ISDS Annual Conference Proceedings 2017. This is an Open Access article distributed under the terms of the Creative Commons Attribution-
Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution,  
and reproduction in any medium, provided the original work is properly cited.

ISDS 2016 Conference Abstracts

Agent-based investigation of sexually transmitted 
infection
Dmytro Chumachenko*1 and Tetyana Chumachenko2
1Informatics Department, National Aerospace University “Kharkiv Aviation Institute”, Kharkiv, Ukraine; 2Kharkiv National Medical 
University, Kharkiv, Ukraine

Objective
To develop agent-based model of sexually transmitted infections 

spreading by example of Syphilis and its analysis.

Introduction
Every year nearly 12 million new cases of syphilis in the world are 

registered. Currently, in many countries of the world the stabilization 
or even reduction of the incidence of syphilis is marked, but this does 
not apply to Ukraine. The current stage of development of the syphilis 
problem in Ukraine is characterized by not only high morbidity, but 
also the fact that in the overwhelming number of cases, we are talking 
about the latent forms and atypical manifestations of the disease and 
resistance to therapy [1].

Preventive and prophylactic measures are important in maintaining 
the public health. Predicting the dynamics of disease spreading allows 
developing appropriate countermeasures and ensuring rational use 
of human and material resources. Qualitative forecast of syphilis 
spreading is possible to implement by means of mathematical 
modeling.

Methods
Deterministic analytical models that are most common in 

epidemiological studies do not take into account the dynamic and 
stochastic nature of epidemics. Agent-based simulation approach 
to modeling allows fixing these shortcomings. It allows conveying 
the social structure of simulated system by the most natural and easy 
way. Each agent has individual state variables and rules of behavior 
that allows detailing the model very deeply. Therefore there is no 
need to describe the complex system of mathematical formulas 
and probability of the dynamics of the epidemic process is defined 
parametrically. The NetLogo software has been used for the program 
realization of the developed model.

Results
The model of morbidity by syphilis spreading has been developed 

by the tradition SIR model expansion. Thus, agents can be in following 
states: S (Susceptible) for health people, IP (Infected Primary) for 
infected people who stay in primary stage and can transmit the 
infection by direct sexual contact with susceptible person, IS (Infected 
Secondary) for infected people who stay in secondary stage and have 
also infectious skin lesions, IL (Infected Latent) for infected people 
who stay in latent stage and change its contagious rate from early 
latent syphilis to late latent syphilis, IT (Infected Tertiary) for infected 
people who stay in tertiary stage and transmit the infection partially, 
and R (Recovered) for people who are recovered from the infection. 
Infecting of agents in the model depends on the number and state 
of agents and the stage of infected agent’s disease. Also, in order 
to correctly determine the intensity of contacts with other agents 
different age groups of agents have been highlighted in the model. 
Screen form of developed agent-based model of syphilis spreading 
is shown in Figure.

The transmission between agent’s states are defined by probabilistic 
way and depends on features of particular states as well as different 

factors, such as coupling tendency, condom use, commitment, test 
frequency etc.

The analysis of experiments under developed model has shown 
that the most influencing factor in the reduction in the percentage of 
patients is frequency of checks on the disease and isolation of patients, 
the second most important factor is constancy of sexual partners, the 
third is the use of condoms, and finally, the fourth is the number of 
exchangeable partners.

Conclusions
The agent-based model of syphilis spreading has been developed. 

The model allows forecasting the morbidity by infection and 
analyzing the disease by changing the initial data. All data has been 
checked by the factual statistics on the syphilis incidence in Kharkiv 
region (Ukraine) from 1975 to 2015 years. The simulation results 
allow us determining the direction of prevention of syphilis treatment 
and the main factors in reducing morbidity. As is evident from the 
simulation results, social factors take precedence over the health 
care that gives grounds for advocacy in health policy among the 
population, especially the youth. Developed model can be configured 
for other sexually transmitted infections by changing the disease 
transition rules.

Figure. The main panel of simulation management and graphic visualization.

Keywords
agent-based simulation; public health; syphilis; mathematical 
modeling; epidemic process

References
1. Herbert L. J., Middleton S. I. An estimate of syphilis incidence in

Eastern Europe. Journal of Global Health, 2012, 2 (1), pp. 1-7.

*Dmytro Chumachenko
E-mail: dichumachenko@gmail.com

Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org * 9(1):e60, 2017