADP
app01-9407	55	6	registration	registration	NOUN
app01-9407	55	7	of	of	ADP
app01-9407	55	8	different	different	ADJ
app01-9407	55	9	resolution	resolution	NOUN
app01-9407	55	10	.	.	PUNCT
app01-9407	56	1	xct	xct	PROPN
app01-9407	56	2	scans	scan	NOUN
app01-9407	56	3	were	be	AUX
app01-9407	56	4	performed	perform	VERB
app01-9407	56	5	on	on	ADP
app01-9407	56	6	a	a	DET
app01-9407	56	7	ge	ge	PROPN
app01-9407	56	8	nanotom	nanotom	PROPN
app01-9407	56	9	180	180	NUM
app01-9407	56	10	nf	nf	ADJ
app01-9407	56	11	xct	xct	ADJ
app01-9407	56	12	-	-	PUNCT
app01-9407	56	13	device	device	NOUN
app01-9407	56	14	.	.	PUNCT
app01-9407	57	1	using	use	VERB
app01-9407	57	2	an	an	DET
app01-9407	57	3	roi	roi	NOUN
app01-9407	57	4	-	-	PUNCT
app01-9407	57	5	ct	ct	NUM
app01-9407	57	6	mode	mode	NOUN
app01-9407	57	7	,	,	PUNCT
app01-9407	57	8	3.3	3.3	NUM
app01-9407	57	9	µm	µm	ADP
app01-9407	57	10	voxel	voxel	PROPN
app01-9407	57	11	size	size	NOUN
app01-9407	57	12	can	can	AUX
app01-9407	57	13	be	be	AUX
app01-9407	57	14	reached	reach	VERB
app01-9407	57	15	.	.	PUNCT
app01-9407	58	1	an	an	DET
app01-9407	58	2	additional	additional	ADJ
app01-9407	58	3	low	low	ADJ
app01-9407	58	4	-	-	PUNCT
app01-9407	58	5	resolution	resolution	NOUN
app01-9407	58	6	scan	scan	NOUN
app01-9407	58	7	with	with	ADP
app01-9407	58	8	10	10	NUM
app01-9407	58	9	µm	µm	ADP
app01-9407	58	10	voxel	voxel	PROPN
app01-9407	58	11	size	size	NOUN
app01-9407	58	12	was	be	AUX
app01-9407	58	13	performed	perform	VERB
app01-9407	58	14	on	on	ADP
app01-9407	58	15	the	the	DET
app01-9407	58	16	vari	vari	ADJ
app01-9407	58	17	material	material	NOUN
app01-9407	58	18	(	(	PUNCT
app01-9407	58	19	sample	sample	NOUN
app01-9407	58	20	3	3	NUM
app01-9407	58	21	)	)	PUNCT
app01-9407	58	22	.	.	PUNCT
app01-9407	59	1	for	for	ADP
app01-9407	59	2	further	further	ADJ
app01-9407	59	3	data	datum	NOUN
app01-9407	59	4	processing	processing	NOUN
app01-9407	59	5	,	,	PUNCT
app01-9407	59	6	following	follow	VERB
app01-9407	59	7	cut	cut	VERB
app01-9407	59	8	-	-	PUNCT
app01-9407	59	9	out	out	ADP
app01-9407	59	10	dimensions	dimension	NOUN
app01-9407	59	11	of	of	ADP
app01-9407	59	12	the	the	DET
app01-9407	59	13	datasets	dataset	NOUN
app01-9407	59	14	were	be	AUX
app01-9407	59	15	used	use	VERB
app01-9407	59	16	,	,	PUNCT
app01-9407	59	17	shown	show	VERB
app01-9407	59	18	in	in	ADP
app01-9407	59	19	table	table	NOUN
app01-9407	59	20	1	1	NUM
app01-9407	59	21	.	.	PUNCT
app01-9407	59	22	resolution	resolution	NOUN
app01-9407	59	23	size	size	NOUN
app01-9407	59	24	sample	sample	NOUN
app01-9407	59	25	1	1	NUM
app01-9407	59	26	3.3	3.3	NUM
app01-9407	59	27	µm	µm	ADP
app01-9407	59	28	1300	1300	NUM
app01-9407	59	29	×	×	NOUN
app01-9407	59	30	900	900	NUM
app01-9407	59	31	×	×	NOUN
app01-9407	59	32	976	976	NUM
app01-9407	59	33	sample	sample	NOUN
app01-9407	59	34	2	2	NUM
app01-9407	59	35	3.3	3.3	NUM
app01-9407	59	36	µm	µm	ADP
app01-9407	59	37	1300	1300	NUM
app01-9407	59	38	×	×	NOUN
app01-9407	59	39	900	900	NUM
app01-9407	59	40	×	×	NOUN
app01-9407	59	41	976	976	NUM
app01-9407	59	42	sample	sample	NOUN
app01-9407	59	43	3	3	NUM
app01-9407	59	44	10	10	NUM
app01-9407	59	45	µm	µm	PART
app01-9407	59	46	366	366	NUM
app01-9407	59	47	×	×	NOUN
app01-9407	59	48	244	244	NUM
app01-9407	59	49	×	×	NOUN
app01-9407	59	50	244	244	NUM
app01-9407	59	51	table	table	NOUN
app01-9407	59	52	1	1	NUM
app01-9407	59	53	.	.	PUNCT
app01-9407	59	54	sample	sample	NOUN
app01-9407	59	55	modality	modality	NOUN
app01-9407	59	56	88	88	NUM
app01-9407	59	57	vol	vol	NOUN
app01-9407	59	58	.	.	PUNCT
app01-9407	60	1	42/2023	42/2023	NUM
app01-9407	60	2	cfrp	cfrp	PROPN
app01-9407	60	3	pore	pore	ADJ
app01-9407	60	4	segmentation	segmentation	NOUN
app01-9407	60	5	via	via	ADP
app01-9407	60	6	u	u	ADJ
app01-9407	60	7	-	-	ADJ
app01-9407	60	8	net	net	ADJ
app01-9407	60	9	figure	figure	NOUN
app01-9407	60	10	2	2	NUM
app01-9407	60	11	.	.	PUNCT
app01-9407	61	1	the	the	DET
app01-9407	61	2	modified	modify	VERB
app01-9407	61	3	3d	3d	PROPN
app01-9407	61	4	u	u	ADJ
app01-9407	61	5	-	-	ADJ
app01-9407	61	6	net	net	ADJ
app01-9407	61	7	architecture	architecture	NOUN
app01-9407	61	8	[	[	X
app01-9407	61	9	5	5	NUM
app01-9407	61	10	]	]	PUNCT
app01-9407	61	11	.	.	PUNCT
app01-9407	62	1	4	4	X
app01-9407	62	2	.	.	X
app01-9407	62	3	methods	method	NOUN
app01-9407	62	4	in	in	ADP
app01-9407	62	5	the	the	DET
app01-9407	62	6	following	follow	VERB
app01-9407	62	7	section	section	NOUN
app01-9407	62	8	we	we	PRON
app01-9407	62	9	describe	describe	VERB
app01-9407	62	10	our	our	PRON
app01-9407	62	11	modified	modified	ADJ
app01-9407	62	12	3d	3d	NUM
app01-9407	62	13	u	u	NOUN
app01-9407	62	14	-	-	NOUN
app01-9407	62	15	net	net	ADJ
app01-9407	62	16	(	(	PUNCT
app01-9407	62	17	section	section	NOUN
app01-9407	62	18	4.1	4.1	NUM
app01-9407	62	19	)	)	PUNCT
app01-9407	62	20	,	,	PUNCT
app01-9407	62	21	the	the	DET
app01-9407	62	22	respective	respective	ADJ
app01-9407	62	23	data	datum	NOUN
app01-9407	62	24	pre	pre	ADJ
app01-9407	62	25	-	-	ADJ
app01-9407	62	26	processing	processing	ADJ
app01-9407	62	27	steps	step	NOUN
app01-9407	62	28	,	,	PUNCT
app01-9407	62	29	as	as	ADV
app01-9407	62	30	well	well	ADV
app01-9407	62	31	as	as	ADP
app01-9407	62	32	the	the	DET
app01-9407	62	33	used	use	VERB
app01-9407	62	34	training	training	NOUN
app01-9407	62	35	process	process	NOUN
app01-9407	62	36	(	(	PUNCT
app01-9407	62	37	section	section	NOUN
app01-9407	62	38	4.2	4.2	NUM
app01-9407	62	39	)	)	PUNCT
app01-9407	62	40	.	.	PUNCT
app01-9407	63	1	4.1	4.1	NUM
app01-9407	63	2	.	.	PUNCT
app01-9407	63	3	modification	modification	NOUN
app01-9407	63	4	of	of	ADP
app01-9407	63	5	3d	3d	NUM
app01-9407	63	6	u	u	NOUN
app01-9407	63	7	-	-	ADJ
app01-9407	63	8	net	net	ADJ
app01-9407	63	9	u	u	NOUN
app01-9407	63	10	-	-	NOUN
app01-9407	63	11	net	net	ADJ
app01-9407	63	12	is	be	AUX
app01-9407	63	13	a	a	DET
app01-9407	63	14	convolutional	convolutional	ADJ
app01-9407	63	15	neural	neural	ADJ
app01-9407	63	16	network	network	NOUN
app01-9407	63	17	that	that	PRON
app01-9407	63	18	features	feature	VERB
app01-9407	63	19	high	high	ADJ
app01-9407	63	20	performance	performance	NOUN
app01-9407	63	21	with	with	ADP
app01-9407	63	22	a	a	DET
app01-9407	63	23	small	small	ADJ
app01-9407	63	24	amount	amount	NOUN
app01-9407	63	25	of	of	ADP
app01-9407	63	26	training	training	NOUN
app01-9407	63	27	,	,	PUNCT
app01-9407	63	28	which	which	PRON
app01-9407	63	29	can	can	AUX
app01-9407	63	30	be	be	AUX
app01-9407	63	31	applied	apply	VERB
app01-9407	63	32	both	both	PRON
app01-9407	63	33	for	for	ADP
app01-9407	63	34	2d	2d	NUM
app01-9407	63	35	and	and	CCONJ
app01-9407	63	36	3d	3d	NUM
app01-9407	63	37	data	datum	NOUN
app01-9407	63	38	.	.	PUNCT
app01-9407	64	1	we	we	PRON
app01-9407	64	2	implemented	implement	VERB
app01-9407	64	3	an	an	DET
app01-9407	64	4	architecture	architecture	NOUN
app01-9407	64	5	in	in	ADP
app01-9407	64	6	python	python	NOUN
app01-9407	65	1	[	[	X
app01-9407	65	2	20	20	NUM
app01-9407	65	3	]	]	PUNCT
app01-9407	65	4	using	use	VERB
app01-9407	65	5	keras	keras	PROPN
app01-9407	65	6	[	[	X
app01-9407	65	7	21	21	NUM
app01-9407	65	8	]	]	PUNCT
app01-9407	65	9	with	with	ADP
app01-9407	65	10	tensorflow	tensorflow	NOUN
app01-9407	65	11	[	[	X
app01-9407	65	12	22	22	NUM
app01-9407	65	13	]	]	PUNCT
app01-9407	65	14	as	as	ADP
app01-9407	65	15	a	a	DET
app01-9407	65	16	backend	backend	NOUN
app01-9407	65	17	which	which	PRON
app01-9407	65	18	was	be	AUX
app01-9407	65	19	inspired	inspire	VERB
app01-9407	65	20	by	by	ADP
app01-9407	65	21	3d	3d	PROPN
app01-9407	65	22	u	u	NOUN
app01-9407	65	23	-	-	NOUN
app01-9407	65	24	net	net	NOUN
app01-9407	65	25	.	.	PUNCT
app01-9407	66	1	the	the	DET
app01-9407	66	2	modified	modify	VERB
app01-9407	66	3	3d	3d	PROPN
app01-9407	66	4	u	u	ADJ
app01-9407	66	5	-	-	ADJ
app01-9407	66	6	net	net	ADJ
app01-9407	66	7	network	network	NOUN
app01-9407	66	8	contains	contain	VERB
app01-9407	66	9	two	two	NUM
app01-9407	66	10	paths	path	NOUN
app01-9407	66	11	respectively	respectively	ADV
app01-9407	66	12	called	call	VERB
app01-9407	66	13	analysis	analysis	NOUN
app01-9407	66	14	and	and	CCONJ
app01-9407	66	15	synthesis	synthesis	NOUN
app01-9407	66	16	path	path	NOUN
app01-9407	66	17	,	,	PUNCT
app01-9407	66	18	with	with	ADP
app01-9407	66	19	a	a	DET
app01-9407	66	20	u	u	NOUN
app01-9407	66	21	-	-	ADJ
app01-9407	66	22	shaped	shape	VERB
app01-9407	66	23	architecture	architecture	NOUN
app01-9407	66	24	containing	contain	VERB
app01-9407	66	25	in	in	ADP
app01-9407	66	26	total	total	ADJ
app01-9407	66	27	46	46	NUM
app01-9407	66	28	layers	layer	NOUN
app01-9407	66	29	(	(	PUNCT
app01-9407	66	30	see	see	VERB
app01-9407	66	31	figure	figure	NOUN
app01-9407	66	32	2	2	NUM
app01-9407	66	33	)	)	PUNCT
app01-9407	66	34	.	.	PUNCT
app01-9407	67	1	the	the	DET
app01-9407	67	2	modified	modify	VERB
app01-9407	67	3	3d	3d	PROPN
app01-9407	67	4	u	u	ADJ
app01-9407	67	5	-	-	ADJ
app01-9407	67	6	net	net	ADJ
app01-9407	67	7	architecture	architecture	NOUN
app01-9407	67	8	contains	contain	VERB
app01-9407	67	9	the	the	DET
app01-9407	67	10	input	input	NOUN
app01-9407	67	11	layer	layer	NOUN
app01-9407	67	12	,	,	PUNCT
app01-9407	67	13	encoder	encoder	NOUN
app01-9407	67	14	and	and	CCONJ
app01-9407	67	15	decoder	decoder	NOUN
app01-9407	67	16	sections	section	NOUN
app01-9407	67	17	,	,	PUNCT
app01-9407	67	18	a	a	DET
app01-9407	67	19	final	final	ADJ
app01-9407	67	20	convolution	convolution	NOUN
app01-9407	67	21	and	and	CCONJ
app01-9407	67	22	transpose	transpose	ADJ
app01-9407	67	23	output	output	NOUN
app01-9407	67	24	layer	layer	NOUN
app01-9407	67	25	.	.	PUNCT
app01-9407	68	1	for	for	ADP
app01-9407	68	2	up	up	ADV
app01-9407	68	3	-	-	PUNCT
app01-9407	68	4	sampling	sampling	NOUN
app01-9407	68	5	,	,	PUNCT
app01-9407	68	6	the	the	DET
app01-9407	68	7	decoder	decoder	NOUN
app01-9407	68	8	subnetworks	subnetwork	NOUN
app01-9407	68	9	also	also	ADV
app01-9407	68	10	contain	contain	VERB
app01-9407	68	11	one	one	NUM
app01-9407	68	12	more	more	ADJ
app01-9407	68	13	convolution	convolution	NOUN
app01-9407	68	14	layer	layer	NOUN
app01-9407	68	15	.	.	PUNCT
app01-9407	69	1	first	first	ADV
app01-9407	69	2	,	,	PUNCT
app01-9407	69	3	encoder	encoder	NOUN
app01-9407	69	4	sections	section	NOUN
app01-9407	69	5	analyze	analyze	VERB
app01-9407	69	6	the	the	DET
app01-9407	69	7	whole	whole	ADJ
app01-9407	69	8	image	image	NOUN
app01-9407	69	9	,	,	PUNCT
app01-9407	69	10	then	then	ADV
app01-9407	69	11	the	the	DET
app01-9407	69	12	decoder	decoder	NOUN
app01-9407	69	13	sections	section	NOUN
app01-9407	69	14	produce	produce	VERB
app01-9407	69	15	and	and	CCONJ
app01-9407	69	16	predict	predict	VERB
app01-9407	69	17	the	the	DET
app01-9407	69	18	segmented	segmented	ADJ
app01-9407	69	19	volume	volume	NOUN
app01-9407	69	20	[	[	X
app01-9407	69	21	5	5	NUM
app01-9407	69	22	]	]	PUNCT
app01-9407	69	23	.	.	PUNCT
app01-9407	70	1	the	the	DET
app01-9407	70	2	modified	modify	VERB
app01-9407	70	3	3d	3d	PROPN
app01-9407	70	4	u	u	NOUN
app01-9407	70	5	-	-	NOUN
app01-9407	70	6	net	net	NOUN
app01-9407	70	7	has	have	VERB
app01-9407	70	8	two	two	NUM
app01-9407	70	9	additional	additional	ADJ
app01-9407	70	10	convolutional	convolutional	ADJ
app01-9407	70	11	and	and	CCONJ
app01-9407	70	12	rectified	rectified	ADJ
app01-9407	70	13	linear	linear	NOUN
app01-9407	70	14	unit	unit	NOUN
app01-9407	70	15	(	(	PUNCT
app01-9407	70	16	relu	relu	NOUN
app01-9407	70	17	)	)	PUNCT
app01-9407	70	18	layers	layer	NOUN
app01-9407	70	19	in	in	ADP
app01-9407	70	20	the	the	DET
app01-9407	70	21	decoder	decoder	NOUN
app01-9407	70	22	stages	stage	NOUN
app01-9407	70	23	if	if	SCONJ
app01-9407	70	24	we	we	PRON
app01-9407	70	25	compare	compare	VERB
app01-9407	70	26	with	with	ADP
app01-9407	70	27	original	original	ADJ
app01-9407	70	28	3d	3d	NUM
app01-9407	70	29	u	u	NOUN
app01-9407	70	30	-	-	NOUN
app01-9407	70	31	net	net	NOUN
app01-9407	70	32	.	.	PUNCT
app01-9407	71	1	rather	rather	ADV
app01-9407	71	2	than	than	ADP
app01-9407	71	3	employing	employ	VERB
app01-9407	71	4	a	a	DET
app01-9407	71	5	softmax	softmax	NOUN
app01-9407	71	6	function	function	NOUN
app01-9407	71	7	,	,	PUNCT
app01-9407	71	8	a	a	DET
app01-9407	71	9	sigmoid	sigmoid	NOUN
app01-9407	71	10	function	function	NOUN
app01-9407	71	11	was	be	AUX
app01-9407	71	12	utilized	utilize	VERB
app01-9407	71	13	in	in	ADP
app01-9407	71	14	this	this	DET
app01-9407	71	15	case	case	NOUN
app01-9407	71	16	to	to	PART
app01-9407	71	17	reduce	reduce	VERB
app01-9407	71	18	complexity	complexity	NOUN
app01-9407	71	19	in	in	ADP
app01-9407	71	20	the	the	DET
app01-9407	71	21	final	final	ADJ
app01-9407	71	22	classification	classification	NOUN
app01-9407	71	23	.	.	PUNCT
app01-9407	72	1	additional	additional	ADJ
app01-9407	72	2	layers	layer	NOUN
app01-9407	72	3	were	be	AUX
app01-9407	72	4	integrated	integrate	VERB
app01-9407	72	5	into	into	ADP
app01-9407	72	6	the	the	DET
app01-9407	72	7	decoder	decoder	NOUN
app01-9407	72	8	stages	stage	NOUN
app01-9407	72	9	with	with	ADP
app01-9407	72	10	the	the	DET
app01-9407	72	11	aim	aim	NOUN
app01-9407	72	12	of	of	ADP
app01-9407	72	13	enhancing	enhance	VERB
app01-9407	72	14	the	the	DET
app01-9407	72	15	final	final	ADJ
app01-9407	72	16	prediction	prediction	NOUN
app01-9407	72	17	accuracy	accuracy	NOUN
app01-9407	72	18	.	.	PUNCT
app01-9407	73	1	as	as	SCONJ
app01-9407	73	2	demonstrated	demonstrate	VERB
app01-9407	73	3	in	in	ADP
app01-9407	73	4	our	our	PRON
app01-9407	73	5	previous	previous	ADJ
app01-9407	73	6	study	study	NOUN
app01-9407	73	7	[	[	X
app01-9407	73	8	5	5	NUM
app01-9407	73	9	]	]	PUNCT
app01-9407	73	10	,	,	PUNCT
app01-9407	73	11	the	the	DET
app01-9407	73	12	segmentation	segmentation	NOUN
app01-9407	73	13	results	result	NOUN
app01-9407	73	14	were	be	AUX
app01-9407	73	15	improved	improve	VERB
app01-9407	73	16	compared	compare	VERB
app01-9407	73	17	to	to	ADP
app01-9407	73	18	the	the	DET
app01-9407	73	19	original	original	ADJ
app01-9407	73	20	u	u	ADJ
app01-9407	73	21	-	-	ADJ
app01-9407	73	22	net	net	ADJ
app01-9407	73	23	architecture	architecture	NOUN
app01-9407	73	24	.	.	PUNCT
app01-9407	74	1	the	the	PRON
app01-9407	74	2	modified	modify	VERB
app01-9407	74	3	3d	3d	PROPN
app01-9407	74	4	u	u	NOUN
app01-9407	74	5	-	-	NOUN
app01-9407	74	6	net	net	NOUN
app01-9407	74	7	uses	use	VERB
app01-9407	74	8	an	an	DET
app01-9407	74	9	overlap	overlap	NOUN
app01-9407	74	10	-	-	PUNCT
app01-9407	74	11	tile	tile	NOUN
app01-9407	74	12	strategy	strategy	NOUN
app01-9407	74	13	to	to	PART
app01-9407	74	14	predict	predict	VERB
app01-9407	74	15	the	the	DET
app01-9407	74	16	labels	label	NOUN
app01-9407	74	17	for	for	ADP
app01-9407	74	18	each	each	DET
app01-9407	74	19	test	test	NOUN
app01-9407	74	20	volume	volume	NOUN
app01-9407	74	21	.	.	PUNCT
app01-9407	75	1	the	the	DET
app01-9407	75	2	input	input	NOUN
app01-9407	75	3	volume	volume	NOUN
app01-9407	75	4	has	have	VERB
app01-9407	75	5	dimensions	dimension	NOUN
app01-9407	75	6	of	of	ADP
app01-9407	75	7	132	132	NUM
app01-9407	75	8	×	×	NOUN
app01-9407	75	9	132	132	NUM
app01-9407	75	10	×	×	NOUN
app01-9407	75	11	132	132	NUM
app01-9407	75	12	,	,	PUNCT
app01-9407	75	13	while	while	SCONJ
app01-9407	75	14	the	the	DET
app01-9407	75	15	prediction	prediction	NOUN
app01-9407	75	16	size	size	NOUN
app01-9407	75	17	is	be	AUX
app01-9407	75	18	122	122	NUM
app01-9407	75	19	×	×	NOUN
app01-9407	75	20	122	122	NUM
app01-9407	75	21	×	×	NOUN
app01-9407	75	22	122	122	NUM
app01-9407	75	23	.	.	PUNCT
app01-9407	76	1	the	the	DET
app01-9407	76	2	variation	variation	NOUN
app01-9407	76	3	in	in	ADP
app01-9407	76	4	input	input	NOUN
app01-9407	76	5	and	and	CCONJ
app01-9407	76	6	output	output	NOUN
app01-9407	76	7	sizes	size	NOUN
app01-9407	76	8	is	be	AUX
app01-9407	76	9	due	due	ADJ
app01-9407	76	10	to	to	ADP
app01-9407	76	11	a	a	DET
app01-9407	76	12	5	5	NUM
app01-9407	76	13	-	-	PUNCT
app01-9407	76	14	voxel	voxel	ADJ
app01-9407	76	15	overlap	overlap	NOUN
app01-9407	76	16	in	in	ADP
app01-9407	76	17	each	each	DET
app01-9407	76	18	dimension	dimension	NOUN
app01-9407	76	19	.	.	PUNCT
app01-9407	77	1	this	this	DET
app01-9407	77	2	overlapping	overlap	VERB
app01-9407	77	3	strategy	strategy	NOUN
app01-9407	77	4	enhances	enhance	VERB
app01-9407	77	5	prediction	prediction	NOUN
app01-9407	77	6	accuracy	accuracy	NOUN
app01-9407	77	7	,	,	PUNCT
app01-9407	77	8	particularly	particularly	ADV
app01-9407	77	9	at	at	ADP
app01-9407	77	10	the	the	DET
app01-9407	77	11	borders	border	NOUN
app01-9407	77	12	of	of	ADP
app01-9407	77	13	the	the	DET
app01-9407	77	14	sub	sub	NOUN
app01-9407	77	15	-	-	NOUN
app01-9407	77	16	volumes	volume	NOUN
app01-9407	77	17	[	[	X
app01-9407	77	18	17	17	NUM
app01-9407	77	19	]	]	PUNCT
app01-9407	77	20	.	.	PUNCT
app01-9407	78	1	the	the	DET
app01-9407	78	2	neural	neural	ADJ
app01-9407	78	3	network	network	NOUN
app01-9407	78	4	model	model	NOUN
app01-9407	78	5	(	(	PUNCT
app01-9407	78	6	the	the	DET
app01-9407	78	7	weights	weight	NOUN
app01-9407	78	8	)	)	PUNCT
app01-9407	78	9	was	be	AUX
app01-9407	78	10	saved	save	VERB
app01-9407	78	11	in	in	ADP
app01-9407	78	12	onnx	onnx	NOUN
app01-9407	78	13	(	(	PUNCT
app01-9407	78	14	open	open	ADJ
app01-9407	78	15	neural	neural	ADJ
app01-9407	78	16	network	network	NOUN
app01-9407	78	17	exchange	exchange	NOUN
app01-9407	78	18	)	)	PUNCT
app01-9407	78	19	format	format	NOUN
app01-9407	78	20	[	[	X
app01-9407	78	21	23	23	NUM
app01-9407	78	22	]	]	PUNCT
app01-9407	78	23	for	for	ADP
app01-9407	78	24	use	use	NOUN
app01-9407	78	25	in	in	ADP
app01-9407	78	26	open_ia	open_ia	NOUN
app01-9407	78	27	[	[	X
app01-9407	78	28	13	13	NUM
app01-9407	78	29	]	]	PUNCT
app01-9407	78	30	.	.	PUNCT
app01-9407	79	1	4.2	4.2	NUM
app01-9407	79	2	.	.	PUNCT
app01-9407	80	1	data	datum	NOUN
app01-9407	80	2	pre	pre	ADJ
app01-9407	80	3	-	-	ADJ
app01-9407	80	4	processing	processing	ADJ
app01-9407	80	5	,	,	PUNCT
app01-9407	80	6	training	training	NOUN
app01-9407	80	7	and	and	CCONJ
app01-9407	80	8	testing	test	VERB
app01-9407	80	9	the	the	DET
app01-9407	80	10	preprocessing	preprocessing	NOUN
app01-9407	80	11	pipeline	pipeline	NOUN
app01-9407	80	12	involved	involve	VERB
app01-9407	80	13	several	several	ADJ
app01-9407	80	14	steps	step	NOUN
app01-9407	80	15	,	,	PUNCT
app01-9407	80	16	including	include	VERB
app01-9407	80	17	volume	volume	NOUN
app01-9407	80	18	normalization	normalization	NOUN
app01-9407	80	19	,	,	PUNCT
app01-9407	80	20	mirror	mirror	NOUN
app01-9407	80	21	-	-	PUNCT
app01-9407	80	22	padding	padding	NOUN
app01-9407	80	23	with	with	ADP
app01-9407	80	24	5	5	NUM
app01-9407	80	25	extra	extra	ADJ
app01-9407	80	26	voxels	voxel	NOUN
app01-9407	80	27	,	,	PUNCT
app01-9407	80	28	and	and	CCONJ
app01-9407	80	29	division	division	NOUN
app01-9407	80	30	of	of	ADP
app01-9407	80	31	the	the	DET
app01-9407	80	32	ct	ct	PROPN
app01-9407	80	33	scan	scan	NOUN
app01-9407	80	34	into	into	ADP
app01-9407	80	35	subvolumes	subvolume	NOUN
app01-9407	80	36	with	with	ADP
app01-9407	80	37	a	a	DET
app01-9407	80	38	5	5	NUM
app01-9407	80	39	-	-	PUNCT
app01-9407	80	40	voxel	voxel	ADJ
app01-9407	80	41	overlap	overlap	NOUN
app01-9407	80	42	in	in	ADP
app01-9407	80	43	all	all	DET
app01-9407	80	44	directions	direction	NOUN
app01-9407	80	45	.	.	PUNCT
app01-9407	81	1	the	the	DET
app01-9407	81	2	labeled	label	VERB
app01-9407	81	3	image	image	NOUN
app01-9407	81	4	,	,	PUNCT
app01-9407	81	5	processed	process	VERB
app01-9407	81	6	using	use	VERB
app01-9407	81	7	otsu	otsu	NOUN
app01-9407	81	8	thresholding	thresholding	NOUN
app01-9407	81	9	,	,	PUNCT
app01-9407	81	10	was	be	AUX
app01-9407	81	11	also	also	ADV
app01-9407	81	12	split	split	VERB
app01-9407	81	13	into	into	ADP
app01-9407	81	14	sub	sub	NOUN
app01-9407	81	15	-	-	NOUN
app01-9407	81	16	volumes	volume	NOUN
app01-9407	81	17	for	for	ADP
app01-9407	81	18	training	training	NOUN
app01-9407	81	19	.	.	PUNCT
app01-9407	82	1	the	the	DET
app01-9407	82	2	primary	primary	ADJ
app01-9407	82	3	reason	reason	NOUN
app01-9407	82	4	for	for	ADP
app01-9407	82	5	splitting	split	VERB
app01-9407	82	6	xct	xct	PROPN
app01-9407	82	7	data	datum	NOUN
app01-9407	82	8	into	into	ADP
app01-9407	82	9	sub	sub	NOUN
app01-9407	82	10	-	-	NOUN
app01-9407	82	11	volumes	volume	NOUN
app01-9407	82	12	with	with	ADP
app01-9407	82	13	a	a	DET
app01-9407	82	14	size	size	NOUN
app01-9407	82	15	of	of	ADP
app01-9407	82	16	132	132	NUM
app01-9407	82	17	×	×	NOUN
app01-9407	82	18	132	132	NUM
app01-9407	82	19	×	×	NOUN
app01-9407	82	20	132	132	NUM
app01-9407	82	21	is	be	AUX
app01-9407	82	22	to	to	PART
app01-9407	82	23	optimize	optimize	VERB
app01-9407	82	24	the	the	DET
app01-9407	82	25	training	training	NOUN
app01-9407	82	26	time	time	NOUN
app01-9407	82	27	and	and	CCONJ
app01-9407	82	28	augment	augment	VERB
app01-9407	82	29	the	the	DET
app01-9407	82	30	training	training	NOUN
app01-9407	82	31	data	datum	NOUN
app01-9407	82	32	,	,	PUNCT
app01-9407	82	33	thereby	thereby	ADV
app01-9407	82	34	reducing	reduce	VERB
app01-9407	82	35	the	the	DET
app01-9407	82	36	need	need	NOUN
app01-9407	82	37	for	for	ADP
app01-9407	82	38	scanning	scan	VERB
app01-9407	82	39	multiple	multiple	ADJ
app01-9407	82	40	specimens	specimen	NOUN
app01-9407	82	41	.	.	PUNCT
app01-9407	83	1	for	for	ADP
app01-9407	83	2	training	training	NOUN
app01-9407	83	3	,	,	PUNCT
app01-9407	83	4	only	only	ADV
app01-9407	83	5	the	the	DET
app01-9407	83	6	sub	sub	NOUN
app01-9407	83	7	-	-	NOUN
app01-9407	83	8	volumes	volume	NOUN
app01-9407	83	9	from	from	ADP
app01-9407	83	10	sample	sample	NOUN
app01-9407	83	11	1	1	NUM
app01-9407	83	12	were	be	AUX
app01-9407	83	13	used	use	VERB
app01-9407	83	14	,	,	PUNCT
app01-9407	83	15	which	which	PRON
app01-9407	83	16	were	be	AUX
app01-9407	83	17	further	far	ADV
app01-9407	83	18	divided	divide	VERB
app01-9407	83	19	into	into	ADP
app01-9407	83	20	three	three	NUM
app01-9407	83	21	parts	part	NOUN
app01-9407	83	22	:	:	PUNCT
app01-9407	83	23	64	64	NUM
app01-9407	83	24	%	%	NOUN
app01-9407	83	25	(	(	PUNCT
app01-9407	83	26	358	358	NUM
app01-9407	83	27	samples	sample	NOUN
app01-9407	83	28	)	)	PUNCT
app01-9407	83	29	,	,	PUNCT
app01-9407	83	30	for	for	ADP
app01-9407	83	31	training	training	NOUN
app01-9407	83	32	,	,	PUNCT
app01-9407	83	33	20	20	NUM
app01-9407	83	34	%	%	NOUN
app01-9407	83	35	(	(	PUNCT
app01-9407	83	36	112	112	NUM
app01-9407	83	37	samples	sample	NOUN
app01-9407	83	38	)	)	PUNCT
app01-9407	83	39	for	for	ADP
app01-9407	83	40	testing	testing	NOUN
app01-9407	83	41	,	,	PUNCT
app01-9407	83	42	and	and	CCONJ
app01-9407	83	43	16	16	NUM
app01-9407	83	44	%	%	NOUN
app01-9407	83	45	(	(	PUNCT
app01-9407	83	46	90	90	NUM
app01-9407	83	47	samples	sample	NOUN
app01-9407	83	48	)	)	PUNCT
app01-9407	83	49	for	for	ADP
app01-9407	83	50	validation	validation	NOUN
app01-9407	83	51	.	.	PUNCT
app01-9407	84	1	to	to	PART
app01-9407	84	2	identify	identify	VERB
app01-9407	84	3	optimal	optimal	ADJ
app01-9407	84	4	training	training	NOUN
app01-9407	84	5	parameters	parameter	NOUN
app01-9407	84	6	,	,	PUNCT
app01-9407	84	7	a	a	DET
app01-9407	84	8	heuristic	heuristic	ADJ
app01-9407	84	9	hyperparameter	hyperparameter	NOUN
app01-9407	84	10	tuning	tune	VERB
app01-9407	84	11	with	with	ADP
app01-9407	84	12	talos	talo	NOUN
app01-9407	84	13	[	[	X
app01-9407	84	14	24	24	NUM
app01-9407	84	15	]	]	PUNCT
app01-9407	84	16	was	be	AUX
app01-9407	84	17	employed	employ	VERB
app01-9407	84	18	.	.	PUNCT
app01-9407	85	1	in	in	ADP
app01-9407	85	2	the	the	DET
app01-9407	85	3	hyperparameter	hyperparameter	NOUN
app01-9407	85	4	tuning	tuning	NOUN
app01-9407	85	5	of	of	ADP
app01-9407	85	6	our	our	PRON
app01-9407	85	7	training	training	NOUN
app01-9407	85	8	we	we	PRON
app01-9407	85	9	tested	test	VERB
app01-9407	85	10	various	various	ADJ
app01-9407	85	11	combinations	combination	NOUN
app01-9407	85	12	of	of	ADP
app01-9407	85	13	learning	learn	VERB
app01-9407	85	14	rates	rate	NOUN
app01-9407	85	15	,	,	PUNCT
app01-9407	85	16	optimizers	optimizer	NOUN
app01-9407	85	17	,	,	PUNCT
app01-9407	85	18	and	and	CCONJ
app01-9407	85	19	epochs	epoch	NOUN
app01-9407	85	20	to	to	PART
app01-9407	85	21	achieve	achieve	VERB
app01-9407	85	22	an	an	DET
app01-9407	85	23	optimum	optimum	NOUN
app01-9407	85	24	between	between	ADP
app01-9407	85	25	accuracy	accuracy	NOUN
app01-9407	85	26	and	and	CCONJ
app01-9407	85	27	training	training	NOUN
app01-9407	85	28	time	time	NOUN
app01-9407	85	29	.	.	PUNCT
app01-9407	86	1	the	the	DET
app01-9407	86	2	hyperparameter	hyperparameter	NOUN
app01-9407	86	3	tuning	tuning	NOUN
app01-9407	86	4	process	process	NOUN
app01-9407	86	5	yielded	yield	VERB
app01-9407	86	6	the	the	DET
app01-9407	86	7	best	good	ADJ
app01-9407	86	8	outcomes	outcome	NOUN
app01-9407	86	9	when	when	SCONJ
app01-9407	86	10	employing	employ	VERB
app01-9407	86	11	the	the	DET
app01-9407	86	12	adam	adam	PROPN
app01-9407	86	13	optimizer	optimizer	NOUN
app01-9407	86	14	[	[	X
app01-9407	86	15	25	25	NUM
app01-9407	86	16	]	]	PUNCT
app01-9407	86	17	with	with	ADP
app01-9407	86	18	a	a	DET
app01-9407	86	19	learning	learn	VERB
app01-9407	86	20	rate	rate	NOUN
app01-9407	86	21	of	of	ADP
app01-9407	86	22	4e-5	4e-5	PROPN
app01-9407	86	23	.	.	PUNCT
app01-9407	87	1	the	the	DET
app01-9407	87	2	training	training	NOUN
app01-9407	87	3	process	process	NOUN
app01-9407	87	4	,	,	PUNCT
app01-9407	87	5	conducted	conduct	VERB
app01-9407	87	6	on	on	ADP
app01-9407	87	7	a	a	DET
app01-9407	87	8	workstation	workstation	NOUN
app01-9407	87	9	powered	power	VERB
app01-9407	87	10	by	by	ADP
app01-9407	87	11	an	an	DET
app01-9407	87	12	nvidia	nvidia	PROPN
app01-9407	87	13	quadro	quadro	PROPN
app01-9407	87	14	rtx	rtx	PROPN
app01-9407	87	15	6000	6000	NUM
app01-9407	87	16	with	with	ADP
app01-9407	87	17	keras	keras	PROPN
app01-9407	87	18	2.3.0	2.3.0	NUM
app01-9407	87	19	and	and	CCONJ
app01-9407	87	20	tensorflow	tensorflow	NOUN
app01-9407	87	21	-	-	PUNCT
app01-9407	87	22	gpu	gpu	NOUN
app01-9407	87	23	2.1.0	2.1.0	NUM
app01-9407	87	24	,	,	PUNCT
app01-9407	87	25	was	be	AUX
app01-9407	87	26	successfully	successfully	ADV
app01-9407	87	27	completed	complete	VERB
app01-9407	87	28	in	in	ADP
app01-9407	87	29	approximately	approximately	ADV
app01-9407	87	30	1	1	NUM
app01-9407	87	31	hour	hour	NOUN
app01-9407	87	32	,	,	PUNCT
app01-9407	87	33	involving	involve	VERB
app01-9407	87	34	10	10	NUM
app01-9407	87	35	epochs	epoch	NOUN
app01-9407	87	36	and	and	CCONJ
app01-9407	87	37	a	a	DET
app01-9407	87	38	batch	batch	NOUN
app01-9407	87	39	size	size	NOUN
app01-9407	87	40	of	of	ADP
app01-9407	87	41	3	3	NUM
app01-9407	87	42	.	.	PUNCT
app01-9407	87	43	to	to	PART
app01-9407	87	44	distinguish	distinguish	VERB
app01-9407	87	45	classes	class	NOUN
app01-9407	87	46	with	with	ADP
app01-9407	87	47	significant	significant	ADJ
app01-9407	87	48	imbalances	imbalance	NOUN
app01-9407	87	49	during	during	ADP
app01-9407	87	50	training	training	NOUN
app01-9407	87	51	,	,	PUNCT
app01-9407	87	52	the	the	DET
app01-9407	87	53	dice	dice	NOUN
app01-9407	87	54	coefficient	coefficient	NOUN
app01-9407	87	55	[	[	X
app01-9407	87	56	26	26	NUM
app01-9407	87	57	]	]	SYM
app01-9407	87	58	89	89	NUM
app01-9407	87	59	m.	m.	NOUN
app01-9407	87	60	yosifov	yosifov	NOUN
app01-9407	87	61	,	,	PUNCT
app01-9407	87	62	p.	p.	NOUN
app01-9407	87	63	weinberger	weinberger	PROPN
app01-9407	87	64	,	,	PUNCT
app01-9407	87	65	b.	b.	PROPN
app01-9407	87	66	plank	plank	PROPN
app01-9407	87	67	et	et	PROPN
app01-9407	87	68	al	al	PROPN
app01-9407	87	69	.	.	PROPN
app01-9407	87	70	acta	acta	PROPN
app01-9407	87	71	polytechnica	polytechnica	PROPN
app01-9407	87	72	ctu	ctu	PROPN
app01-9407	87	73	proceedings	proceeding	NOUN
app01-9407	87	74	figure	figure	VERB
app01-9407	87	75	3	3	NUM
app01-9407	87	76	.	.	PUNCT
app01-9407	88	1	the	the	DET
app01-9407	88	2	visualization	visualization	NOUN
app01-9407	88	3	consists	consist	VERB
app01-9407	88	4	of	of	ADP
app01-9407	88	5	2d	2d	NUM
app01-9407	88	6	and	and	CCONJ
app01-9407	88	7	3d	3d	NUM
app01-9407	88	8	representations	representation	NOUN
app01-9407	88	9	of	of	ADP
app01-9407	88	10	a	a	DET
app01-9407	88	11	sub	sub	NOUN
app01-9407	88	12	-	-	NOUN
app01-9407	88	13	volume	volume	NOUN
app01-9407	88	14	extracted	extract	VERB
app01-9407	88	15	from	from	ADP
app01-9407	88	16	sample	sample	NOUN
app01-9407	88	17	2	2	NUM
app01-9407	88	18	.	.	PUNCT
app01-9407	89	1	(	(	PUNCT
app01-9407	89	2	a	a	X
app01-9407	89	3	)	)	PUNCT
app01-9407	89	4	displays	display	VERB
app01-9407	89	5	the	the	DET
app01-9407	89	6	superimposed	superimpose	VERB
app01-9407	89	7	predictions	prediction	NOUN
app01-9407	89	8	(	(	PUNCT
app01-9407	89	9	yellow	yellow	ADJ
app01-9407	89	10	)	)	PUNCT
app01-9407	89	11	on	on	ADP
app01-9407	89	12	the	the	DET
app01-9407	89	13	input	input	NOUN
app01-9407	89	14	volume	volume	NOUN
app01-9407	89	15	and	and	CCONJ
app01-9407	89	16	the	the	DET
app01-9407	89	17	otsu	otsu	NOUN
app01-9407	89	18	thresholding	thresholding	NOUN
app01-9407	89	19	(	(	PUNCT
app01-9407	89	20	red	red	ADJ
app01-9407	89	21	)	)	PUNCT
app01-9407	89	22	in	in	ADP
app01-9407	89	23	a	a	DET
app01-9407	89	24	3d	3d	NUM
app01-9407	89	25	visualization	visualization	NOUN
app01-9407	89	26	.	.	PUNCT
app01-9407	90	1	(	(	PUNCT
app01-9407	90	2	b	b	NOUN
app01-9407	90	3	)	)	PUNCT
app01-9407	90	4	,	,	PUNCT
app01-9407	90	5	(	(	PUNCT
app01-9407	90	6	c	c	X
app01-9407	90	7	)	)	PUNCT
app01-9407	90	8	and	and	CCONJ
app01-9407	90	9	(	(	PUNCT
app01-9407	90	10	d	d	X
app01-9407	90	11	)	)	PUNCT
app01-9407	90	12	respectively	respectively	ADV
app01-9407	90	13	display	display	VERB
app01-9407	90	14	axis	axis	NOUN
app01-9407	90	15	-	-	PUNCT
app01-9407	90	16	aligned	align	VERB
app01-9407	90	17	2d	2d	NOUN
app01-9407	90	18	slices	slice	NOUN
app01-9407	90	19	of	of	ADP
app01-9407	90	20	predictions	prediction	NOUN
app01-9407	90	21	(	(	PUNCT
app01-9407	90	22	yellow	yellow	NOUN
app01-9407	90	23	)	)	PUNCT
app01-9407	90	24	overlaid	overlay	VERB
app01-9407	90	25	on	on	ADP
app01-9407	90	26	the	the	DET
app01-9407	90	27	input	input	NOUN
app01-9407	90	28	volume	volume	NOUN
app01-9407	90	29	and	and	CCONJ
app01-9407	90	30	otsu	otsu	NOUN
app01-9407	90	31	thresholding	thresholding	NOUN
app01-9407	90	32	(	(	PUNCT
app01-9407	90	33	red	red	ADJ
app01-9407	90	34	)	)	PUNCT
app01-9407	90	35	.	.	PUNCT
app01-9407	91	1	orange	orange	PROPN
app01-9407	91	2	colour	colour	PROPN
app01-9407	91	3	represents	represent	VERB
app01-9407	91	4	the	the	DET
app01-9407	91	5	agreement	agreement	NOUN
app01-9407	91	6	between	between	ADP
app01-9407	91	7	the	the	DET
app01-9407	91	8	prediction	prediction	NOUN
app01-9407	91	9	and	and	CCONJ
app01-9407	91	10	otsu	otsu	ADJ
app01-9407	91	11	thresholding	thresholding	NOUN
app01-9407	91	12	regions	region	NOUN
app01-9407	91	13	.	.	PUNCT
app01-9407	92	1	additionally	additionally	ADV
app01-9407	92	2	,	,	PUNCT
app01-9407	92	3	the	the	DET
app01-9407	92	4	detail	detail	NOUN
app01-9407	92	5	images	image	NOUN
app01-9407	92	6	provide	provide	VERB
app01-9407	92	7	a	a	DET
app01-9407	92	8	closer	close	ADJ
app01-9407	92	9	look	look	NOUN
app01-9407	92	10	at	at	ADP
app01-9407	92	11	a	a	DET
app01-9407	92	12	specific	specific	ADJ
app01-9407	92	13	region	region	NOUN
app01-9407	92	14	marked	mark	VERB
app01-9407	92	15	in	in	ADP
app01-9407	92	16	blue	blue	ADJ
app01-9407	92	17	,	,	PUNCT
app01-9407	92	18	magnified	magnify	VERB
app01-9407	92	19	four	four	NUM
app01-9407	92	20	times	time	NOUN
app01-9407	92	21	.	.	PUNCT
app01-9407	93	1	function	function	NOUN
app01-9407	93	2	was	be	AUX
app01-9407	93	3	employed	employ	VERB
app01-9407	93	4	as	as	ADP
app01-9407	93	5	loss	loss	NOUN
app01-9407	93	6	function	function	NOUN
app01-9407	93	7	.	.	PUNCT
app01-9407	94	1	following	follow	VERB
app01-9407	94	2	the	the	DET
app01-9407	94	3	training	training	NOUN
app01-9407	94	4	process	process	NOUN
app01-9407	94	5	,	,	PUNCT
app01-9407	94	6	the	the	DET
app01-9407	94	7	obtained	obtain	VERB
app01-9407	94	8	results	result	NOUN
app01-9407	94	9	were	be	AUX
app01-9407	94	10	evaluated	evaluate	VERB
app01-9407	94	11	using	use	VERB
app01-9407	94	12	the	the	DET
app01-9407	94	13	test	test	NOUN
app01-9407	94	14	dataset	dataset	NOUN
app01-9407	94	15	.	.	PUNCT
app01-9407	95	1	in	in	ADP
app01-9407	95	2	the	the	DET
app01-9407	95	3	training	training	NOUN
app01-9407	95	4	process	process	NOUN
app01-9407	95	5	,	,	PUNCT
app01-9407	95	6	the	the	DET
app01-9407	95	7	input	input	NOUN
app01-9407	95	8	size	size	NOUN
app01-9407	95	9	is	be	AUX
app01-9407	95	10	fixed	fix	VERB
app01-9407	95	11	at	at	ADP
app01-9407	95	12	132	132	NUM
app01-9407	95	13	×	×	NOUN
app01-9407	95	14	132	132	NUM
app01-9407	95	15	×	×	NOUN
app01-9407	95	16	132	132	NUM
app01-9407	95	17	;	;	PUNCT
app01-9407	95	18	however	however	ADV
app01-9407	95	19	,	,	PUNCT
app01-9407	95	20	during	during	ADP
app01-9407	95	21	the	the	DET
app01-9407	95	22	testing	testing	NOUN
app01-9407	95	23	stage	stage	NOUN
app01-9407	95	24	,	,	PUNCT
app01-9407	95	25	data	datum	NOUN
app01-9407	95	26	of	of	ADP
app01-9407	95	27	any	any	DET
app01-9407	95	28	size	size	NOUN
app01-9407	95	29	can	can	AUX
app01-9407	95	30	be	be	AUX
app01-9407	95	31	applied	apply	VERB
app01-9407	95	32	.	.	PUNCT
app01-9407	96	1	in	in	ADP
app01-9407	96	2	our	our	PRON
app01-9407	96	3	case	case	NOUN
app01-9407	96	4	,	,	PUNCT
app01-9407	96	5	we	we	PRON
app01-9407	96	6	segmented	segment	VERB
app01-9407	96	7	the	the	DET
app01-9407	96	8	entire	entire	ADJ
app01-9407	96	9	xct	xct	PROPN
app01-9407	96	10	volumes	volume	NOUN
app01-9407	96	11	with	with	ADP
app01-9407	96	12	the	the	DET
app01-9407	96	13	trained	train	VERB
app01-9407	96	14	model	model	NOUN
app01-9407	96	15	,	,	PUNCT
app01-9407	96	16	as	as	ADV
app01-9407	96	17	well	well	ADV
app01-9407	96	18	as	as	ADP
app01-9407	96	19	exemplary	exemplary	ADJ
app01-9407	96	20	sub	sub	NOUN
app01-9407	96	21	-	-	NOUN
app01-9407	96	22	volumes	volume	NOUN
app01-9407	96	23	.	.	PUNCT
app01-9407	97	1	for	for	ADP
app01-9407	97	2	the	the	DET
app01-9407	97	3	prediction	prediction	NOUN
app01-9407	97	4	phase	phase	NOUN
app01-9407	97	5	,	,	PUNCT
app01-9407	97	6	a	a	DET
app01-9407	97	7	different	different	ADJ
app01-9407	97	8	system	system	NOUN
app01-9407	97	9	was	be	AUX
app01-9407	97	10	employed	employ	VERB
app01-9407	97	11	,	,	PUNCT
app01-9407	97	12	featuring	feature	VERB
app01-9407	97	13	a	a	DET
app01-9407	97	14	consumer	consumer	NOUN
app01-9407	97	15	-	-	PUNCT
app01-9407	97	16	grade	grade	NOUN
app01-9407	97	17	gpu	gpu	NOUN
app01-9407	97	18	(	(	PUNCT
app01-9407	97	19	nvidia	nvidia	PROPN
app01-9407	97	20	geforce	geforce	NOUN
app01-9407	97	21	gtx	gtx	PROPN
app01-9407	97	22	1080	1080	NUM
app01-9407	97	23	)	)	PUNCT
app01-9407	97	24	.	.	PUNCT
app01-9407	98	1	predicting	predict	VERB
app01-9407	98	2	the	the	DET
app01-9407	98	3	sample	sample	NOUN
app01-9407	98	4	2	2	NUM
app01-9407	98	5	(	(	PUNCT
app01-9407	98	6	1300×	1300×	NUM
app01-9407	98	7	900	900	NUM
app01-9407	98	8	×	×	NOUN
app01-9407	98	9	976	976	NUM
app01-9407	98	10	)	)	PUNCT
app01-9407	98	11	using	use	VERB
app01-9407	98	12	open_ia	open_ia	NOUN
app01-9407	98	13	and	and	CCONJ
app01-9407	98	14	the	the	DET
app01-9407	98	15	onnx	onnx	PROPN
app01-9407	98	16	runtime	runtime	NOUN
app01-9407	98	17	took	take	VERB
app01-9407	98	18	approximately	approximately	ADV
app01-9407	98	19	four	four	NUM
app01-9407	98	20	minutes	minute	NOUN
app01-9407	98	21	.	.	PUNCT
app01-9407	99	1	this	this	DET
app01-9407	99	2	time	time	NOUN
app01-9407	99	3	includes	include	VERB
app01-9407	99	4	the	the	DET
app01-9407	99	5	normalization	normalization	NOUN
app01-9407	99	6	of	of	ADP
app01-9407	99	7	the	the	DET
app01-9407	99	8	input	input	NOUN
app01-9407	99	9	,	,	PUNCT
app01-9407	99	10	splitting	split	VERB
app01-9407	99	11	it	it	PRON
app01-9407	99	12	into	into	ADP
app01-9407	99	13	subvolumes	subvolume	NOUN
app01-9407	99	14	,	,	PUNCT
app01-9407	99	15	performing	perform	VERB
app01-9407	99	16	the	the	DET
app01-9407	99	17	prediction	prediction	NOUN
app01-9407	99	18	,	,	PUNCT
app01-9407	99	19	and	and	CCONJ
app01-9407	99	20	merging	merge	VERB
app01-9407	99	21	the	the	DET
app01-9407	99	22	sub	sub	NOUN
app01-9407	99	23	-	-	NOUN
app01-9407	99	24	volumes	volume	NOUN
app01-9407	99	25	into	into	ADP
app01-9407	99	26	the	the	DET
app01-9407	99	27	final	final	ADJ
app01-9407	99	28	result	result	NOUN
app01-9407	99	29	.	.	PUNCT
app01-9407	100	1	5	5	X
app01-9407	100	2	.	.	NOUN
app01-9407	100	3	results	result	NOUN
app01-9407	100	4	and	and	CCONJ
app01-9407	100	5	discussion	discussion	NOUN
app01-9407	100	6	in	in	ADP
app01-9407	100	7	this	this	DET
app01-9407	100	8	section	section	NOUN
app01-9407	100	9	,	,	PUNCT
app01-9407	100	10	we	we	PRON
app01-9407	100	11	present	present	VERB
app01-9407	100	12	some	some	PRON
app01-9407	100	13	of	of	ADP
app01-9407	100	14	the	the	DET
app01-9407	100	15	results	result	NOUN
app01-9407	100	16	of	of	ADP
app01-9407	100	17	the	the	DET
app01-9407	100	18	predictions	prediction	NOUN
app01-9407	100	19	in	in	ADP
app01-9407	100	20	comparison	comparison	NOUN
app01-9407	100	21	to	to	ADP
app01-9407	100	22	otsu	otsu	NOUN
app01-9407	100	23	thresholding	thresholding	NOUN
app01-9407	100	24	.	.	PUNCT
app01-9407	101	1	we	we	PRON
app01-9407	101	2	measure	measure	VERB
app01-9407	101	3	the	the	DET
app01-9407	101	4	segmentation	segmentation	NOUN
app01-9407	101	5	accuracy	accuracy	NOUN
app01-9407	101	6	using	use	VERB
app01-9407	101	7	the	the	DET
app01-9407	101	8	dice	dice	NOUN
app01-9407	101	9	coefficient	coefficient	NOUN
app01-9407	101	10	function	function	NOUN
app01-9407	101	11	[	[	X
app01-9407	101	12	26	26	NUM
app01-9407	101	13	]	]	PUNCT
app01-9407	101	14	.	.	PUNCT
app01-9407	102	1	this	this	DET
app01-9407	102	2	function	function	NOUN
app01-9407	102	3	computes	compute	VERB
app01-9407	102	4	the	the	DET
app01-9407	102	5	dice	dice	NOUN
app01-9407	102	6	similarity	similarity	NOUN
app01-9407	102	7	coefficient	coefficient	NOUN
app01-9407	102	8	between	between	ADP
app01-9407	102	9	the	the	DET
app01-9407	102	10	prediction	prediction	NOUN
app01-9407	102	11	and	and	CCONJ
app01-9407	102	12	reference	reference	NOUN
app01-9407	102	13	segmentations	segmentation	NOUN
app01-9407	102	14	.	.	PUNCT
app01-9407	103	1	training	training	NOUN
app01-9407	103	2	and	and	CCONJ
app01-9407	103	3	validation	validation	NOUN
app01-9407	103	4	accuracy	accuracy	NOUN
app01-9407	103	5	were	be	AUX
app01-9407	103	6	both	both	PRON
app01-9407	103	7	determined	determine	VERB
app01-9407	103	8	at	at	ADP
app01-9407	103	9	appr	appr	PROPN
app01-9407	103	10	.	.	PROPN
app01-9407	104	1	99	99	NUM
app01-9407	104	2	%	%	NOUN
app01-9407	104	3	.	.	PUNCT
app01-9407	105	1	on	on	ADP
app01-9407	105	2	the	the	DET
app01-9407	105	3	training	training	NOUN
app01-9407	105	4	process	process	NOUN
app01-9407	105	5	,	,	PUNCT
app01-9407	105	6	those	those	DET
app01-9407	105	7	values	value	NOUN
app01-9407	105	8	were	be	AUX
app01-9407	105	9	continuously	continuously	ADV
app01-9407	105	10	growing	grow	VERB
app01-9407	105	11	,	,	PUNCT
app01-9407	105	12	and	and	CCONJ
app01-9407	105	13	the	the	DET
app01-9407	105	14	final	final	ADJ
app01-9407	105	15	model	model	NOUN
app01-9407	105	16	achieved	achieve	VERB
app01-9407	105	17	satisfactory	satisfactory	ADJ
app01-9407	105	18	results	result	NOUN
app01-9407	105	19	.	.	PUNCT
app01-9407	106	1	the	the	DET
app01-9407	106	2	result	result	NOUN
app01-9407	106	3	of	of	ADP
app01-9407	106	4	the	the	DET
app01-9407	106	5	evaluation	evaluation	NOUN
app01-9407	106	6	with	with	ADP
app01-9407	106	7	the	the	DET
app01-9407	106	8	test	test	NOUN
app01-9407	106	9	dataset	dataset	NOUN
app01-9407	106	10	was	be	AUX
app01-9407	106	11	a	a	DET
app01-9407	106	12	dice	dice	NOUN
app01-9407	106	13	coefficient	coefficient	NOUN
app01-9407	106	14	of	of	ADP
app01-9407	106	15	0.9822	0.9822	NUM
app01-9407	106	16	and	and	CCONJ
app01-9407	106	17	an	an	DET
app01-9407	106	18	accuracy	accuracy	NOUN
app01-9407	106	19	of	of	ADP
app01-9407	106	20	0.9990	0.9990	NUM
app01-9407	106	21	.	.	PUNCT
app01-9407	107	1	the	the	DET
app01-9407	107	2	grey	grey	PROPN
app01-9407	107	3	values	value	NOUN
app01-9407	107	4	were	be	AUX
app01-9407	107	5	normalized	normalize	VERB
app01-9407	107	6	to	to	PART
app01-9407	107	7	be	be	AUX
app01-9407	107	8	in	in	ADP
app01-9407	107	9	the	the	DET
app01-9407	107	10	range	range	NOUN
app01-9407	107	11	from	from	ADP
app01-9407	107	12	zero	zero	NUM
app01-9407	107	13	to	to	ADP
app01-9407	107	14	one	one	NUM
app01-9407	107	15	for	for	ADP
app01-9407	107	16	all	all	PRON
app01-9407	107	17	of	of	ADP
app01-9407	107	18	the	the	DET
app01-9407	107	19	samples	sample	NOUN
app01-9407	107	20	before	before	ADP
app01-9407	107	21	the	the	DET
app01-9407	107	22	training	training	NOUN
app01-9407	107	23	and	and	CCONJ
app01-9407	107	24	testing	testing	NOUN
app01-9407	107	25	process	process	NOUN
app01-9407	107	26	.	.	PUNCT
app01-9407	108	1	otsu	otsu	ADJ
app01-9407	108	2	thresholding	thresholde	VERB
app01-9407	108	3	segmentations	segmentation	NOUN
app01-9407	108	4	(	(	PUNCT
app01-9407	108	5	labels	label	NOUN
app01-9407	108	6	)	)	PUNCT
app01-9407	108	7	are	be	AUX
app01-9407	108	8	binarized	binarize	VERB
app01-9407	108	9	to	to	ADP
app01-9407	108	10	zero	zero	NUM
app01-9407	108	11	or	or	CCONJ
app01-9407	108	12	one	one	NUM
app01-9407	108	13	(	(	PUNCT
app01-9407	108	14	white	white	ADJ
app01-9407	108	15	values	value	NOUN
app01-9407	108	16	=	=	SYM
app01-9407	108	17	1	1	NUM
app01-9407	108	18	,	,	PUNCT
app01-9407	108	19	black	black	ADJ
app01-9407	108	20	values	value	NOUN
app01-9407	108	21	=	=	SYM
app01-9407	108	22	0	0	X
app01-9407	108	23	)	)	PUNCT
app01-9407	108	24	for	for	ADP
app01-9407	108	25	all	all	DET
app01-9407	108	26	the	the	DET
app01-9407	108	27	samples	sample	NOUN
app01-9407	108	28	.	.	PUNCT
app01-9407	109	1	for	for	ADP
app01-9407	109	2	each	each	DET
app01-9407	109	3	voxel	voxel	PROPN
app01-9407	109	4	,	,	PUNCT
app01-9407	109	5	the	the	DET
app01-9407	109	6	prediction	prediction	NOUN
app01-9407	109	7	delivers	deliver	VERB
app01-9407	109	8	a	a	DET
app01-9407	109	9	value	value	NOUN
app01-9407	109	10	between	between	ADP
app01-9407	109	11	0	0	NUM
app01-9407	109	12	and	and	CCONJ
app01-9407	109	13	1	1	NUM
app01-9407	109	14	,	,	PUNCT
app01-9407	109	15	denoting	denote	VERB
app01-9407	109	16	the	the	DET
app01-9407	109	17	probability	probability	NOUN
app01-9407	109	18	for	for	SCONJ
app01-9407	109	19	this	this	DET
app01-9407	109	20	voxel	voxel	PROPN
app01-9407	109	21	to	to	PART
app01-9407	109	22	be	be	AUX
app01-9407	109	23	a	a	DET
app01-9407	109	24	pore	pore	NOUN
app01-9407	109	25	.	.	PUNCT
app01-9407	110	1	we	we	PRON
app01-9407	110	2	have	have	AUX
app01-9407	110	3	used	use	VERB
app01-9407	110	4	our	our	PRON
app01-9407	110	5	prediction	prediction	NOUN
app01-9407	110	6	without	without	ADP
app01-9407	110	7	any	any	DET
app01-9407	110	8	postprocessing	postprocessing	NOUN
app01-9407	110	9	to	to	PART
app01-9407	110	10	show	show	VERB
app01-9407	110	11	the	the	DET
app01-9407	110	12	probability	probability	NOUN
app01-9407	110	13	of	of	ADP
app01-9407	110	14	the	the	DET
app01-9407	110	15	predicted	predict	VERB
app01-9407	110	16	voxels	voxel	NOUN
app01-9407	110	17	.	.	PUNCT
app01-9407	111	1	5.1	5.1	NUM
app01-9407	111	2	.	.	PUNCT
app01-9407	112	1	sub	sub	ADJ
app01-9407	112	2	-	-	ADJ
app01-9407	112	3	volume	volume	ADJ
app01-9407	112	4	prediction	prediction	NOUN
app01-9407	112	5	and	and	CCONJ
app01-9407	112	6	visualization	visualization	NOUN
app01-9407	112	7	the	the	DET
app01-9407	112	8	neural	neural	ADJ
app01-9407	112	9	network	network	NOUN
app01-9407	112	10	model	model	NOUN
app01-9407	112	11	in	in	ADP
app01-9407	112	12	open_ia	open_ia	PROPN
app01-9407	112	13	was	be	AUX
app01-9407	112	14	utilized	utilize	VERB
app01-9407	112	15	to	to	PART
app01-9407	112	16	perform	perform	VERB
app01-9407	112	17	segmentation	segmentation	NOUN
app01-9407	112	18	on	on	ADP
app01-9407	112	19	a	a	DET
app01-9407	112	20	sub	sub	NOUN
app01-9407	112	21	-	-	NOUN
app01-9407	112	22	volume	volume	NOUN
app01-9407	112	23	extracted	extract	VERB
app01-9407	112	24	from	from	ADP
app01-9407	112	25	sample	sample	NOUN
app01-9407	112	26	2	2	NUM
app01-9407	112	27	,	,	PUNCT
app01-9407	112	28	which	which	PRON
app01-9407	112	29	had	have	VERB
app01-9407	112	30	dimensions	dimension	NOUN
app01-9407	112	31	of	of	ADP
app01-9407	112	32	122	122	NUM
app01-9407	112	33	×	×	NOUN
app01-9407	112	34	122	122	NUM
app01-9407	112	35	×	×	NOUN
app01-9407	112	36	122	122	NUM
app01-9407	112	37	.	.	PUNCT
app01-9407	113	1	the	the	DET
app01-9407	113	2	obtained	obtain	VERB
app01-9407	113	3	segmentation	segmentation	NOUN
app01-9407	113	4	results	result	NOUN
app01-9407	113	5	are	be	AUX
app01-9407	113	6	presented	present	VERB
app01-9407	113	7	in	in	ADP
app01-9407	113	8	figure	figure	NOUN
app01-9407	113	9	3	3	NUM
app01-9407	113	10	.	.	PUNCT
app01-9407	114	1	this	this	DET
app01-9407	114	2	subsection	subsection	NOUN
app01-9407	114	3	of	of	ADP
app01-9407	114	4	the	the	DET
app01-9407	114	5	dataset	dataset	NOUN
app01-9407	114	6	has	have	AUX
app01-9407	114	7	been	be	AUX
app01-9407	114	8	chosen	choose	VERB
app01-9407	114	9	since	since	SCONJ
app01-9407	114	10	it	it	PRON
app01-9407	114	11	contains	contain	VERB
app01-9407	114	12	many	many	ADJ
app01-9407	114	13	different	different	ADJ
app01-9407	114	14	shapes	shape	NOUN
app01-9407	114	15	of	of	ADP
app01-9407	114	16	pores	pore	NOUN
app01-9407	114	17	.	.	PUNCT
app01-9407	115	1	according	accord	VERB
app01-9407	115	2	to	to	ADP
app01-9407	115	3	the	the	DET
app01-9407	115	4	predictions	prediction	NOUN
app01-9407	115	5	,	,	PUNCT
app01-9407	115	6	all	all	DET
app01-9407	115	7	kind	kind	ADV
app01-9407	115	8	of	of	ADP
app01-9407	115	9	pores	pore	NOUN
app01-9407	115	10	with	with	ADP
app01-9407	115	11	different	different	ADJ
app01-9407	115	12	shapes	shape	NOUN
app01-9407	115	13	are	be	AUX
app01-9407	115	14	segmented	segment	VERB
app01-9407	115	15	both	both	CCONJ
app01-9407	115	16	with	with	ADP
app01-9407	115	17	otsu	otsu	NOUN
app01-9407	115	18	thresholding	thresholding	NOUN
app01-9407	115	19	and	and	CCONJ
app01-9407	115	20	with	with	ADP
app01-9407	115	21	our	our	PRON
app01-9407	115	22	modified	modify	VERB
app01-9407	115	23	3d	3d	NUM
app01-9407	115	24	u	u	ADJ
app01-9407	115	25	-	-	ADJ
app01-9407	115	26	net	net	ADJ
app01-9407	115	27	approach	approach	NOUN
app01-9407	115	28	.	.	PUNCT
app01-9407	116	1	for	for	ADP
app01-9407	116	2	bigger	big	ADJ
app01-9407	116	3	pores	pore	NOUN
app01-9407	116	4	,	,	PUNCT
app01-9407	116	5	the	the	DET
app01-9407	116	6	overall	overall	ADJ
app01-9407	116	7	results	result	NOUN
app01-9407	116	8	seem	seem	VERB
app01-9407	116	9	quite	quite	ADV
app01-9407	116	10	similar	similar	ADJ
app01-9407	116	11	with	with	ADP
app01-9407	116	12	both	both	DET
app01-9407	116	13	methods	method	NOUN
app01-9407	116	14	.	.	PUNCT
app01-9407	117	1	in	in	ADP
app01-9407	117	2	the	the	DET
app01-9407	117	3	magnifications	magnification	NOUN
app01-9407	117	4	,	,	PUNCT
app01-9407	117	5	it	it	PRON
app01-9407	117	6	becomes	become	VERB
app01-9407	117	7	obvious	obvious	ADJ
app01-9407	117	8	that	that	SCONJ
app01-9407	117	9	the	the	DET
app01-9407	117	10	otsu	otsu	NOUN
app01-9407	117	11	thresholding	thresholde	VERB
app01-9407	117	12	segmentations	segmentation	NOUN
app01-9407	117	13	are	be	AUX
app01-9407	117	14	90	90	NUM
app01-9407	117	15	vol	vol	NOUN
app01-9407	117	16	.	.	PUNCT
app01-9407	118	1	42/2023	42/2023	NUM
app01-9407	118	2	cfrp	cfrp	PROPN
app01-9407	118	3	pore	pore	ADJ
app01-9407	118	4	segmentation	segmentation	NOUN
app01-9407	118	5	via	via	ADP
app01-9407	118	6	u	u	ADJ
app01-9407	118	7	-	-	ADJ
app01-9407	118	8	net	net	ADJ
app01-9407	118	9	figure	figure	NOUN
app01-9407	118	10	4	4	NUM
app01-9407	118	11	.	.	PUNCT
app01-9407	118	12	original	original	ADJ
app01-9407	118	13	input	input	NOUN
app01-9407	118	14	data	datum	NOUN
app01-9407	118	15	slice	slice	NOUN
app01-9407	118	16	from	from	ADP
app01-9407	118	17	sample	sample	NOUN
app01-9407	118	18	2	2	NUM
app01-9407	118	19	(	(	PUNCT
app01-9407	118	20	a	a	NOUN
app01-9407	118	21	)	)	PUNCT
app01-9407	118	22	,	,	PUNCT
app01-9407	118	23	respectively	respectively	ADV
app01-9407	118	24	show	show	VERB
app01-9407	118	25	2d	2d	NUM
app01-9407	118	26	slices	slice	NOUN
app01-9407	118	27	of	of	ADP
app01-9407	118	28	otsu	otsu	NOUN
app01-9407	118	29	thresholding	thresholding	NOUN
app01-9407	118	30	(	(	PUNCT
app01-9407	118	31	red	red	ADJ
app01-9407	118	32	)	)	PUNCT
app01-9407	118	33	(	(	PUNCT
app01-9407	118	34	b	b	NOUN
app01-9407	118	35	)	)	PUNCT
app01-9407	118	36	and	and	CCONJ
app01-9407	118	37	prediction	prediction	NOUN
app01-9407	118	38	(	(	PUNCT
app01-9407	118	39	yellow	yellow	ADJ
app01-9407	118	40	)	)	PUNCT
app01-9407	118	41	(	(	PUNCT
app01-9407	118	42	c	c	X
app01-9407	118	43	)	)	PUNCT
app01-9407	118	44	overlaid	overlay	VERB
app01-9407	118	45	on	on	ADP
app01-9407	118	46	an	an	DET
app01-9407	118	47	input	input	NOUN
app01-9407	118	48	volume	volume	NOUN
app01-9407	118	49	slice	slice	NOUN
app01-9407	118	50	.	.	PUNCT
app01-9407	119	1	figure	figure	NOUN
app01-9407	119	2	5	5	NUM
app01-9407	119	3	.	.	PUNCT
app01-9407	119	4	zoomed	zoomed	PROPN
app01-9407	119	5	version	version	PROPN
app01-9407	119	6	of	of	ADP
app01-9407	119	7	region	region	NOUN
app01-9407	119	8	1	1	NUM
app01-9407	119	9	and	and	CCONJ
app01-9407	119	10	2	2	NUM
app01-9407	119	11	:	:	PUNCT
app01-9407	119	12	input	input	NOUN
app01-9407	119	13	data	data	NOUN
app01-9407	119	14	slice	slice	NOUN
app01-9407	119	15	(	(	PUNCT
app01-9407	119	16	a	a	NOUN
app01-9407	119	17	)	)	PUNCT
app01-9407	119	18	,	,	PUNCT
app01-9407	119	19	otsu	otsu	NOUN
app01-9407	119	20	thresholding	thresholding	NOUN
app01-9407	119	21	(	(	PUNCT
app01-9407	119	22	red	red	ADJ
app01-9407	119	23	)	)	PUNCT
app01-9407	119	24	segmentation	segmentation	NOUN
app01-9407	119	25	results	result	NOUN
app01-9407	119	26	overlaid	overlay	VERB
app01-9407	119	27	on	on	ADP
app01-9407	119	28	input	input	NOUN
app01-9407	119	29	data	datum	NOUN
app01-9407	119	30	slice	slice	NOUN
app01-9407	119	31	(	(	PUNCT
app01-9407	119	32	b	b	NOUN
app01-9407	119	33	)	)	PUNCT
app01-9407	119	34	,	,	PUNCT
app01-9407	119	35	prediction	prediction	NOUN
app01-9407	119	36	(	(	PUNCT
app01-9407	119	37	yellow	yellow	NOUN
app01-9407	119	38	)	)	PUNCT
app01-9407	119	39	overlaid	overlay	VERB
app01-9407	119	40	on	on	ADP
app01-9407	119	41	input	input	NOUN
app01-9407	119	42	data	datum	NOUN
app01-9407	119	43	slice	slice	NOUN
app01-9407	119	44	(	(	PUNCT
app01-9407	119	45	c	c	NOUN
app01-9407	119	46	)	)	PUNCT
app01-9407	119	47	.	.	PUNCT
app01-9407	120	1	the	the	DET
app01-9407	120	2	detail	detail	NOUN
app01-9407	120	3	images	image	NOUN
app01-9407	120	4	showcase	showcase	VERB
app01-9407	120	5	a	a	DET
app01-9407	120	6	closer	close	ADJ
app01-9407	120	7	view	view	NOUN
app01-9407	120	8	of	of	ADP
app01-9407	120	9	the	the	DET
app01-9407	120	10	edge	edge	NOUN
app01-9407	120	11	region	region	NOUN
app01-9407	120	12	marked	mark	VERB
app01-9407	120	13	in	in	ADP
app01-9407	120	14	blue	blue	ADJ
app01-9407	120	15	,	,	PUNCT
app01-9407	120	16	magnified	magnify	VERB
app01-9407	120	17	12	12	NUM
app01-9407	120	18	times	time	NOUN
app01-9407	120	19	and	and	CCONJ
app01-9407	120	20	4	4	NUM
app01-9407	120	21	times	time	NOUN
app01-9407	120	22	respectively	respectively	ADV
app01-9407	120	23	.	.	PUNCT
app01-9407	121	1	under	under	ADV
app01-9407	121	2	-	-	PUNCT
app01-9407	121	3	segmented	segment	VERB
app01-9407	121	4	,	,	PUNCT
app01-9407	121	5	especially	especially	ADV
app01-9407	121	6	with	with	ADP
app01-9407	121	7	long	long	ADJ
app01-9407	121	8	and	and	CCONJ
app01-9407	121	9	thin	thin	ADJ
app01-9407	121	10	(	(	PUNCT
app01-9407	121	11	figure	figure	NOUN
app01-9407	121	12	3c	3c	NUM
app01-9407	121	13	pores	pore	NOUN
app01-9407	121	14	and	and	CCONJ
app01-9407	121	15	small	small	ADJ
app01-9407	121	16	pores	pore	NOUN
app01-9407	121	17	(	(	PUNCT
app01-9407	121	18	see	see	VERB
app01-9407	121	19	figure	figure	NOUN
app01-9407	121	20	3b	3b	NOUN
app01-9407	121	21	.	.	PUNCT
app01-9407	122	1	the	the	DET
app01-9407	122	2	neural	neural	ADJ
app01-9407	122	3	network	network	NOUN
app01-9407	122	4	predictions	prediction	NOUN
app01-9407	122	5	are	be	AUX
app01-9407	122	6	performing	perform	VERB
app01-9407	122	7	better	well	ADV
app01-9407	122	8	than	than	ADP
app01-9407	122	9	otsu	otsu	ADJ
app01-9407	122	10	thresholding	thresholding	NOUN
app01-9407	122	11	as	as	SCONJ
app01-9407	122	12	seen	see	VERB
app01-9407	122	13	by	by	ADP
app01-9407	122	14	visual	visual	ADJ
app01-9407	122	15	inspection	inspection	NOUN
app01-9407	122	16	.	.	PUNCT
app01-9407	123	1	3d	3d	NUM
app01-9407	123	2	visualization	visualization	NOUN
app01-9407	123	3	of	of	ADP
app01-9407	123	4	the	the	DET
app01-9407	123	5	sub	sub	NOUN
app01-9407	123	6	-	-	NOUN
app01-9407	123	7	volume	volume	NOUN
app01-9407	123	8	and	and	CCONJ
app01-9407	123	9	its	its	PRON
app01-9407	123	10	prediction	prediction	NOUN
app01-9407	123	11	with	with	ADP
app01-9407	123	12	overlaid	overlaid	ADJ
app01-9407	123	13	otsu	otsu	NOUN
app01-9407	123	14	thresholding	thresholding	NOUN
app01-9407	123	15	is	be	AUX
app01-9407	123	16	shown	show	VERB
app01-9407	123	17	in	in	ADP
app01-9407	123	18	figure	figure	NOUN
app01-9407	123	19	3a	3a	NUM
app01-9407	123	20	.	.	PUNCT
app01-9407	124	1	differences	difference	NOUN
app01-9407	124	2	between	between	ADP
app01-9407	124	3	otsu	otsu	NOUN
app01-9407	124	4	thresholding	thresholding	NOUN
app01-9407	124	5	and	and	CCONJ
app01-9407	124	6	predictions	prediction	NOUN
app01-9407	124	7	are	be	AUX
app01-9407	124	8	clearly	clearly	ADV
app01-9407	124	9	visible	visible	ADJ
app01-9407	124	10	in	in	ADP
app01-9407	124	11	the	the	DET
app01-9407	124	12	zoomed	zoomed	PROPN
app01-9407	124	13	region	region	NOUN
app01-9407	124	14	in	in	ADP
app01-9407	124	15	the	the	DET
app01-9407	124	16	3d	3d	NUM
app01-9407	124	17	image	image	NOUN
app01-9407	124	18	.	.	PUNCT
app01-9407	125	1	the	the	DET
app01-9407	125	2	predicted	predict	VERB
app01-9407	125	3	area	area	NOUN
app01-9407	125	4	covers	cover	VERB
app01-9407	125	5	most	most	ADJ
app01-9407	125	6	of	of	ADP
app01-9407	125	7	the	the	DET
app01-9407	125	8	matching	match	VERB
app01-9407	125	9	otsu	otsu	NOUN
app01-9407	125	10	thresholding	thresholding	NOUN
app01-9407	125	11	region	region	NOUN
app01-9407	125	12	.	.	PUNCT
app01-9407	126	1	according	accord	VERB
app01-9407	126	2	to	to	ADP
app01-9407	126	3	the	the	DET
app01-9407	126	4	2d	2d	NOUN
app01-9407	126	5	and	and	CCONJ
app01-9407	126	6	3d	3d	NUM
app01-9407	126	7	visualizations	visualization	NOUN
app01-9407	126	8	of	of	ADP
app01-9407	126	9	the	the	DET
app01-9407	126	10	sub	sub	NOUN
app01-9407	126	11	-	-	NOUN
app01-9407	126	12	volume	volume	NOUN
app01-9407	126	13	shown	show	VERB
app01-9407	126	14	in	in	ADP
app01-9407	126	15	figure	figure	NOUN
app01-9407	126	16	3	3	NUM
app01-9407	126	17	,	,	PUNCT
app01-9407	126	18	both	both	PRON
app01-9407	126	19	show	show	VERB
app01-9407	126	20	that	that	SCONJ
app01-9407	126	21	the	the	DET
app01-9407	126	22	results	result	NOUN
app01-9407	126	23	from	from	ADP
app01-9407	126	24	otsu	otsu	ADJ
app01-9407	126	25	thresholding	thresholding	NOUN
app01-9407	126	26	are	be	AUX
app01-9407	126	27	under	under	ADV
app01-9407	126	28	-	-	PUNCT
app01-9407	126	29	segmented	segment	VERB
app01-9407	126	30	when	when	SCONJ
app01-9407	126	31	compared	compare	VERB
app01-9407	126	32	to	to	ADP
app01-9407	126	33	predictions	prediction	NOUN
app01-9407	126	34	of	of	ADP
app01-9407	126	35	the	the	DET
app01-9407	126	36	neural	neural	ADJ
app01-9407	126	37	network	network	NOUN
app01-9407	126	38	model	model	NOUN
app01-9407	126	39	.	.	PUNCT
app01-9407	127	1	5.2	5.2	NUM
app01-9407	127	2	.	.	PUNCT
app01-9407	127	3	prediction	prediction	NOUN
app01-9407	127	4	and	and	CCONJ
app01-9407	127	5	visualization	visualization	NOUN
app01-9407	127	6	of	of	ADP
app01-9407	127	7	sample	sample	NOUN
app01-9407	127	8	2	2	NUM
app01-9407	127	9	this	this	DET
app01-9407	127	10	section	section	NOUN
app01-9407	127	11	presents	present	VERB
app01-9407	127	12	the	the	DET
app01-9407	127	13	prediction	prediction	NOUN
app01-9407	127	14	outcome	outcome	NOUN
app01-9407	127	15	for	for	ADP
app01-9407	127	16	sample	sample	NOUN
app01-9407	127	17	2	2	NUM
app01-9407	127	18	(	(	PUNCT
app01-9407	127	19	see	see	VERB
app01-9407	127	20	figure	figure	NOUN
app01-9407	127	21	4	4	NUM
app01-9407	127	22	)	)	PUNCT
app01-9407	127	23	.	.	PUNCT
app01-9407	128	1	the	the	DET
app01-9407	128	2	goal	goal	NOUN
app01-9407	128	3	is	be	AUX
app01-9407	128	4	here	here	ADV
app01-9407	128	5	to	to	PART
app01-9407	128	6	show	show	VERB
app01-9407	128	7	that	that	SCONJ
app01-9407	128	8	the	the	DET
app01-9407	128	9	model	model	NOUN
app01-9407	128	10	is	be	AUX
app01-9407	128	11	also	also	ADV
app01-9407	128	12	able	able	ADJ
app01-9407	128	13	to	to	PART
app01-9407	128	14	effectively	effectively	ADV
app01-9407	128	15	segment	segment	VERB
app01-9407	128	16	unseen	unseen	ADJ
app01-9407	128	17	data	datum	NOUN
app01-9407	128	18	.	.	PUNCT
app01-9407	129	1	train	train	NOUN
app01-9407	129	2	data	datum	NOUN
app01-9407	129	3	and	and	CCONJ
app01-9407	129	4	test	test	NOUN
app01-9407	129	5	data	datum	NOUN
app01-9407	129	6	have	have	VERB
app01-9407	129	7	an	an	DET
app01-9407	129	8	isotropic	isotropic	ADJ
app01-9407	129	9	resolution	resolution	NOUN
app01-9407	129	10	of	of	ADP
app01-9407	129	11	3.3	3.3	NUM
app01-9407	129	12	µm	µm	NOUN
app01-9407	129	13	.	.	PUNCT
app01-9407	130	1	in	in	ADP
app01-9407	130	2	figure	figure	NOUN
app01-9407	130	3	4c	4c	NOUN
app01-9407	130	4	,	,	PUNCT
app01-9407	130	5	yellow	yellow	ADJ
app01-9407	130	6	colors	color	NOUN
app01-9407	130	7	(	(	PUNCT
app01-9407	130	8	prediction	prediction	NOUN
app01-9407	130	9	)	)	PUNCT
app01-9407	130	10	show	show	VERB
app01-9407	130	11	that	that	SCONJ
app01-9407	130	12	long	long	ADJ
app01-9407	130	13	thin	thin	ADJ
app01-9407	130	14	pores	pore	NOUN
app01-9407	130	15	are	be	AUX
app01-9407	130	16	segmented	segment	VERB
app01-9407	130	17	much	much	ADV
app01-9407	130	18	better	well	ADV
app01-9407	130	19	with	with	ADP
app01-9407	130	20	the	the	DET
app01-9407	130	21	neural	neural	ADJ
app01-9407	130	22	network	network	NOUN
app01-9407	130	23	.	.	PUNCT
app01-9407	131	1	otsu	otsu	ADJ
app01-9407	131	2	thresholding	thresholding	NOUN
app01-9407	131	3	is	be	AUX
app01-9407	131	4	not	not	PART
app01-9407	131	5	as	as	ADV
app01-9407	131	6	effective	effective	ADJ
app01-9407	131	7	for	for	ADP
app01-9407	131	8	this	this	DET
app01-9407	131	9	kind	kind	NOUN
app01-9407	131	10	of	of	ADP
app01-9407	131	11	pore	pore	ADJ
app01-9407	131	12	segmentation	segmentation	NOUN
app01-9407	131	13	as	as	ADP
app01-9407	131	14	the	the	DET
app01-9407	131	15	neural	neural	ADJ
app01-9407	131	16	network	network	NOUN
app01-9407	131	17	.	.	PUNCT
app01-9407	132	1	in	in	ADP
app01-9407	132	2	figure	figure	NOUN
app01-9407	132	3	4b	4b	PROPN
app01-9407	132	4	,	,	PUNCT
app01-9407	132	5	region	region	NOUN
app01-9407	132	6	2	2	NUM
app01-9407	132	7	clearly	clearly	ADV
app01-9407	132	8	shows	show	VERB
app01-9407	132	9	that	that	SCONJ
app01-9407	132	10	otsu	otsu	NOUN
app01-9407	132	11	thresholding	thresholding	NOUN
app01-9407	132	12	produces	produce	VERB
app01-9407	132	13	under	under	ADV
app01-9407	132	14	-	-	PUNCT
app01-9407	132	15	segmented	segment	VERB
app01-9407	132	16	results	result	NOUN
app01-9407	132	17	.	.	PUNCT
app01-9407	133	1	the	the	DET
app01-9407	133	2	qualitative	qualitative	ADJ
app01-9407	133	3	result	result	NOUN
app01-9407	133	4	of	of	ADP
app01-9407	133	5	prediction	prediction	NOUN
app01-9407	133	6	is	be	AUX
app01-9407	133	7	similar	similar	ADJ
app01-9407	133	8	to	to	ADP
app01-9407	133	9	the	the	DET
app01-9407	133	10	sub	sub	ADJ
app01-9407	133	11	-	-	ADJ
app01-9407	133	12	volume	volume	ADJ
app01-9407	133	13	segmentation	segmentation	NOUN
app01-9407	133	14	presented	present	VERB
app01-9407	133	15	in	in	ADP
app01-9407	133	16	section	section	NOUN
app01-9407	133	17	section	section	NOUN
app01-9407	133	18	5.1	5.1	NUM
app01-9407	133	19	.	.	PUNCT
app01-9407	134	1	sample	sample	NOUN
app01-9407	134	2	2	2	NUM
app01-9407	134	3	has	have	VERB
app01-9407	134	4	many	many	ADJ
app01-9407	134	5	different	different	ADJ
app01-9407	134	6	shapes	shape	NOUN
app01-9407	134	7	of	of	ADP
app01-9407	134	8	pores	pore	NOUN
app01-9407	134	9	such	such	ADJ
app01-9407	134	10	as	as	ADP
app01-9407	134	11	rounded	rounded	ADJ
app01-9407	134	12	pores	pore	NOUN
app01-9407	134	13	(	(	PUNCT
app01-9407	134	14	big	big	ADJ
app01-9407	134	15	and	and	CCONJ
app01-9407	134	16	small	small	ADJ
app01-9407	134	17	)	)	PUNCT
app01-9407	134	18	,	,	PUNCT
app01-9407	134	19	long	long	ADV
app01-9407	134	20	thin	thin	ADJ
app01-9407	134	21	and	and	CCONJ
app01-9407	134	22	some	some	DET
app01-9407	134	23	big	big	ADJ
app01-9407	134	24	complex	complex	ADJ
app01-9407	134	25	pores	pore	NOUN
app01-9407	134	26	.	.	PUNCT
app01-9407	135	1	segmentation	segmentation	NOUN
app01-9407	135	2	results	result	NOUN
app01-9407	135	3	differ	differ	VERB
app01-9407	135	4	substantially	substantially	ADV
app01-9407	135	5	between	between	ADP
app01-9407	135	6	the	the	DET
app01-9407	135	7	otsu	otsu	NOUN
app01-9407	135	8	technique	technique	NOUN
app01-9407	135	9	and	and	CCONJ
app01-9407	135	10	the	the	DET
app01-9407	135	11	neural	neural	ADJ
app01-9407	135	12	network	network	NOUN
app01-9407	135	13	prediction	prediction	NOUN
app01-9407	135	14	depend	depend	VERB
app01-9407	135	15	on	on	ADP
app01-9407	135	16	the	the	DET
app01-9407	135	17	shapes	shape	NOUN
app01-9407	135	18	of	of	ADP
app01-9407	135	19	the	the	DET
app01-9407	135	20	pores	pore	NOUN
app01-9407	135	21	.	.	PUNCT
app01-9407	136	1	region	region	NOUN
app01-9407	136	2	1	1	NUM
app01-9407	136	3	and	and	CCONJ
app01-9407	136	4	2	2	NUM
app01-9407	136	5	of	of	ADP
app01-9407	136	6	figure	figure	NOUN
app01-9407	136	7	4	4	NUM
app01-9407	136	8	are	be	AUX
app01-9407	136	9	shown	show	VERB
app01-9407	136	10	in	in	ADP
app01-9407	136	11	a	a	DET
app01-9407	136	12	magnified	magnify	VERB
app01-9407	136	13	version	version	NOUN
app01-9407	136	14	in	in	ADP
app01-9407	136	15	figure	figure	NOUN
app01-9407	136	16	5	5	NUM
app01-9407	136	17	.	.	PUNCT
app01-9407	137	1	the	the	DET
app01-9407	137	2	original	original	ADJ
app01-9407	137	3	slice	slice	NOUN
app01-9407	137	4	images	image	NOUN
app01-9407	137	5	are	be	AUX
app01-9407	137	6	transparently	transparently	ADV
app01-9407	137	7	overlaid	overlay	VERB
app01-9407	137	8	with	with	ADP
app01-9407	137	9	the	the	DET
app01-9407	137	10	otsu	otsu	NOUN
app01-9407	137	11	thresholding	thresholde	VERB
app01-9407	137	12	segmentation	segmentation	NOUN
app01-9407	137	13	and	and	CCONJ
app01-9407	137	14	the	the	DET
app01-9407	137	15	neural	neural	ADJ
app01-9407	137	16	network	network	NOUN
app01-9407	137	17	prediction	prediction	NOUN
app01-9407	137	18	.	.	PUNCT
app01-9407	138	1	region	region	NOUN
app01-9407	138	2	1	1	NUM
app01-9407	138	3	contains	contain	VERB
app01-9407	138	4	two	two	NUM
app01-9407	138	5	pores	pore	NOUN
app01-9407	138	6	with	with	ADP
app01-9407	138	7	different	different	ADJ
app01-9407	138	8	characteristics	characteristic	NOUN
app01-9407	138	9	:	:	PUNCT
app01-9407	138	10	one	one	NUM
app01-9407	138	11	is	be	AUX
app01-9407	138	12	clearly	clearly	ADV
app01-9407	138	13	visible	visible	ADJ
app01-9407	138	14	,	,	PUNCT
app01-9407	138	15	while	while	SCONJ
app01-9407	138	16	the	the	DET
app01-9407	138	17	other	other	ADJ
app01-9407	138	18	one	one	NOUN
app01-9407	138	19	is	be	AUX
app01-9407	138	20	only	only	ADV
app01-9407	138	21	barely	barely	ADV
app01-9407	138	22	visible	visible	ADJ
app01-9407	138	23	.	.	PUNCT
app01-9407	139	1	segmentations	segmentation	NOUN
app01-9407	139	2	are	be	AUX
app01-9407	139	3	almost	almost	ADV
app01-9407	139	4	the	the	DET
app01-9407	139	5	same	same	ADJ
app01-9407	139	6	and	and	CCONJ
app01-9407	139	7	reasonably	reasonably	ADV
app01-9407	139	8	good	good	ADJ
app01-9407	139	9	for	for	ADP
app01-9407	139	10	the	the	DET
app01-9407	139	11	clearly	clearly	ADV
app01-9407	139	12	visible	visible	ADJ
app01-9407	139	13	pore	pore	NOUN
app01-9407	139	14	in	in	ADP
app01-9407	139	15	both	both	DET
app01-9407	139	16	methods	method	NOUN
app01-9407	139	17	.	.	PUNCT
app01-9407	140	1	for	for	ADP
app01-9407	140	2	barely	barely	ADV
app01-9407	140	3	visible	visible	ADJ
app01-9407	140	4	pores	pore	NOUN
app01-9407	140	5	,	,	PUNCT
app01-9407	140	6	it	it	PRON
app01-9407	140	7	is	be	AUX
app01-9407	140	8	hard	hard	ADJ
app01-9407	140	9	to	to	PART
app01-9407	140	10	evaluate	evaluate	VERB
app01-9407	140	11	which	which	DET
app01-9407	140	12	method	method	NOUN
app01-9407	140	13	corresponds	correspond	VERB
app01-9407	140	14	better	well	ADV
app01-9407	140	15	with	with	ADP
app01-9407	140	16	reality	reality	NOUN
app01-9407	140	17	by	by	ADP
app01-9407	140	18	visual	visual	ADJ
app01-9407	140	19	inspection	inspection	NOUN
app01-9407	140	20	.	.	PUNCT
app01-9407	141	1	there	there	PRON
app01-9407	141	2	is	be	VERB
app01-9407	141	3	not	not	PART
app01-9407	141	4	a	a	DET
app01-9407	141	5	sharp	sharp	ADJ
app01-9407	141	6	edge	edge	NOUN
app01-9407	141	7	between	between	ADP
app01-9407	141	8	pore	pore	ADJ
app01-9407	141	9	and	and	CCONJ
app01-9407	141	10	background	background	NOUN
app01-9407	141	11	transitions	transition	NOUN
app01-9407	141	12	and	and	CCONJ
app01-9407	141	13	thus	thus	ADV
app01-9407	141	14	it	it	PRON
app01-9407	141	15	is	be	AUX
app01-9407	141	16	also	also	ADV
app01-9407	141	17	hard	hard	ADJ
app01-9407	141	18	to	to	PART
app01-9407	141	19	make	make	VERB
app01-9407	141	20	decisions	decision	NOUN
app01-9407	141	21	for	for	ADP
app01-9407	141	22	the	the	DET
app01-9407	141	23	network	network	NOUN
app01-9407	141	24	.	.	PUNCT
app01-9407	142	1	even	even	ADV
app01-9407	142	2	so	so	ADV
app01-9407	142	3	,	,	PUNCT
app01-9407	142	4	it	it	PRON
app01-9407	142	5	is	be	AUX
app01-9407	142	6	91	91	NUM
app01-9407	142	7	m.	m.	NOUN
app01-9407	142	8	yosifov	yosifov	NOUN
app01-9407	142	9	,	,	PUNCT
app01-9407	142	10	p.	p.	NOUN
app01-9407	142	11	weinberger	weinberger	PROPN
app01-9407	142	12	,	,	PUNCT
app01-9407	142	13	b.	b.	PROPN
app01-9407	142	14	plank	plank	PROPN
app01-9407	142	15	et	et	PROPN
app01-9407	142	16	al	al	PROPN
app01-9407	142	17	.	.	PROPN
app01-9407	142	18	acta	acta	PROPN
app01-9407	142	19	polytechnica	polytechnica	PROPN
app01-9407	142	20	ctu	ctu	PROPN
app01-9407	142	21	proceedings	proceeding	NOUN
app01-9407	142	22	figure	figure	VERB
app01-9407	142	23	6	6	NUM
app01-9407	142	24	.	.	PUNCT
app01-9407	143	1	input	input	NOUN
app01-9407	143	2	data	datum	NOUN
app01-9407	143	3	slice	slice	NOUN
app01-9407	143	4	from	from	ADP
app01-9407	143	5	sample	sample	NOUN
app01-9407	143	6	3	3	NUM
app01-9407	143	7	.	.	PUNCT
app01-9407	144	1	(	(	PUNCT
app01-9407	144	2	a	a	NOUN
app01-9407	144	3	)	)	PUNCT
app01-9407	144	4	,	,	PUNCT
app01-9407	144	5	and	and	CCONJ
app01-9407	144	6	(	(	PUNCT
app01-9407	144	7	b	b	NOUN
app01-9407	144	8	)	)	PUNCT
app01-9407	144	9	,	,	PUNCT
app01-9407	144	10	(	(	PUNCT
app01-9407	144	11	c	c	X
app01-9407	144	12	)	)	PUNCT
app01-9407	144	13	respectively	respectively	ADV
app01-9407	144	14	show	show	VERB
app01-9407	144	15	2d	2d	NUM
app01-9407	144	16	slices	slice	NOUN
app01-9407	144	17	of	of	ADP
app01-9407	144	18	otsu	otsu	NOUN
app01-9407	144	19	thresholding	thresholding	NOUN
app01-9407	144	20	and	and	CCONJ
app01-9407	144	21	prediction	prediction	NOUN
app01-9407	144	22	overlaid	overlay	VERB
app01-9407	144	23	on	on	ADP
app01-9407	144	24	an	an	DET
app01-9407	144	25	input	input	NOUN
app01-9407	144	26	volume	volume	NOUN
app01-9407	144	27	slice	slice	NOUN
app01-9407	144	28	from	from	ADP
app01-9407	144	29	sample	sample	NOUN
app01-9407	144	30	2	2	NUM
app01-9407	144	31	(	(	PUNCT
app01-9407	144	32	red=	red=	ADJ
app01-9407	144	33	otsu	otsu	NOUN
app01-9407	144	34	thresholding	thresholding	NOUN
app01-9407	144	35	,	,	PUNCT
app01-9407	144	36	yellow	yellow	ADJ
app01-9407	144	37	=	=	NOUN
app01-9407	144	38	predictions	prediction	NOUN
app01-9407	144	39	)	)	PUNCT
app01-9407	144	40	.	.	PUNCT
app01-9407	145	1	figure	figure	VERB
app01-9407	145	2	7	7	NUM
app01-9407	145	3	.	.	PUNCT
app01-9407	145	4	zoomed	zoomed	PROPN
app01-9407	145	5	version	version	PROPN
app01-9407	145	6	of	of	ADP
app01-9407	145	7	region	region	NOUN
app01-9407	146	1	3	3	NUM
app01-9407	146	2	:	:	PUNCT
app01-9407	146	3	input	input	NOUN
app01-9407	146	4	data	data	NOUN
app01-9407	146	5	slice	slice	NOUN
app01-9407	146	6	(	(	PUNCT
app01-9407	146	7	a	a	NOUN
app01-9407	146	8	)	)	PUNCT
app01-9407	146	9	,	,	PUNCT
app01-9407	146	10	otsu	otsu	NOUN
app01-9407	146	11	thresholding	thresholding	NOUN
app01-9407	146	12	overlaid	overlay	VERB
app01-9407	146	13	on	on	ADP
app01-9407	146	14	input	input	NOUN
app01-9407	146	15	data	datum	NOUN
app01-9407	146	16	slice	slice	NOUN
app01-9407	146	17	(	(	PUNCT
app01-9407	146	18	b	b	NOUN
app01-9407	146	19	)	)	PUNCT
app01-9407	146	20	,	,	PUNCT
app01-9407	146	21	prediction	prediction	NOUN
app01-9407	146	22	overlaid	overlay	VERB
app01-9407	146	23	on	on	ADP
app01-9407	146	24	input	input	NOUN
app01-9407	146	25	data	datum	NOUN
app01-9407	146	26	slice	slice	NOUN
app01-9407	146	27	(	(	PUNCT
app01-9407	146	28	c	c	NOUN
app01-9407	146	29	)	)	PUNCT
app01-9407	146	30	(	(	PUNCT
app01-9407	146	31	red=	red=	ADJ
app01-9407	146	32	otsu	otsu	NOUN
app01-9407	146	33	thresholding	thresholding	NOUN
app01-9407	146	34	,	,	PUNCT
app01-9407	146	35	yellow	yellow	ADJ
app01-9407	146	36	=	=	NOUN
app01-9407	146	37	predictions	prediction	NOUN
app01-9407	146	38	)	)	PUNCT
app01-9407	146	39	.	.	PUNCT
app01-9407	147	1	the	the	DET
app01-9407	147	2	detailed	detailed	ADJ
app01-9407	147	3	images	image	NOUN
app01-9407	147	4	display	display	VERB
app01-9407	147	5	a	a	DET
app01-9407	147	6	2x	2x	NUM
app01-9407	147	7	zoom	zoom	NOUN
app01-9407	147	8	of	of	ADP
app01-9407	147	9	the	the	DET
app01-9407	147	10	edge	edge	NOUN
app01-9407	147	11	region	region	NOUN
app01-9407	147	12	highlighted	highlight	VERB
app01-9407	147	13	in	in	ADP
app01-9407	147	14	blue	blue	NOUN
app01-9407	147	15	.	.	PUNCT
app01-9407	148	1	also	also	ADV
app01-9407	148	2	hard	hard	ADJ
app01-9407	148	3	to	to	PART
app01-9407	148	4	decide	decide	VERB
app01-9407	148	5	which	which	DET
app01-9407	148	6	area	area	NOUN
app01-9407	148	7	should	should	AUX
app01-9407	148	8	be	be	AUX
app01-9407	148	9	segmented	segment	VERB
app01-9407	148	10	as	as	ADP
app01-9407	148	11	pore	pore	ADJ
app01-9407	148	12	for	for	ADP
app01-9407	148	13	a	a	DET
app01-9407	148	14	human	human	NOUN
app01-9407	148	15	.	.	PUNCT
app01-9407	149	1	5.3	5.3	NUM
app01-9407	149	2	.	.	PUNCT
app01-9407	149	3	prediction	prediction	NOUN
app01-9407	149	4	and	and	CCONJ
app01-9407	149	5	visualization	visualization	NOUN
app01-9407	149	6	of	of	ADP
app01-9407	149	7	sample	sample	NOUN
app01-9407	149	8	3	3	NUM
app01-9407	149	9	the	the	DET
app01-9407	149	10	model	model	NOUN
app01-9407	149	11	trained	train	VERB
app01-9407	149	12	on	on	ADP
app01-9407	149	13	3.3	3.3	NUM
app01-9407	149	14	µm	µm	ADP
app01-9407	149	15	data	datum	NOUN
app01-9407	149	16	is	be	AUX
app01-9407	149	17	also	also	ADV
app01-9407	149	18	successfully	successfully	ADV
app01-9407	149	19	working	work	VERB
app01-9407	149	20	on	on	ADP
app01-9407	149	21	datasets	dataset	NOUN
app01-9407	149	22	with	with	ADP
app01-9407	149	23	different	different	ADJ
app01-9407	149	24	resolution	resolution	NOUN
app01-9407	149	25	.	.	PUNCT
app01-9407	150	1	to	to	PART
app01-9407	150	2	show	show	VERB
app01-9407	150	3	that	that	SCONJ
app01-9407	150	4	,	,	PUNCT
app01-9407	150	5	sample	sample	NOUN
app01-9407	150	6	3	3	NUM
app01-9407	150	7	is	be	AUX
app01-9407	150	8	segmented	segment	VERB
app01-9407	150	9	in	in	ADP
app01-9407	150	10	open_ia	open_ia	NOUN
app01-9407	150	11	with	with	ADP
app01-9407	150	12	the	the	DET
app01-9407	150	13	neural	neural	ADJ
app01-9407	150	14	network	network	NOUN
app01-9407	150	15	and	and	CCONJ
app01-9407	150	16	otsu	otsu	NOUN
app01-9407	150	17	thresholding	thresholding	NOUN
app01-9407	150	18	method	method	NOUN
app01-9407	150	19	,	,	PUNCT
app01-9407	150	20	the	the	DET
app01-9407	150	21	resulting	result	VERB
app01-9407	150	22	segmentation	segmentation	NOUN
app01-9407	150	23	outcomes	outcome	NOUN
app01-9407	150	24	are	be	AUX
app01-9407	150	25	presented	present	VERB
app01-9407	150	26	in	in	ADP
app01-9407	150	27	figure	figure	NOUN
app01-9407	150	28	6	6	NUM
app01-9407	150	29	.	.	PUNCT
app01-9407	151	1	however	however	ADV
app01-9407	151	2	,	,	PUNCT
app01-9407	151	3	there	there	PRON
app01-9407	151	4	are	be	VERB
app01-9407	151	5	some	some	DET
app01-9407	151	6	limitations	limitation	NOUN
app01-9407	151	7	with	with	ADP
app01-9407	151	8	applying	apply	VERB
app01-9407	151	9	the	the	DET
app01-9407	151	10	neural	neural	ADJ
app01-9407	151	11	network	network	NOUN
app01-9407	151	12	on	on	ADP
app01-9407	151	13	the	the	DET
app01-9407	151	14	different	different	ADJ
app01-9407	151	15	resolution	resolution	NOUN
app01-9407	151	16	data	datum	NOUN
app01-9407	151	17	:	:	PUNCT
app01-9407	151	18	the	the	DET
app01-9407	151	19	pores	pore	NOUN
app01-9407	151	20	(	(	PUNCT
app01-9407	151	21	long	long	ADJ
app01-9407	151	22	and	and	CCONJ
app01-9407	151	23	thin	thin	ADJ
app01-9407	151	24	)	)	PUNCT
app01-9407	151	25	are	be	AUX
app01-9407	151	26	barely	barely	ADV
app01-9407	151	27	visible	visible	ADJ
app01-9407	151	28	,	,	PUNCT
app01-9407	151	29	much	much	ADV
app01-9407	151	30	less	less	ADV
app01-9407	151	31	visible	visible	ADJ
app01-9407	151	32	than	than	ADP
app01-9407	151	33	in	in	ADP
app01-9407	151	34	the	the	DET
app01-9407	151	35	datasets	dataset	NOUN
app01-9407	151	36	with	with	ADP
app01-9407	151	37	3.3	3.3	NUM
app01-9407	151	38	µm	µm	ADP
app01-9407	151	39	resolution	resolution	NOUN
app01-9407	151	40	,	,	PUNCT
app01-9407	151	41	and	and	CCONJ
app01-9407	151	42	the	the	DET
app01-9407	151	43	transition	transition	NOUN
app01-9407	151	44	is	be	AUX
app01-9407	151	45	more	more	ADV
app01-9407	151	46	blurred	blurred	ADJ
app01-9407	151	47	between	between	ADP
app01-9407	151	48	pore	pore	ADJ
app01-9407	151	49	and	and	CCONJ
app01-9407	151	50	background	background	NOUN
app01-9407	151	51	.	.	PUNCT
app01-9407	152	1	the	the	DET
app01-9407	152	2	results	result	NOUN
app01-9407	152	3	show	show	VERB
app01-9407	152	4	that	that	SCONJ
app01-9407	152	5	the	the	DET
app01-9407	152	6	prediction	prediction	NOUN
app01-9407	152	7	of	of	ADP
app01-9407	152	8	the	the	DET
app01-9407	152	9	neural	neural	ADJ
app01-9407	152	10	network	network	NOUN
app01-9407	152	11	is	be	AUX
app01-9407	152	12	also	also	ADV
app01-9407	152	13	successfully	successfully	ADV
app01-9407	152	14	working	work	VERB
app01-9407	152	15	for	for	ADP
app01-9407	152	16	the	the	DET
app01-9407	152	17	segmentation	segmentation	NOUN
app01-9407	152	18	of	of	ADP
app01-9407	152	19	different	different	ADJ
app01-9407	152	20	shapes	shape	NOUN
app01-9407	152	21	of	of	ADP
app01-9407	152	22	pores	pore	NOUN
app01-9407	152	23	on	on	ADP
app01-9407	152	24	different	different	ADJ
app01-9407	152	25	resolutions	resolution	NOUN
app01-9407	152	26	(	(	PUNCT
app01-9407	152	27	see	see	VERB
app01-9407	152	28	figure	figure	NOUN
app01-9407	152	29	6	6	NUM
app01-9407	152	30	)	)	PUNCT
app01-9407	152	31	.	.	PUNCT
app01-9407	153	1	another	another	DET
app01-9407	153	2	result	result	NOUN
app01-9407	153	3	is	be	AUX
app01-9407	153	4	that	that	SCONJ
app01-9407	153	5	the	the	DET
app01-9407	153	6	otsu	otsu	NOUN
app01-9407	153	7	thresholding	thresholde	VERB
app01-9407	153	8	over	over	ADP
app01-9407	153	9	-	-	PUNCT
app01-9407	153	10	segments	segment	NOUN
app01-9407	153	11	pores	pore	NOUN
app01-9407	153	12	(	(	PUNCT
app01-9407	153	13	see	see	VERB
app01-9407	153	14	figure	figure	NOUN
app01-9407	153	15	7b	7b	NOUN
app01-9407	153	16	)	)	PUNCT
app01-9407	153	17	.	.	PUNCT
app01-9407	154	1	the	the	DET
app01-9407	154	2	main	main	ADJ
app01-9407	154	3	challenge	challenge	NOUN
app01-9407	154	4	in	in	ADP
app01-9407	154	5	this	this	DET
app01-9407	154	6	scenario	scenario	NOUN
app01-9407	154	7	is	be	AUX
app01-9407	154	8	the	the	DET
app01-9407	154	9	segmentation	segmentation	NOUN
app01-9407	154	10	of	of	ADP
app01-9407	154	11	long	long	ADJ
app01-9407	154	12	and	and	CCONJ
app01-9407	154	13	thin	thin	ADJ
app01-9407	154	14	pores	pore	NOUN
app01-9407	154	15	,	,	PUNCT
app01-9407	154	16	particularly	particularly	ADV
app01-9407	154	17	in	in	ADP
app01-9407	154	18	cases	case	NOUN
app01-9407	154	19	where	where	SCONJ
app01-9407	154	20	they	they	PRON
app01-9407	154	21	are	be	AUX
app01-9407	154	22	in	in	ADP
app01-9407	154	23	close	close	ADJ
app01-9407	154	24	proximity	proximity	NOUN
app01-9407	154	25	to	to	ADP
app01-9407	154	26	one	one	NUM
app01-9407	154	27	another	another	DET
app01-9407	154	28	(	(	PUNCT
app01-9407	154	29	see	see	VERB
app01-9407	154	30	figure	figure	NOUN
app01-9407	154	31	7	7	NUM
app01-9407	154	32	)	)	PUNCT
app01-9407	154	33	.	.	PUNCT
app01-9407	155	1	proper	proper	ADJ
app01-9407	155	2	segmentation	segmentation	NOUN
app01-9407	155	3	is	be	AUX
app01-9407	155	4	challenging	challenge	VERB
app01-9407	155	5	for	for	ADP
app01-9407	155	6	the	the	DET
app01-9407	155	7	network	network	NOUN
app01-9407	155	8	model	model	NOUN
app01-9407	155	9	and	and	CCONJ
app01-9407	155	10	other	other	ADJ
app01-9407	155	11	methods	method	NOUN
app01-9407	155	12	.	.	PUNCT
app01-9407	156	1	nevertheless	nevertheless	ADV
app01-9407	156	2	,	,	PUNCT
app01-9407	156	3	results	result	NOUN
app01-9407	156	4	are	be	AUX
app01-9407	156	5	impressive	impressive	ADJ
app01-9407	156	6	and	and	CCONJ
app01-9407	156	7	promising	promise	VERB
app01-9407	156	8	for	for	ADP
app01-9407	156	9	future	future	ADJ
app01-9407	156	10	improvements	improvement	NOUN
app01-9407	156	11	.	.	PUNCT
app01-9407	157	1	6	6	X
app01-9407	157	2	.	.	X
app01-9407	157	3	conclusions	conclusion	NOUN
app01-9407	157	4	this	this	DET
app01-9407	157	5	paper	paper	NOUN
app01-9407	157	6	demonstrates	demonstrate	VERB
app01-9407	157	7	that	that	SCONJ
app01-9407	157	8	the	the	DET
app01-9407	157	9	modified	modify	VERB
app01-9407	157	10	3d	3d	PROPN
app01-9407	157	11	u	u	ADJ
app01-9407	157	12	-	-	ADJ
app01-9407	157	13	net	net	ADJ
app01-9407	157	14	architecture	architecture	NOUN
app01-9407	157	15	achieves	achieve	VERB
app01-9407	157	16	satisfactory	satisfactory	ADJ
app01-9407	157	17	performance	performance	NOUN
app01-9407	157	18	,	,	PUNCT
app01-9407	157	19	even	even	ADV
app01-9407	157	20	when	when	SCONJ
app01-9407	157	21	trained	train	VERB
app01-9407	157	22	on	on	ADP
app01-9407	157	23	a	a	DET
app01-9407	157	24	limited	limited	ADJ
app01-9407	157	25	training	training	NOUN
app01-9407	157	26	dataset	dataset	NOUN
app01-9407	157	27	.	.	PUNCT
app01-9407	158	1	through	through	ADP
app01-9407	158	2	visual	visual	ADJ
app01-9407	158	3	inspection	inspection	NOUN
app01-9407	158	4	and	and	CCONJ
app01-9407	158	5	a	a	DET
app01-9407	158	6	comparative	comparative	ADJ
app01-9407	158	7	assessment	assessment	NOUN
app01-9407	158	8	with	with	ADP
app01-9407	158	9	the	the	DET
app01-9407	158	10	findings	finding	NOUN
app01-9407	158	11	,	,	PUNCT
app01-9407	158	12	the	the	DET
app01-9407	158	13	modified	modify	VERB
app01-9407	158	14	3d	3d	NUM
app01-9407	158	15	u	u	NOUN
app01-9407	158	16	-	-	NOUN
app01-9407	158	17	net	net	ADJ
app01-9407	158	18	performs	perform	VERB
app01-9407	158	19	better	well	ADV
app01-9407	158	20	in	in	ADP
app01-9407	158	21	certain	certain	ADJ
app01-9407	158	22	areas	area	NOUN
app01-9407	158	23	compared	compare	VERB
app01-9407	158	24	to	to	ADP
app01-9407	158	25	otsu	otsu	ADJ
app01-9407	158	26	thresholding	thresholding	NOUN
app01-9407	158	27	.	.	PUNCT
app01-9407	159	1	significantly	significantly	ADV
app01-9407	159	2	,	,	PUNCT
app01-9407	159	3	the	the	DET
app01-9407	159	4	generated	generate	VERB
app01-9407	159	5	model	model	NOUN
app01-9407	159	6	with	with	ADP
app01-9407	159	7	3	3	NUM
app01-9407	159	8	µm	µm	ADP
app01-9407	159	9	resolution	resolution	NOUN
app01-9407	159	10	also	also	ADV
app01-9407	159	11	demonstrated	demonstrate	VERB
app01-9407	159	12	a	a	DET
app01-9407	159	13	satisfactory	satisfactory	ADJ
app01-9407	159	14	performance	performance	NOUN
app01-9407	159	15	on	on	ADP
app01-9407	159	16	the	the	DET
app01-9407	159	17	datasets	dataset	NOUN
app01-9407	159	18	with	with	ADP
app01-9407	159	19	10	10	NUM
app01-9407	159	20	µm	µm	NOUN
app01-9407	159	21	resolution	resolution	NOUN
app01-9407	159	22	.	.	PUNCT
app01-9407	160	1	it	it	PRON
app01-9407	160	2	is	be	AUX
app01-9407	160	3	important	important	ADJ
app01-9407	160	4	to	to	PART
app01-9407	160	5	note	note	VERB
app01-9407	160	6	that	that	SCONJ
app01-9407	160	7	during	during	ADP
app01-9407	160	8	training	training	NOUN
app01-9407	160	9	,	,	PUNCT
app01-9407	160	10	only	only	ADV
app01-9407	160	11	the	the	DET
app01-9407	160	12	3	3	NUM
app01-9407	160	13	µm	µm	ADP
app01-9407	160	14	resolution	resolution	NOUN
app01-9407	160	15	(	(	PUNCT
app01-9407	160	16	sample	sample	NOUN
app01-9407	160	17	1	1	NUM
app01-9407	160	18	)	)	PUNCT
app01-9407	160	19	was	be	AUX
app01-9407	160	20	utilized	utilize	VERB
app01-9407	160	21	.	.	PUNCT
app01-9407	161	1	quality	quality	NOUN
app01-9407	161	2	assessment	assessment	NOUN
app01-9407	161	3	and	and	CCONJ
app01-9407	161	4	comparisons	comparison	NOUN
app01-9407	161	5	in	in	ADP
app01-9407	161	6	this	this	DET
app01-9407	161	7	workflow	workflow	NOUN
app01-9407	161	8	are	be	AUX
app01-9407	161	9	conducted	conduct	VERB
app01-9407	161	10	through	through	ADP
app01-9407	161	11	manual	manual	ADJ
app01-9407	161	12	visual	visual	ADJ
app01-9407	161	13	inspection	inspection	NOUN
app01-9407	161	14	.	.	PUNCT
app01-9407	162	1	for	for	ADP
app01-9407	162	2	future	future	ADJ
app01-9407	162	3	work	work	NOUN
app01-9407	162	4	,	,	PUNCT
app01-9407	162	5	we	we	PRON
app01-9407	162	6	plan	plan	VERB
app01-9407	162	7	to	to	PART
app01-9407	162	8	measure	measure	VERB
app01-9407	162	9	the	the	DET
app01-9407	162	10	quality	quality	NOUN
app01-9407	162	11	numerically	numerically	ADV
app01-9407	162	12	by	by	ADP
app01-9407	162	13	using	use	VERB
app01-9407	162	14	xct	xct	PROPN
app01-9407	162	15	simulated	simulated	ADJ
app01-9407	162	16	data	datum	NOUN
app01-9407	162	17	on	on	ADP
app01-9407	162	18	cfrp	cfrp	NOUN
app01-9407	162	19	.	.	PUNCT
app01-9407	163	1	however	however	ADV
app01-9407	163	2	,	,	PUNCT
app01-9407	163	3	to	to	PART
app01-9407	163	4	measure	measure	VERB
app01-9407	163	5	the	the	DET
app01-9407	163	6	quality	quality	NOUN
app01-9407	163	7	of	of	ADP
app01-9407	163	8	estimations	estimation	NOUN
app01-9407	163	9	and	and	CCONJ
app01-9407	163	10	the	the	DET
app01-9407	163	11	performance	performance	NOUN
app01-9407	163	12	of	of	ADP
app01-9407	163	13	the	the	DET
app01-9407	163	14	deep	deep	ADJ
app01-9407	163	15	learning	learning	NOUN
app01-9407	163	16	model	model	NOUN
app01-9407	163	17	,	,	PUNCT
app01-9407	163	18	the	the	DET
app01-9407	163	19	generation	generation	NOUN
app01-9407	163	20	of	of	ADP
app01-9407	163	21	accurate	accurate	ADJ
app01-9407	163	22	ground	ground	NOUN
app01-9407	163	23	truth	truth	NOUN
app01-9407	163	24	is	be	AUX
app01-9407	163	25	a	a	DET
app01-9407	163	26	necessity	necessity	NOUN
app01-9407	163	27	.	.	PUNCT
app01-9407	164	1	therefore	therefore	ADV
app01-9407	164	2	,	,	PUNCT
app01-9407	164	3	we	we	PRON
app01-9407	164	4	will	will	AUX
app01-9407	164	5	improve	improve	VERB
app01-9407	164	6	our	our	PRON
app01-9407	164	7	reference	reference	NOUN
app01-9407	164	8	segmentation	segmentation	NOUN
app01-9407	164	9	by	by	ADP
app01-9407	164	10	incorporating	incorporate	VERB
app01-9407	164	11	both	both	CCONJ
app01-9407	164	12	xct	xct	NOUN
app01-9407	164	13	images	image	NOUN
app01-9407	164	14	segmented	segment	VERB
app01-9407	164	15	by	by	ADP
app01-9407	164	16	human	human	ADJ
app01-9407	164	17	experts	expert	NOUN
app01-9407	164	18	and	and	CCONJ
app01-9407	164	19	xct	xct	PROPN
app01-9407	164	20	simulation	simulation	NOUN
app01-9407	164	21	techniques	technique	NOUN
app01-9407	164	22	.	.	PUNCT
app01-9407	165	1	another	another	DET
app01-9407	165	2	important	important	ADJ
app01-9407	165	3	point	point	NOUN
app01-9407	165	4	is	be	AUX
app01-9407	165	5	the	the	DET
app01-9407	165	6	uncertainty	uncertainty	NOUN
app01-9407	165	7	determination	determination	NOUN
app01-9407	165	8	for	for	ADP
app01-9407	165	9	generation	generation	NOUN
app01-9407	165	10	of	of	ADP
app01-9407	165	11	xct	xct	PROPN
app01-9407	165	12	images	image	NOUN
app01-9407	165	13	(	(	PUNCT
app01-9407	165	14	radiographs	radiograph	NOUN
app01-9407	165	15	)	)	PUNCT
app01-9407	165	16	,	,	PUNCT
app01-9407	165	17	reconstruction	reconstruction	NOUN
app01-9407	165	18	,	,	PUNCT
app01-9407	165	19	segmentation	segmentation	NOUN
app01-9407	165	20	and	and	CCONJ
app01-9407	165	21	deep	deep	ADJ
app01-9407	165	22	learning	learning	NOUN
app01-9407	165	23	training	training	NOUN
app01-9407	165	24	.	.	PUNCT
app01-9407	166	1	each	each	PRON
app01-9407	166	2	of	of	ADP
app01-9407	166	3	these	these	DET
app01-9407	166	4	steps	step	NOUN
app01-9407	166	5	can	can	AUX
app01-9407	166	6	influence	influence	VERB
app01-9407	166	7	the	the	DET
app01-9407	166	8	final	final	ADJ
app01-9407	166	9	image	image	NOUN
app01-9407	166	10	segmentation	segmentation	NOUN
app01-9407	166	11	or	or	CCONJ
app01-9407	166	12	estimation	estimation	NOUN
app01-9407	166	13	.	.	PUNCT
app01-9407	167	1	to	to	PART
app01-9407	167	2	improve	improve	VERB
app01-9407	167	3	segmentation	segmentation	NOUN
app01-9407	167	4	accuracy	accuracy	NOUN
app01-9407	167	5	,	,	PUNCT
app01-9407	167	6	uncertainty	uncertainty	NOUN
app01-9407	167	7	sources	source	NOUN
app01-9407	167	8	and	and	CCONJ
app01-9407	167	9	their	their	PRON
app01-9407	167	10	levels	level	NOUN
app01-9407	167	11	have	have	VERB
app01-9407	167	12	to	to	PART
app01-9407	167	13	be	be	AUX
app01-9407	167	14	identified	identify	VERB
app01-9407	167	15	.	.	PUNCT
app01-9407	168	1	additionally	additionally	ADV
app01-9407	168	2	,	,	PUNCT
app01-9407	168	3	we	we	PRON
app01-9407	168	4	intend	intend	VERB
app01-9407	168	5	to	to	PART
app01-9407	168	6	explore	explore	VERB
app01-9407	168	7	the	the	DET
app01-9407	168	8	assessment	assessment	NOUN
app01-9407	168	9	of	of	ADP
app01-9407	168	10	algorithm	algorithm	NOUN
app01-9407	168	11	reliability	reliability	NOUN
app01-9407	168	12	through	through	ADP
app01-9407	168	13	the	the	DET
app01-9407	168	14	creation	creation	NOUN
app01-9407	168	15	of	of	ADP
app01-9407	168	16	artificial	artificial	ADJ
app01-9407	168	17	xct	xct	PROPN
app01-9407	168	18	images	image	NOUN
app01-9407	168	19	containing	contain	VERB
app01-9407	168	20	known	know	VERB
app01-9407	168	21	pore	pore	ADJ
app01-9407	168	22	sizes	size	NOUN
app01-9407	168	23	.	.	PUNCT
app01-9407	169	1	moreover	moreover	ADV
app01-9407	169	2	,	,	PUNCT
app01-9407	169	3	we	we	PRON
app01-9407	169	4	will	will	AUX
app01-9407	169	5	investigate	investigate	VERB
app01-9407	169	6	the	the	DET
app01-9407	169	7	impact	impact	NOUN
app01-9407	169	8	of	of	ADP
app01-9407	169	9	introducing	introduce	VERB
app01-9407	169	10	noise	noise	NOUN
app01-9407	169	11	or	or	CCONJ
app01-9407	169	12	image	image	NOUN
app01-9407	169	13	blurring	blur	VERB
app01-9407	169	14	on	on	ADP
app01-9407	169	15	the	the	DET
app01-9407	169	16	algorithm	algorithm	NOUN
app01-9407	169	17	’s	’s	PART
app01-9407	169	18	sensitivity	sensitivity	NOUN
app01-9407	169	19	to	to	ADP
app01-9407	169	20	various	various	ADJ
app01-9407	169	21	side	side	NOUN
app01-9407	169	22	effects	effect	NOUN
app01-9407	169	23	.	.	PUNCT
app01-9407	170	1	92	92	NUM
app01-9407	170	2	vol	vol	NOUN
app01-9407	170	3	.	.	PUNCT
app01-9407	171	1	42/2023	42/2023	NUM
app01-9407	171	2	cfrp	cfrp	PROPN
app01-9407	171	3	pore	pore	ADJ
app01-9407	171	4	segmentation	segmentation	NOUN
app01-9407	171	5	via	via	ADP
app01-9407	171	6	u	u	ADJ
app01-9407	171	7	-	-	ADJ
app01-9407	171	8	net	net	ADJ
app01-9407	171	9	acknowledgements	acknowledgement	NOUN
app01-9407	171	10	this	this	DET
app01-9407	171	11	research	research	NOUN
app01-9407	171	12	was	be	AUX
app01-9407	171	13	co	co	VERB
app01-9407	171	14	-	-	VERB
app01-9407	171	15	financed	finance	VERB
app01-9407	171	16	by	by	ADP
app01-9407	171	17	the	the	DET
app01-9407	171	18	european	european	PROPN
app01-9407	171	19	union	union	PROPN
app01-9407	171	20	h2020	h2020	PROPN
app01-9407	171	21	-	-	PUNCT
app01-9407	171	22	msca	msca	NOUN
app01-9407	171	23	-	-	PUNCT
app01-9407	171	24	itn-2020	itn-2020	NOUN
app01-9407	171	25	under	under	ADP
app01-9407	171	26	grant	grant	NOUN
app01-9407	171	27	agreement	agreement	NOUN
app01-9407	171	28	no	no	INTJ
app01-9407	171	29	.	.	PUNCT
app01-9407	171	30	956172	956172	NUM
app01-9407	171	31	(	(	PUNCT
app01-9407	171	32	xcting	xcte	VERB
app01-9407	171	33	)	)	PUNCT
app01-9407	171	34	.	.	PUNCT
app01-9407	172	1	references	reference	NOUN
app01-9407	172	2	[	[	X
app01-9407	172	3	1	1	NUM
app01-9407	172	4	]	]	PUNCT
app01-9407	172	5	c.	c.	PROPN
app01-9407	172	6	heinzl	heinzl	PROPN
app01-9407	172	7	,	,	PUNCT
app01-9407	172	8	s.	s.	PROPN
app01-9407	172	9	stappen	stappen	PROPN
app01-9407	172	10	.	.	PUNCT
app01-9407	173	1	star	star	NOUN
app01-9407	173	2	:	:	PUNCT
app01-9407	173	3	visual	visual	ADJ
app01-9407	173	4	computing	computing	NOUN
app01-9407	173	5	in	in	ADP
app01-9407	173	6	materials	material	NOUN
app01-9407	173	7	science	science	NOUN
app01-9407	173	8	.	.	PUNCT
app01-9407	174	1	computer	computer	NOUN
app01-9407	174	2	graphics	graphics	PROPN
app01-9407	174	3	forum	forum	PROPN
app01-9407	174	4	36:647	36:647	PROPN
app01-9407	174	5	–	–	PUNCT
app01-9407	174	6	666	666	NUM
app01-9407	174	7	,	,	PUNCT
app01-9407	174	8	2017	2017	NUM
app01-9407	174	9	.	.	PUNCT
app01-9407	175	1	https://doi.org/10.1111/cgf.13214	https://doi.org/10.1111/cgf.13214	PROPN
app01-9407	175	2	[	[	X
app01-9407	175	3	2	2	X
app01-9407	175	4	]	]	X
app01-9407	175	5	d.	d.	PROPN
app01-9407	175	6	cireşan	cireşan	PROPN
app01-9407	175	7	,	,	PUNCT
app01-9407	175	8	u.	u.	PROPN
app01-9407	175	9	meier	meier	PROPN
app01-9407	175	10	,	,	PUNCT
app01-9407	175	11	j.	j.	PROPN
app01-9407	175	12	schmidhuber	schmidhuber	PROPN
app01-9407	175	13	.	.	PUNCT
app01-9407	176	1	multi	multi	ADJ
app01-9407	176	2	-	-	ADJ
app01-9407	176	3	column	column	ADJ
app01-9407	176	4	deep	deep	ADJ
app01-9407	176	5	neural	neural	ADJ
app01-9407	176	6	networks	network	NOUN
app01-9407	176	7	for	for	ADP
app01-9407	176	8	image	image	NOUN
app01-9407	176	9	classification	classification	NOUN
app01-9407	176	10	.	.	PUNCT
app01-9407	177	1	proceedings	proceeding	NOUN
app01-9407	177	2	/	/	SYM
app01-9407	177	3	cvpr	cvpr	NOUN
app01-9407	177	4	,	,	PUNCT
app01-9407	177	5	ieee	ieee	NOUN
app01-9407	177	6	computer	computer	NOUN
app01-9407	177	7	society	society	PROPN
app01-9407	177	8	conference	conference	NOUN
app01-9407	177	9	on	on	ADP
app01-9407	177	10	computer	computer	NOUN
app01-9407	177	11	vision	vision	NOUN
app01-9407	177	12	and	and	CCONJ
app01-9407	177	13	pattern	pattern	NOUN
app01-9407	177	14	recognition	recognition	NOUN
app01-9407	177	15	ieee	ieee	PROPN
app01-9407	177	16	computer	computer	NOUN
app01-9407	177	17	society	society	PROPN
app01-9407	177	18	conference	conference	NOUN
app01-9407	177	19	on	on	ADP
app01-9407	177	20	computer	computer	NOUN
app01-9407	177	21	vision	vision	NOUN
app01-9407	177	22	and	and	CCONJ
app01-9407	177	23	pattern	pattern	NOUN
app01-9407	177	24	recognition	recognition	NOUN
app01-9407	177	25	2012	2012	NUM
app01-9407	177	26	.	.	PUNCT
app01-9407	178	1	[	[	X
app01-9407	178	2	202306	202306	NUM
app01-9407	178	3	-	-	SYM
app01-9407	178	4	30	30	NUM
app01-9407	178	5	]	]	PUNCT
app01-9407	178	6	.	.	PUNCT
app01-9407	179	1	https://doi.org/10.48550/arxiv.1202.2745	https://doi.org/10.48550/arxiv.1202.2745	PROPN
app01-9407	180	1	[	[	X
app01-9407	180	2	3	3	X
app01-9407	180	3	]	]	X
app01-9407	180	4	j.	j.	PROPN
app01-9407	180	5	ba	ba	PROPN
app01-9407	180	6	,	,	PUNCT
app01-9407	180	7	v.	v.	PROPN
app01-9407	180	8	mnih	mnih	PROPN
app01-9407	180	9	,	,	PUNCT
app01-9407	180	10	k.	k.	PROPN
app01-9407	180	11	kavukcuoglu	kavukcuoglu	PROPN
app01-9407	180	12	.	.	PUNCT
app01-9407	181	1	multiple	multiple	ADJ
app01-9407	181	2	object	object	NOUN
app01-9407	181	3	recognition	recognition	NOUN
app01-9407	181	4	with	with	ADP
app01-9407	181	5	visual	visual	ADJ
app01-9407	181	6	attention	attention	NOUN
app01-9407	181	7	.	.	PUNCT
app01-9407	182	1	arxiv	arxiv	PROPN
app01-9407	182	2	2014	2014	NUM
app01-9407	182	3	.	.	PUNCT
app01-9407	183	1	[	[	X
app01-9407	183	2	2023	2023	NUM
app01-9407	183	3	-	-	SYM
app01-9407	183	4	0630	0630	NUM
app01-9407	183	5	]	]	PUNCT
app01-9407	183	6	.	.	PUNCT
app01-9407	184	1	https://doi.org/10.48550/arxiv.1412.7755	https://doi.org/10.48550/arxiv.1412.7755	PROPN
app01-9407	184	2	[	[	X
app01-9407	184	3	4	4	NUM
app01-9407	184	4	]	]	PUNCT
app01-9407	184	5	j.	j.	PROPN
app01-9407	184	6	ker	ker	PROPN
app01-9407	184	7	,	,	PUNCT
app01-9407	184	8	l.	l.	PROPN
app01-9407	184	9	wang	wang	PROPN
app01-9407	184	10	,	,	PUNCT
app01-9407	184	11	j.	j.	PROPN
app01-9407	184	12	rao	rao	PROPN
app01-9407	184	13	,	,	PUNCT
app01-9407	184	14	t.	t.	PROPN
app01-9407	184	15	lim	lim	PROPN
app01-9407	184	16	.	.	PUNCT
app01-9407	185	1	deep	deep	ADJ
app01-9407	185	2	learning	learning	NOUN
app01-9407	185	3	applications	application	NOUN
app01-9407	185	4	in	in	ADP
app01-9407	185	5	medical	medical	ADJ
app01-9407	185	6	image	image	NOUN
app01-9407	185	7	analysis	analysis	NOUN
app01-9407	185	8	.	.	PUNCT
app01-9407	186	1	ieee	ieee	NOUN
app01-9407	186	2	access	access	NOUN
app01-9407	186	3	6:9375–9389	6:9375–9389	PROPN
app01-9407	186	4	,	,	PUNCT
app01-9407	186	5	2018	2018	NUM
app01-9407	186	6	.	.	PUNCT
app01-9407	187	1	https://doi.org/10.1109/access.2017.2788044	https://doi.org/10.1109/access.2017.2788044	PROPN
app01-9407	188	1	[	[	X
app01-9407	188	2	5	5	NUM
app01-9407	188	3	]	]	PUNCT
app01-9407	188	4	m.	m.	NOUN
app01-9407	188	5	yosifov	yosifov	NOUN
app01-9407	188	6	.	.	PUNCT
app01-9407	188	7	extraction	extraction	NOUN
app01-9407	188	8	and	and	CCONJ
app01-9407	188	9	quantification	quantification	NOUN
app01-9407	188	10	of	of	ADP
app01-9407	188	11	features	feature	NOUN
app01-9407	188	12	in	in	ADP
app01-9407	188	13	xct	xct	PROPN
app01-9407	188	14	datasets	dataset	NOUN
app01-9407	188	15	of	of	ADP
app01-9407	188	16	fibre	fibre	NOUN
app01-9407	188	17	reinforced	reinforce	VERB
app01-9407	188	18	polymers	polymer	NOUN
app01-9407	188	19	using	use	VERB
app01-9407	188	20	machine	machine	NOUN
app01-9407	188	21	learning	learning	NOUN
app01-9407	188	22	techniques	technique	NOUN
app01-9407	188	23	.	.	PUNCT
app01-9407	189	1	master	master	NOUN
app01-9407	189	2	’s	’s	PART
app01-9407	189	3	thesis	thesis	NOUN
app01-9407	189	4	,	,	PUNCT
app01-9407	189	5	umeå	umeå	PROPN
app01-9407	189	6	university	university	PROPN
app01-9407	189	7	,	,	PUNCT
app01-9407	189	8	department	department	NOUN
app01-9407	189	9	of	of	ADP
app01-9407	189	10	computing	compute	VERB
app01-9407	189	11	science	science	NOUN
app01-9407	189	12	,	,	PUNCT
app01-9407	189	13	2020	2020	NUM
app01-9407	189	14	.	.	PUNCT
app01-9407	190	1	[	[	X
app01-9407	190	2	6	6	NUM
app01-9407	190	3	]	]	PUNCT
app01-9407	190	4	m.	m.	NOUN
app01-9407	190	5	yosifov	yosifov	NOUN
app01-9407	190	6	,	,	PUNCT
app01-9407	190	7	m.	m.	NOUN
app01-9407	190	8	reiter	reiter	PROPN
app01-9407	190	9	,	,	PUNCT
app01-9407	190	10	s.	s.	PROPN
app01-9407	190	11	heupl	heupl	PROPN
app01-9407	190	12	,	,	PUNCT
app01-9407	190	13	et	et	PROPN
app01-9407	190	14	al	al	PROPN
app01-9407	190	15	.	.	PROPN
app01-9407	190	16	probability	probability	NOUN
app01-9407	190	17	of	of	ADP
app01-9407	190	18	detection	detection	NOUN
app01-9407	190	19	applied	apply	VERB
app01-9407	190	20	to	to	ADP
app01-9407	190	21	x	x	ADJ
app01-9407	190	22	-	-	NOUN
app01-9407	190	23	ray	ray	NOUN
app01-9407	190	24	inspection	inspection	NOUN
app01-9407	190	25	using	use	VERB
app01-9407	190	26	numerical	numerical	ADJ
app01-9407	190	27	simulations	simulation	NOUN
app01-9407	190	28	.	.	PUNCT
app01-9407	191	1	nondestructive	nondestructive	ADJ
app01-9407	191	2	testing	testing	NOUN
app01-9407	191	3	and	and	CCONJ
app01-9407	191	4	evaluation	evaluation	NOUN
app01-9407	191	5	0(0):1–16	0(0):1–16	NOUN
app01-9407	191	6	,	,	PUNCT
app01-9407	191	7	2022	2022	NUM
app01-9407	191	8	.	.	PUNCT
app01-9407	192	1	https://doi.org/10.1080/10589759.2022.2071892	https://doi.org/10.1080/10589759.2022.2071892	VERB
app01-9407	193	1	[	[	X
app01-9407	193	2	7	7	X
app01-9407	193	3	]	]	X
app01-9407	193	4	m.	m.	NOUN
app01-9407	193	5	yosifov	yosifov	NOUN
app01-9407	193	6	,	,	PUNCT
app01-9407	193	7	p.	p.	NOUN
app01-9407	193	8	weinberger	weinberger	PROPN
app01-9407	193	9	,	,	PUNCT
app01-9407	193	10	m.	m.	NOUN
app01-9407	193	11	reiter	reiter	PROPN
app01-9407	193	12	,	,	PUNCT
app01-9407	193	13	et	et	PROPN
app01-9407	194	1	al	al	PROPN
app01-9407	194	2	.	.	PUNCT
app01-9407	194	3	defect	defect	VERB
app01-9407	194	4	detectability	detectability	NOUN
app01-9407	194	5	analysis	analysis	NOUN
app01-9407	194	6	via	via	ADP
app01-9407	194	7	probability	probability	NOUN
app01-9407	194	8	of	of	ADP
app01-9407	194	9	defect	defect	ADJ
app01-9407	194	10	detection	detection	NOUN
app01-9407	194	11	between	between	ADP
app01-9407	194	12	traditional	traditional	ADJ
app01-9407	194	13	and	and	CCONJ
app01-9407	194	14	deep	deep	ADJ
app01-9407	194	15	learning	learning	NOUN
app01-9407	194	16	methods	method	NOUN
app01-9407	194	17	in	in	ADP
app01-9407	194	18	numerical	numerical	ADJ
app01-9407	194	19	simulations	simulation	NOUN
app01-9407	194	20	.	.	PUNCT
app01-9407	195	1	e	e	X
app01-9407	195	2	-	-	NOUN
app01-9407	195	3	journal	journal	NOUN
app01-9407	195	4	of	of	ADP
app01-9407	195	5	nondestructive	nondestructive	ADJ
app01-9407	195	6	testing	testing	NOUN
app01-9407	195	7	28	28	NUM
app01-9407	195	8	,	,	PUNCT
app01-9407	195	9	2023	2023	NUM
app01-9407	195	10	.	.	PUNCT
app01-9407	196	1	https://doi.org/10.58286/27716	https://doi.org/10.58286/27716	NUM
app01-9407	196	2	[	[	X
app01-9407	196	3	8	8	NUM
app01-9407	196	4	]	]	PUNCT
app01-9407	196	5	s.	s.	PROPN
app01-9407	196	6	ray	ray	PROPN
app01-9407	196	7	,	,	PUNCT
app01-9407	196	8	r.	r.	PROPN
app01-9407	196	9	h.	h.	PROPN
app01-9407	196	10	turi	turi	PROPN
app01-9407	196	11	.	.	PUNCT
app01-9407	197	1	determination	determination	NOUN
app01-9407	197	2	of	of	ADP
app01-9407	197	3	number	number	NOUN
app01-9407	197	4	of	of	ADP
app01-9407	197	5	clusters	cluster	NOUN
app01-9407	197	6	in	in	ADP
app01-9407	197	7	k	k	PROPN
app01-9407	197	8	-	-	PUNCT
app01-9407	197	9	means	mean	VERB
app01-9407	197	10	clustering	clustering	NOUN
app01-9407	197	11	and	and	CCONJ
app01-9407	197	12	application	application	NOUN
app01-9407	197	13	in	in	ADP
app01-9407	197	14	colour	colour	NOUN
app01-9407	197	15	image	image	NOUN
app01-9407	197	16	segmentation	segmentation	NOUN
app01-9407	197	17	.	.	PUNCT
app01-9407	198	1	in	in	ADP
app01-9407	198	2	international	international	ADJ
app01-9407	198	3	conference	conference	NOUN
app01-9407	198	4	on	on	ADP
app01-9407	198	5	advances	advance	NOUN
app01-9407	198	6	in	in	ADP
app01-9407	198	7	pattern	pattern	NOUN
app01-9407	198	8	recognition	recognition	NOUN
app01-9407	198	9	and	and	CCONJ
app01-9407	198	10	digital	digital	ADJ
app01-9407	198	11	techniques	technique	NOUN
app01-9407	198	12	,	,	PUNCT
app01-9407	198	13	pp	pp	ADV
app01-9407	198	14	.	.	PUNCT
app01-9407	199	1	137–143	137–143	NUM
app01-9407	199	2	.	.	PUNCT
app01-9407	199	3	1999	1999	NUM
app01-9407	199	4	.	.	PUNCT
app01-9407	200	1	[	[	X
app01-9407	200	2	9	9	NUM
app01-9407	200	3	]	]	PUNCT
app01-9407	200	4	j.	j.	PROPN
app01-9407	200	5	a.	a.	PROPN
app01-9407	200	6	hartigan	hartigan	PROPN
app01-9407	200	7	,	,	PUNCT
app01-9407	200	8	m.	m.	NOUN
app01-9407	200	9	a.	a.	PROPN
app01-9407	200	10	wong	wong	PROPN
app01-9407	200	11	.	.	PUNCT
app01-9407	200	12	algorithm	algorithm	PROPN
app01-9407	200	13	as	as	ADP
app01-9407	200	14	136	136	NUM
app01-9407	200	15	:	:	PUNCT
app01-9407	200	16	a	a	DET
app01-9407	200	17	k	k	X
app01-9407	200	18	-	-	PUNCT
app01-9407	200	19	means	mean	VERB
app01-9407	200	20	clustering	clustering	ADJ
app01-9407	200	21	algorithm	algorithm	NOUN
app01-9407	200	22	.	.	PUNCT
app01-9407	201	1	journal	journal	NOUN
app01-9407	201	2	of	of	ADP
app01-9407	201	3	the	the	DET
app01-9407	201	4	royal	royal	ADJ
app01-9407	201	5	statistical	statistical	ADJ
app01-9407	201	6	society	society	NOUN
app01-9407	201	7	series	series	NOUN
app01-9407	201	8	c	c	PROPN
app01-9407	201	9	(	(	PUNCT
app01-9407	201	10	applied	apply	VERB
app01-9407	201	11	statistics	statistic	NOUN
app01-9407	201	12	)	)	PUNCT
app01-9407	201	13	28(1):100	28(1):100	NUM
app01-9407	201	14	–	–	PUNCT
app01-9407	201	15	108	108	NUM
app01-9407	201	16	,	,	PUNCT
app01-9407	201	17	1979	1979	NUM
app01-9407	201	18	.	.	PUNCT
app01-9407	202	1	https://doi.org/10.2307/2346830	https://doi.org/10.2307/2346830	NOUN
app01-9407	203	1	[	[	X
app01-9407	203	2	10	10	NUM
app01-9407	203	3	]	]	X
app01-9407	203	4	l.	l.	PROPN
app01-9407	203	5	j.	j.	PROPN
app01-9407	203	6	belaid	belaid	PROPN
app01-9407	203	7	,	,	PUNCT
app01-9407	203	8	w.	w.	PROPN
app01-9407	203	9	mourou	mourou	PROPN
app01-9407	203	10	.	.	PUNCT
app01-9407	204	1	image	image	NOUN
app01-9407	204	2	segmentation	segmentation	NOUN
app01-9407	204	3	:	:	PUNCT
app01-9407	204	4	a	a	DET
app01-9407	204	5	watershed	watershe	VERB
app01-9407	204	6	transformation	transformation	NOUN
app01-9407	204	7	algorithm	algorithm	NOUN
app01-9407	204	8	.	.	PUNCT
app01-9407	205	1	image	image	NOUN
app01-9407	205	2	analysis	analysis	NOUN
app01-9407	205	3	&	&	CCONJ
app01-9407	205	4	stereology	stereology	NOUN
app01-9407	205	5	28(2):93–102	28(2):93–102	NUM
app01-9407	205	6	,	,	PUNCT
app01-9407	205	7	2009	2009	NUM
app01-9407	205	8	.	.	PUNCT
app01-9407	206	1	https://doi.org/10.5566/ias.v28.p93-102	https://doi.org/10.5566/ias.v28.p93-102	PRON
app01-9407	207	1	[	[	X
app01-9407	207	2	11	11	NUM
app01-9407	207	3	]	]	PUNCT
app01-9407	207	4	j.	j.	PROPN
app01-9407	207	5	kastner	kastner	PROPN
app01-9407	207	6	,	,	PUNCT
app01-9407	207	7	b.	b.	PROPN
app01-9407	207	8	plank	plank	PROPN
app01-9407	207	9	,	,	PUNCT
app01-9407	207	10	d.	d.	PROPN
app01-9407	207	11	salaberger	salaberger	PROPN
app01-9407	207	12	,	,	PUNCT
app01-9407	207	13	j.	j.	PROPN
app01-9407	207	14	sekelja	sekelja	PROPN
app01-9407	207	15	.	.	PUNCT
app01-9407	208	1	defect	defect	VERB
app01-9407	208	2	and	and	CCONJ
app01-9407	208	3	porosity	porosity	NOUN
app01-9407	208	4	determination	determination	NOUN
app01-9407	208	5	of	of	ADP
app01-9407	208	6	fibre	fibre	NOUN
app01-9407	208	7	reinforced	reinforce	VERB
app01-9407	208	8	polymers	polymer	NOUN
app01-9407	208	9	by	by	ADP
app01-9407	208	10	x	x	NOUN
app01-9407	208	11	-	-	NOUN
app01-9407	208	12	ray	ray	NOUN
app01-9407	208	13	computed	compute	VERB
app01-9407	208	14	tomography	tomography	NOUN
app01-9407	208	15	.	.	PUNCT
app01-9407	209	1	2nd	2nd	ADJ
app01-9407	209	2	international	international	ADJ
app01-9407	209	3	symposium	symposium	NOUN
app01-9407	209	4	on	on	ADP
app01-9407	209	5	ndt	ndt	PROPN
app01-9407	209	6	in	in	ADP
app01-9407	209	7	aerospace	aerospace	NOUN
app01-9407	209	8	2010	2010	NUM
app01-9407	209	9	.	.	PUNCT
app01-9407	210	1	[	[	X
app01-9407	210	2	12	12	NUM
app01-9407	210	3	]	]	X
app01-9407	210	4	g.	g.	PROPN
app01-9407	210	5	rao	rao	PROPN
app01-9407	210	6	,	,	PUNCT
app01-9407	210	7	b.	b.	PROPN
app01-9407	210	8	plank	plank	PROPN
app01-9407	210	9	,	,	PUNCT
app01-9407	210	10	j.	j.	PROPN
app01-9407	210	11	kastner	kastner	PROPN
app01-9407	210	12	.	.	PUNCT
app01-9407	211	1	comparison	comparison	NOUN
app01-9407	211	2	of	of	ADP
app01-9407	211	3	different	different	ADJ
app01-9407	211	4	segmentation	segmentation	NOUN
app01-9407	211	5	methods	method	NOUN
app01-9407	211	6	for	for	ADP
app01-9407	211	7	porosity	porosity	NOUN
app01-9407	211	8	evaluation	evaluation	NOUN
app01-9407	211	9	in	in	ADP
app01-9407	211	10	cfrp	cfrp	NOUN
app01-9407	211	11	-	-	PUNCT
app01-9407	211	12	reference	reference	NOUN
app01-9407	211	13	samples	sample	NOUN
app01-9407	211	14	with	with	ADP
app01-9407	211	15	real	real	ADJ
app01-9407	211	16	porosity	porosity	NOUN
app01-9407	211	17	samples	sample	NOUN
app01-9407	211	18	.	.	PUNCT
app01-9407	212	1	in	in	ADP
app01-9407	212	2	conference	conference	NOUN
app01-9407	212	3	on	on	ADP
app01-9407	212	4	industrial	industrial	ADJ
app01-9407	212	5	computed	compute	VERB
app01-9407	212	6	tomography	tomography	NOUN
app01-9407	212	7	,	,	PUNCT
app01-9407	212	8	pp	pp	ADP
app01-9407	212	9	.	.	PUNCT
app01-9407	213	1	1–8	1–8	X
app01-9407	213	2	.	.	NOUN
app01-9407	213	3	2016	2016	NUM
app01-9407	213	4	.	.	PUNCT
app01-9407	214	1	[	[	X
app01-9407	214	2	13	13	NUM
app01-9407	214	3	]	]	X
app01-9407	214	4	b.	b.	PROPN
app01-9407	214	5	fröhler	fröhler	PROPN
app01-9407	214	6	,	,	PUNCT
app01-9407	214	7	j.	j.	PROPN
app01-9407	214	8	weissenböck	weissenböck	PROPN
app01-9407	214	9	,	,	PUNCT
app01-9407	214	10	m.	m.	NOUN
app01-9407	214	11	schiwarth	schiwarth	NOUN
app01-9407	214	12	,	,	PUNCT
app01-9407	214	13	et	et	PROPN
app01-9407	214	14	al	al	PROPN
app01-9407	214	15	.	.	PROPN
app01-9407	214	16	open_ia	open_ia	PROPN
app01-9407	214	17	:	:	PUNCT
app01-9407	214	18	a	a	DET
app01-9407	214	19	tool	tool	NOUN
app01-9407	214	20	for	for	ADP
app01-9407	214	21	processing	processing	NOUN
app01-9407	214	22	and	and	CCONJ
app01-9407	214	23	visual	visual	ADJ
app01-9407	214	24	analysis	analysis	NOUN
app01-9407	214	25	of	of	ADP
app01-9407	214	26	industrial	industrial	ADJ
app01-9407	214	27	computed	compute	VERB
app01-9407	214	28	tomography	tomography	NOUN
app01-9407	214	29	datasets	dataset	NOUN
app01-9407	214	30	.	.	PUNCT
app01-9407	215	1	journal	journal	PROPN
app01-9407	215	2	of	of	ADP
app01-9407	215	3	open	open	ADJ
app01-9407	215	4	source	source	NOUN
app01-9407	215	5	software	software	NOUN
app01-9407	215	6	4(35):1185	4(35):1185	NOUN
app01-9407	215	7	,	,	PUNCT
app01-9407	215	8	2019	2019	NUM
app01-9407	215	9	.	.	PUNCT
app01-9407	216	1	https://doi.org/10.21105/joss.01185	https://doi.org/10.21105/joss.01185	ADJ
app01-9407	217	1	[	[	X
app01-9407	217	2	14	14	NUM
app01-9407	217	3	]	]	PUNCT
app01-9407	217	4	m.	m.	NOUN
app01-9407	217	5	sezgin	sezgin	PROPN
app01-9407	217	6	,	,	PUNCT
app01-9407	217	7	b.	b.	PROPN
app01-9407	217	8	sankur	sankur	PROPN
app01-9407	217	9	.	.	PUNCT
app01-9407	218	1	survey	survey	NOUN
app01-9407	218	2	over	over	ADP
app01-9407	218	3	image	image	NOUN
app01-9407	218	4	thresholding	thresholde	VERB
app01-9407	218	5	techniques	technique	NOUN
app01-9407	218	6	and	and	CCONJ
app01-9407	218	7	quantitative	quantitative	ADJ
app01-9407	218	8	performance	performance	NOUN
app01-9407	218	9	evaluation	evaluation	NOUN
app01-9407	218	10	.	.	PUNCT
app01-9407	219	1	journal	journal	NOUN
app01-9407	219	2	of	of	ADP
app01-9407	219	3	electronic	electronic	ADJ
app01-9407	219	4	imaging	imaging	NOUN
app01-9407	219	5	13:146–168	13:146–168	PROPN
app01-9407	219	6	,	,	PUNCT
app01-9407	219	7	2004	2004	NUM
app01-9407	219	8	.	.	PUNCT
app01-9407	220	1	https://doi.org/10.1117/1.1631315	https://doi.org/10.1117/1.1631315	PROPN
app01-9407	220	2	[	[	X
app01-9407	220	3	15	15	NUM
app01-9407	220	4	]	]	X
app01-9407	220	5	n.	n.	NOUN
app01-9407	220	6	otsu	otsu	PROPN
app01-9407	220	7	.	.	PUNCT
app01-9407	221	1	a	a	DET
app01-9407	221	2	threshold	threshold	NOUN
app01-9407	221	3	selection	selection	NOUN
app01-9407	221	4	method	method	NOUN
app01-9407	221	5	from	from	ADP
app01-9407	221	6	gray	gray	ADJ
app01-9407	221	7	-	-	PUNCT
app01-9407	221	8	level	level	NOUN
app01-9407	221	9	histograms	histogram	NOUN
app01-9407	221	10	.	.	PUNCT
app01-9407	222	1	ieee	ieee	NOUN
app01-9407	222	2	transactions	transaction	NOUN
app01-9407	222	3	on	on	ADP
app01-9407	222	4	systems	system	NOUN
app01-9407	222	5	,	,	PUNCT
app01-9407	222	6	man	man	NOUN
app01-9407	222	7	,	,	PUNCT
app01-9407	222	8	and	and	CCONJ
app01-9407	222	9	cybernetics	cybernetic	NOUN
app01-9407	222	10	9(1):62–66	9(1):62–66	NUM
app01-9407	222	11	,	,	PUNCT
app01-9407	222	12	1979	1979	NUM
app01-9407	222	13	.	.	PUNCT
app01-9407	223	1	https://doi.org/10.1109/tsmc.1979.4310076	https://doi.org/10.1109/tsmc.1979.4310076	PROPN
app01-9407	224	1	[	[	X
app01-9407	224	2	16	16	NUM
app01-9407	224	3	]	]	PUNCT
app01-9407	224	4	a.	a.	NOUN
app01-9407	224	5	reh	reh	PROPN
app01-9407	224	6	,	,	PUNCT
app01-9407	224	7	b.	b.	PROPN
app01-9407	224	8	plank	plank	PROPN
app01-9407	224	9	,	,	PUNCT
app01-9407	224	10	j.	j.	PROPN
app01-9407	224	11	kastner	kastner	PROPN
app01-9407	224	12	,	,	PUNCT
app01-9407	224	13	et	et	PROPN
app01-9407	224	14	al	al	PROPN
app01-9407	224	15	.	.	PROPN
app01-9407	224	16	porosity	porosity	NOUN
app01-9407	224	17	maps	map	NOUN
app01-9407	224	18	–	–	PUNCT
app01-9407	224	19	interactive	interactive	ADJ
app01-9407	224	20	exploration	exploration	NOUN
app01-9407	224	21	and	and	CCONJ
app01-9407	224	22	visual	visual	ADJ
app01-9407	224	23	analysis	analysis	NOUN
app01-9407	224	24	of	of	ADP
app01-9407	224	25	porosity	porosity	NOUN
app01-9407	224	26	in	in	ADP
app01-9407	224	27	carbon	carbon	NOUN
app01-9407	224	28	fiber	fiber	NOUN
app01-9407	224	29	reinforced	reinforce	VERB
app01-9407	224	30	polymers	polymer	NOUN
app01-9407	224	31	.	.	PUNCT
app01-9407	225	1	computer	computer	NOUN
app01-9407	225	2	graphics	graphic	NOUN
app01-9407	225	3	forum	forum	PROPN
app01-9407	225	4	31:1185–1194	31:1185–1194	NUM
app01-9407	225	5	,	,	PUNCT
app01-9407	225	6	2012	2012	NUM
app01-9407	225	7	.	.	PUNCT
app01-9407	226	1	https	https	NOUN
app01-9407	226	2	:	:	PUNCT
app01-9407	227	1	//doi.org/10.1111	//doi.org/10.1111	PROPN
app01-9407	227	2	/	/	SYM
app01-9407	227	3	j.1467	j.1467	PROPN
app01-9407	227	4	-	-	PUNCT
app01-9407	227	5	8659.2012.03111.x	8659.2012.03111.x	NOUN
app01-9407	228	1	[	[	X
app01-9407	228	2	17	17	NUM
app01-9407	228	3	]	]	X
app01-9407	228	4	o.	o.	NOUN
app01-9407	228	5	ronneberger	ronneberger	NOUN
app01-9407	228	6	,	,	PUNCT
app01-9407	228	7	p.	p.	NOUN
app01-9407	228	8	fischer	fischer	PROPN
app01-9407	228	9	,	,	PUNCT
app01-9407	228	10	t.	t.	PROPN
app01-9407	228	11	brox	brox	PROPN
app01-9407	228	12	.	.	PUNCT
app01-9407	229	1	u	u	NOUN
app01-9407	229	2	-	-	NOUN
app01-9407	229	3	net	net	ADJ
app01-9407	229	4	:	:	PUNCT
app01-9407	229	5	convolutional	convolutional	ADJ
app01-9407	229	6	networks	network	NOUN
app01-9407	229	7	for	for	ADP
app01-9407	229	8	biomedical	biomedical	ADJ
app01-9407	229	9	image	image	NOUN
app01-9407	229	10	segmentation	segmentation	NOUN
app01-9407	229	11	.	.	PUNCT
app01-9407	230	1	in	in	ADP
app01-9407	230	2	international	international	ADJ
app01-9407	230	3	conference	conference	NOUN
app01-9407	230	4	on	on	ADP
app01-9407	230	5	medical	medical	ADJ
app01-9407	230	6	image	image	NOUN
app01-9407	230	7	computing	computing	NOUN
app01-9407	230	8	and	and	CCONJ
app01-9407	230	9	computer	computer	NOUN
app01-9407	230	10	-	-	PUNCT
app01-9407	230	11	assisted	assist	VERB
app01-9407	230	12	intervention	intervention	NOUN
app01-9407	230	13	(	(	PUNCT
app01-9407	230	14	miccai	miccai	PROPN
app01-9407	230	15	)	)	PUNCT
app01-9407	230	16	,	,	PUNCT
app01-9407	230	17	pp	pp	ADV
app01-9407	230	18	.	.	PUNCT
app01-9407	231	1	234–241	234–241	NUM
app01-9407	231	2	.	.	PUNCT
app01-9407	232	1	springer	springer	NOUN
app01-9407	232	2	,	,	PUNCT
app01-9407	232	3	cham	cham	PROPN
app01-9407	232	4	,	,	PUNCT
app01-9407	232	5	2015	2015	NUM
app01-9407	232	6	.	.	PUNCT
app01-9407	233	1	https://doi.org/10.1007/978-3-319-24574-4_28	https://doi.org/10.1007/978-3-319-24574-4_28	PROPN
app01-9407	234	1	[	[	X
app01-9407	234	2	18	18	NUM
app01-9407	234	3	]	]	PUNCT
app01-9407	234	4	ö.	ö.	PROPN
app01-9407	234	5	çiçek	çiçek	PROPN
app01-9407	234	6	,	,	PUNCT
app01-9407	234	7	a.	a.	NOUN
app01-9407	234	8	abdulkadir	abdulkadir	PROPN
app01-9407	234	9	,	,	PUNCT
app01-9407	234	10	s.	s.	PROPN
app01-9407	234	11	s.	s.	PROPN
app01-9407	234	12	lienkamp	lienkamp	PROPN
app01-9407	234	13	,	,	PUNCT
app01-9407	234	14	et	et	PROPN
app01-9407	234	15	al	al	PROPN
app01-9407	234	16	.	.	PROPN
app01-9407	235	1	3d	3d	NUM
app01-9407	235	2	u	u	NOUN
app01-9407	235	3	-	-	NOUN
app01-9407	235	4	net	net	ADJ
app01-9407	235	5	:	:	PUNCT
app01-9407	235	6	learning	learn	VERB
app01-9407	235	7	dense	dense	ADJ
app01-9407	235	8	volumetric	volumetric	NOUN
app01-9407	235	9	segmentation	segmentation	NOUN
app01-9407	235	10	from	from	ADP
app01-9407	235	11	sparse	sparse	ADJ
app01-9407	235	12	annotation	annotation	NOUN
app01-9407	235	13	.	.	PUNCT
app01-9407	236	1	in	in	ADP
app01-9407	236	2	international	international	ADJ
app01-9407	236	3	conference	conference	NOUN
app01-9407	236	4	on	on	ADP
app01-9407	236	5	medical	medical	ADJ
app01-9407	236	6	image	image	NOUN
app01-9407	236	7	computing	computing	NOUN
app01-9407	236	8	and	and	CCONJ
app01-9407	236	9	computer	computer	NOUN
app01-9407	236	10	-	-	PUNCT
app01-9407	236	11	assisted	assist	VERB
app01-9407	236	12	intervention	intervention	NOUN
app01-9407	236	13	(	(	PUNCT
app01-9407	236	14	miccai	miccai	PROPN
app01-9407	236	15	)	)	PUNCT
app01-9407	236	16	,	,	PUNCT
app01-9407	236	17	pp	pp	ADP
app01-9407	236	18	.	.	PUNCT
app01-9407	237	1	424–432	424–432	NUM
app01-9407	237	2	.	.	PUNCT
app01-9407	237	3	springer	springer	NOUN
app01-9407	237	4	,	,	PUNCT
app01-9407	237	5	cham	cham	PROPN
app01-9407	237	6	,	,	PUNCT
app01-9407	237	7	2016	2016	NUM
app01-9407	237	8	.	.	PUNCT
app01-9407	238	1	https://doi.org/10.1007/978-3-319-46723-849	https://doi.org/10.1007/978-3-319-46723-849	ADP
app01-9407	238	2	[	[	X
app01-9407	238	3	19	19	NUM
app01-9407	238	4	]	]	X
app01-9407	238	5	j.	j.	PROPN
app01-9407	238	6	walsh	walsh	PROPN
app01-9407	238	7	,	,	PUNCT
app01-9407	238	8	a.	a.	PROPN
app01-9407	238	9	othmani	othmani	PROPN
app01-9407	238	10	,	,	PUNCT
app01-9407	238	11	m.	m.	NOUN
app01-9407	238	12	jain	jain	PROPN
app01-9407	238	13	,	,	PUNCT
app01-9407	238	14	s.	s.	PROPN
app01-9407	238	15	dev	dev	PROPN
app01-9407	238	16	.	.	PUNCT
app01-9407	239	1	using	use	VERB
app01-9407	239	2	u	u	ADJ
app01-9407	239	3	-	-	ADJ
app01-9407	239	4	net	net	ADJ
app01-9407	239	5	network	network	NOUN
app01-9407	239	6	for	for	ADP
app01-9407	239	7	efficient	efficient	ADJ
app01-9407	239	8	brain	brain	NOUN
app01-9407	239	9	tumor	tumor	NOUN
app01-9407	239	10	segmentation	segmentation	NOUN
app01-9407	239	11	in	in	ADP
app01-9407	239	12	mri	mri	NOUN
app01-9407	239	13	images	image	NOUN
app01-9407	239	14	.	.	PUNCT
app01-9407	240	1	healthcare	healthcare	NOUN
app01-9407	240	2	analytics	analytic	NOUN
app01-9407	240	3	2:100098	2:100098	NUM
app01-9407	240	4	,	,	PUNCT
app01-9407	240	5	2022	2022	NUM
app01-9407	240	6	.	.	PUNCT
app01-9407	241	1	https://doi.org/https	https://doi.org/https	PUNCT
app01-9407	241	2	:	:	PUNCT
app01-9407	242	1	//doi.org/10.1016	//doi.org/10.1016	X
app01-9407	242	2	/	/	SYM
app01-9407	242	3	j.health.2022.100098	j.health.2022.100098	PROPN
app01-9407	242	4	[	[	X
app01-9407	242	5	20	20	NUM
app01-9407	242	6	]	]	X
app01-9407	242	7	g.	g.	PROPN
app01-9407	242	8	van	van	PROPN
app01-9407	242	9	rossum	rossum	PROPN
app01-9407	242	10	,	,	PUNCT
app01-9407	242	11	p.	p.	PROPN
app01-9407	242	12	d.	d.	PROPN
app01-9407	242	13	team	team	PROPN
app01-9407	242	14	.	.	PUNCT
app01-9407	243	1	python	python	NOUN
app01-9407	243	2	3.6	3.6	NUM
app01-9407	243	3	language	language	NOUN
app01-9407	243	4	reference	reference	NOUN
app01-9407	243	5	.	.	PUNCT
app01-9407	244	1	samurai	samurai	NOUN
app01-9407	244	2	media	medium	NOUN
app01-9407	244	3	limited	limit	VERB
app01-9407	244	4	,	,	PUNCT
app01-9407	244	5	2016	2016	NUM
app01-9407	244	6	.	.	PUNCT
app01-9407	245	1	[	[	X
app01-9407	245	2	21	21	NUM
app01-9407	245	3	]	]	X
app01-9407	245	4	keras	keras	PROPN
app01-9407	245	5	.	.	PUNCT
app01-9407	245	6	f.	f.	PROPN
app01-9407	245	7	chollet	chollet	PROPN
app01-9407	245	8	and	and	CCONJ
app01-9407	245	9	others	other	NOUN
app01-9407	245	10	.	.	PUNCT
app01-9407	246	1	keras	keras	PROPN
app01-9407	246	2	:	:	PUNCT
app01-9407	247	1	deep	deep	ADJ
app01-9407	247	2	learning	learning	NOUN
app01-9407	247	3	for	for	ADP
app01-9407	247	4	humans	human	NOUN
app01-9407	247	5	.	.	PUNCT
app01-9407	248	1	2015	2015	NUM
app01-9407	248	2	.	.	PUNCT
app01-9407	249	1	https://keras.io	https://keras.io	X
app01-9407	249	2	.	.	PUNCT
app01-9407	250	1	[	[	X
app01-9407	250	2	22	22	NUM
app01-9407	250	3	]	]	X
app01-9407	250	4	tensorflow	tensorflow	NOUN
app01-9407	250	5	:	:	PUNCT
app01-9407	250	6	large	large	ADJ
app01-9407	250	7	-	-	PUNCT
app01-9407	250	8	scale	scale	NOUN
app01-9407	250	9	machine	machine	NOUN
app01-9407	250	10	learning	learn	VERB
app01-9407	250	11	on	on	ADP
app01-9407	250	12	heterogeneous	heterogeneous	ADJ
app01-9407	250	13	systems	system	NOUN
app01-9407	250	14	.	.	PUNCT
app01-9407	251	1	2015	2015	NUM
app01-9407	251	2	.	.	PUNCT
app01-9407	252	1	https://tensorflow.org	https://tensorflow.org	NUM
app01-9407	252	2	.	.	PUNCT
app01-9407	253	1	[	[	X
app01-9407	253	2	23	23	NUM
app01-9407	253	3	]	]	X
app01-9407	253	4	onnx	onnx	PROPN
app01-9407	253	5	.	.	PUNCT
app01-9407	254	1	open	open	ADJ
app01-9407	254	2	neural	neural	ADJ
app01-9407	254	3	network	network	NOUN
app01-9407	254	4	exchange	exchange	NOUN
app01-9407	254	5	:	:	PUNCT
app01-9407	254	6	the	the	DET
app01-9407	254	7	open	open	ADJ
app01-9407	254	8	standard	standard	NOUN
app01-9407	254	9	for	for	ADP
app01-9407	254	10	machine	machine	NOUN
app01-9407	254	11	learning	learning	NOUN
app01-9407	254	12	interoperability	interoperability	NOUN
app01-9407	254	13	.	.	PUNCT
app01-9407	255	1	2017	2017	NUM
app01-9407	255	2	.	.	PUNCT
app01-9407	256	1	https://onnx.ai	https://onnx.ai	NOUN
app01-9407	256	2	.	.	PUNCT
app01-9407	257	1	[	[	X
app01-9407	257	2	24	24	NUM
app01-9407	257	3	]	]	PUNCT
app01-9407	257	4	autonomio	autonomio	NOUN
app01-9407	257	5	talos	talos	PROPN
app01-9407	257	6	[	[	X
app01-9407	257	7	computer	computer	NOUN
app01-9407	257	8	software	software	NOUN
app01-9407	257	9	]	]	PUNCT
app01-9407	257	10	.	.	PUNCT
app01-9407	258	1	2020	2020	NUM
app01-9407	258	2	.	.	PUNCT
app01-9407	259	1	http://github.com/autonomio/talos	http://github.com/autonomio/talo	NOUN
app01-9407	259	2	.	.	PUNCT
app01-9407	260	1	[	[	X
app01-9407	260	2	25	25	NUM
app01-9407	260	3	]	]	X
app01-9407	260	4	d.	d.	PROPN
app01-9407	260	5	p.	p.	PROPN
app01-9407	260	6	kingma	kingma	PROPN
app01-9407	260	7	,	,	PUNCT
app01-9407	260	8	j.	j.	PROPN
app01-9407	260	9	ba	ba	PROPN
app01-9407	260	10	.	.	PUNCT
app01-9407	261	1	adam	adam	PROPN
app01-9407	261	2	:	:	PUNCT
app01-9407	261	3	a	a	DET
app01-9407	261	4	method	method	NOUN
app01-9407	261	5	for	for	ADP
app01-9407	261	6	stochastic	stochastic	ADJ
app01-9407	261	7	optimization	optimization	NOUN
app01-9407	261	8	.	.	PUNCT
app01-9407	262	1	in	in	ADP
app01-9407	262	2	international	international	ADJ
app01-9407	262	3	conference	conference	NOUN
app01-9407	262	4	for	for	ADP
app01-9407	262	5	learning	learn	VERB
app01-9407	262	6	representations	representation	NOUN
app01-9407	262	7	.	.	PUNCT
app01-9407	262	8	2017	2017	NUM
app01-9407	262	9	.	.	PUNCT
app01-9407	263	1	[	[	X
app01-9407	263	2	2023	2023	NUM
app01-9407	263	3	-	-	SYM
app01-9407	263	4	06	06	NUM
app01-9407	263	5	-	-	SYM
app01-9407	263	6	30	30	NUM
app01-9407	263	7	]	]	PUNCT
app01-9407	263	8	.	.	PUNCT
app01-9407	264	1	https://doi.org/10.48550/arxiv.1412.6980	https://doi.org/10.48550/arxiv.1412.6980	PROPN
app01-9407	265	1	[	[	X
app01-9407	265	2	26	26	NUM
app01-9407	265	3	]	]	PUNCT
app01-9407	265	4	t.	t.	NOUN
app01-9407	265	5	sørensen	sørensen	NOUN
app01-9407	265	6	.	.	PUNCT
app01-9407	266	1	a	a	DET
app01-9407	266	2	method	method	NOUN
app01-9407	266	3	of	of	ADP
app01-9407	266	4	establishing	establish	VERB
app01-9407	266	5	groups	group	NOUN
app01-9407	266	6	of	of	ADP
app01-9407	266	7	equal	equal	ADJ
app01-9407	266	8	amplitude	amplitude	NOUN
app01-9407	266	9	in	in	ADP
app01-9407	266	10	plant	plant	NOUN
app01-9407	266	11	sociology	sociology	NOUN
app01-9407	266	12	based	base	VERB
app01-9407	266	13	on	on	ADP
app01-9407	266	14	similarity	similarity	NOUN
app01-9407	266	15	of	of	ADP
app01-9407	266	16	species	specie	NOUN
app01-9407	266	17	and	and	CCONJ
app01-9407	266	18	its	its	PRON
app01-9407	266	19	application	application	NOUN
app01-9407	266	20	to	to	ADP
app01-9407	266	21	analyses	analysis	NOUN
app01-9407	266	22	of	of	ADP
app01-9407	266	23	the	the	DET
app01-9407	266	24	vegetation	vegetation	NOUN
app01-9407	266	25	on	on	ADP
app01-9407	266	26	danish	danish	ADJ
app01-9407	266	27	commons	common	NOUN
app01-9407	266	28	.	.	PUNCT
app01-9407	267	1	kongelige	kongelige	PROPN
app01-9407	267	2	danske	danske	PROPN
app01-9407	267	3	videnskabernes	videnskaberne	NOUN
app01-9407	267	4	selskab	selskab	NOUN
app01-9407	267	5	,	,	PUNCT
app01-9407	267	6	biologiske	biologiske	PROPN
app01-9407	267	7	skrifter	skrifter	PROPN
app01-9407	267	8	pp	pp	PROPN
app01-9407	267	9	.	.	PUNCT
app01-9407	268	1	1–34	1–34	NUM
app01-9407	268	2	,	,	PUNCT
app01-9407	268	3	1948	1948	NUM
app01-9407	268	4	.	.	PUNCT
app01-9407	269	1	93	93	NUM
app01-9407	269	2	https://doi.org/10.1111/cgf.13214	https://doi.org/10.1111/cgf.13214	PROPN
app01-9407	269	3	https://doi.org/10.48550/arxiv.1202.2745	https://doi.org/10.48550/arxiv.1202.2745	PROPN
app01-9407	269	4	https://doi.org/10.48550/arxiv.1412.7755	https://doi.org/10.48550/arxiv.1412.7755	PROPN
app01-9407	269	5	https://doi.org/10.1109/access.2017.2788044	https://doi.org/10.1109/access.2017.2788044	PROPN
app01-9407	269	6	https://doi.org/10.1080/10589759.2022.2071892	https://doi.org/10.1080/10589759.2022.2071892	PROPN
app01-9407	270	1	https://doi.org/10.58286/27716	https://doi.org/10.58286/27716	NOUN
app01-9407	270	2	https://doi.org/10.2307/2346830	https://doi.org/10.2307/2346830	ADP
app01-9407	270	3	https://doi.org/10.5566/ias.v28.p93-102	https://doi.org/10.5566/ias.v28.p93-102	PRON
app01-9407	270	4	https://doi.org/10.21105/joss.01185	https://doi.org/10.21105/joss.01185	ADJ
app01-9407	270	5	https://doi.org/10.1117/1.1631315	https://doi.org/10.1117/1.1631315	PROPN
app01-9407	270	6	https://doi.org/10.1109/tsmc.1979.4310076	https://doi.org/10.1109/tsmc.1979.4310076	PROPN
app01-9407	270	7	https://doi.org/10.1111/j.1467-8659.2012.03111.x	https://doi.org/10.1111/j.1467-8659.2012.03111.x	NOUN
app01-9407	270	8	https://doi.org/10.1111/j.1467-8659.2012.03111.x	https://doi.org/10.1111/j.1467-8659.2012.03111.x	NOUN
app01-9407	270	9	https://doi.org/10.1007/978-3-319-24574-4_28	https://doi.org/10.1007/978-3-319-24574-4_28	PROPN
app01-9407	271	1	https://doi.org/10.1007/978-3-319-46723-8	https://doi.org/10.1007/978-3-319-46723-8	NOUN
app01-9407	271	2	49	49	NUM
app01-9407	271	3	https://doi.org/https://doi.org/10.1016/j.health.2022.100098	https://doi.org/https://doi.org/10.1016/j.health.2022.100098	NOUN
app01-9407	271	4	https://doi.org/https://doi.org/10.1016/j.health.2022.100098	https://doi.org/https://doi.org/10.1016/j.health.2022.100098	NOUN
app01-9407	271	5	https://keras.io	https://keras.io	NUM
app01-9407	271	6	https://tensorflow.org	https://tensorflow.org	PROPN
app01-9407	271	7	https://onnx.ai	https://onnx.ai	NUM
app01-9407	271	8	http://github.com/autonomio/talos	http://github.com/autonomio/talos	NOUN
app01-9407	271	9	https://doi.org/10.48550/arxiv.1412.6980	https://doi.org/10.48550/arxiv.1412.6980	PROPN
app01-9407	271	10	acta	acta	PROPN
app01-9407	271	11	polytechnica	polytechnica	PROPN
app01-9407	271	12	ctu	ctu	NOUN
app01-9407	271	13	proceedings	proceeding	NOUN
app01-9407	271	14	42:87–93	42:87–93	PROPN
app01-9407	271	15	,	,	PUNCT
app01-9407	271	16	2023	2023	NUM
app01-9407	271	17	1	1	NUM
app01-9407	271	18	introduction	introduction	NOUN
app01-9407	271	19	2	2	NUM
app01-9407	271	20	background	background	NOUN
app01-9407	271	21	and	and	CCONJ
app01-9407	271	22	related	relate	VERB
app01-9407	271	23	work	work	NOUN
app01-9407	271	24	2.1	2.1	NUM
app01-9407	271	25	thresholding	thresholding	NOUN
app01-9407	271	26	-	-	PUNCT
app01-9407	271	27	based	base	VERB
app01-9407	271	28	segmentation	segmentation	NOUN
app01-9407	271	29	2.2	2.2	NUM
app01-9407	271	30	convolutional	convolutional	ADJ
app01-9407	271	31	neural	neural	ADJ
app01-9407	271	32	networks	network	NOUN
app01-9407	271	33	3	3	NUM
app01-9407	271	34	data	datum	NOUN
app01-9407	271	35	characteristic	characteristic	ADJ
app01-9407	271	36	4	4	NUM
app01-9407	271	37	methods	method	NOUN
app01-9407	271	38	4.1	4.1	NUM
app01-9407	271	39	modification	modification	NOUN
app01-9407	271	40	of	of	ADP
app01-9407	271	41	3d	3d	NUM
app01-9407	271	42	u	u	NOUN
app01-9407	271	43	-	-	PROPN
app01-9407	271	44	net	net	ADJ
app01-9407	271	45	4.2	4.2	NUM
app01-9407	271	46	data	datum	NOUN
app01-9407	271	47	pre	pre	ADJ
app01-9407	271	48	-	-	ADJ
app01-9407	271	49	processing	processing	ADJ
app01-9407	271	50	,	,	PUNCT
app01-9407	271	51	training	training	NOUN
app01-9407	271	52	and	and	CCONJ
app01-9407	271	53	testing	test	VERB
app01-9407	271	54	5	5	NUM
app01-9407	271	55	results	result	NOUN
app01-9407	271	56	and	and	CCONJ
app01-9407	271	57	discussion	discussion	NOUN
app01-9407	271	58	5.1	5.1	NUM
app01-9407	271	59	sub	sub	ADJ
app01-9407	271	60	-	-	ADJ
app01-9407	271	61	volume	volume	NOUN
app01-9407	271	62	prediction	prediction	NOUN
app01-9407	271	63	and	and	CCONJ
app01-9407	271	64	visualization	visualization	NOUN
app01-9407	271	65	5.2	5.2	NUM
app01-9407	271	66	prediction	prediction	NOUN
app01-9407	271	67	and	and	CCONJ
app01-9407	271	68	visualization	visualization	NOUN
app01-9407	271	69	of	of	ADP
app01-9407	271	70	sample	sample	NOUN
app01-9407	271	71	2	2	NUM
app01-9407	271	72	5.3	5.3	NUM
app01-9407	271	73	prediction	prediction	NOUN
app01-9407	271	74	and	and	CCONJ
app01-9407	271	75	visualization	visualization	NOUN
app01-9407	271	76	of	of	ADP
app01-9407	271	77	sample	sample	NOUN
app01-9407	271	78	3	3	NUM
app01-9407	271	79	6	6	NUM
app01-9407	271	80	conclusions	conclusion	NOUN
app01-9407	271	81	acknowledgements	acknowledgement	NOUN
app01-9407	271	82	references	reference	NOUN