Medical decision-making support system based on bayesian networks in medical diagnostics

Abstract

Modern level of medical science development provides the attending doctor with thousands of various diagnostic and therapeutic techniques as well as medicines. Practically applicating them, the clinician has to take into account a variety of factors: indications and contraindications of the method or modalities of treatment, characteristics of the patient and the course of the disease, compatibility or strengthening of the influence of certain examination methods, medications on each other, individual drug intolerance and contraindications in the patient. It becomes more difficult to keep all this in memory and make error-free, correct and timely decisions. Moreover, the situation is rapidly aggravated by the fact that the volume of knowledge in medicine is growing incrementally, and the time for a doctor to make an appropriate decision when making a diagnosis does not increase. In this regard, the question of creating a system that will minimize the time for a doctor to make a decision on the presence of a particular disease arises.

AIM: To develop a medical decision-making support system based on Bayesian networks in diagnosing patients.

RESULTS: A variant of the medical decision support system for the diagnosis of cold, flu and coronavirus are considered. A Bayesian network model using the GeNIe Academic software is proposed. The results of the percentages of possible diseases of the patient based on the existing symptoms are obtained.

CONCLUSION: The approach to the decision support system construction considered in the article is intended to assist doctors in making a diagnosis to a patient based on his anamnesis. It should be noted that the constructed Bayesian network can be modified by adding other symptoms with their conditional probabilities and adjusting the existing ones after expert judgment (6 figs, bibliography: 6 refs).

Keywords

Bayesian network / cold / flu / conditional probability / coronavirus / decision making support system / unconditional probability

References

[1] Gusev AV, Zarubina TV. Support of medical decision-making in medical information systems of a medical organization. Doctor and information technologies. 2017;(2):60–72. (In Russ.)

[2] Гусев А.В., Зарубина Т.В. Поддержка принятия врачебных решений в медицинских информационных системах медицинской организации // Врач и информационные технологии. 2017. № 2. С. 60–72.

[3] Zvyagin LS. Bayesian Network Method and Key Aspects of Bayesian Modeling. Proceedings of the International Conference on Soft Computing and Measurements. 2019;1:30–34. (In Russ.)

[4] Звягин Л.С. Метод байесовских сетей и ключевые аспекты байесовского моделирования // Материалы международной конференции по мягким вычислениям и измерениям. 2019. Т. 1. С. 30–34.

[5] Ayvazyan SA. Bayesian approach in econometric analysis. Applied econometrics. 2008;1(9):93–108. (In Russ.)

[6] Айвазян С.А. Байесовский подход в эконометрическом анализе // Прикладная эконометрика. 2008. № 1 (9). С. 93–108.

[7] MacKay DJC. Information Theory, Inference, and Learning Algorithms. Cambridge: Published by Cambridge University Press; 2003. 640 p.

[8] MacKay D.J.C. Information Theory, Inference, and Learning Algorithms. Cambridge: Published by Cambridge University Press, 2003. 640 p.

[9] Vetrov DP, Kropotov DA. Algorithms for choosing models and synthesizing collective solutions in classification problems based on the principle of stability. Moscow: URSS Publisher; 2006.

[10] Ветров Д.П., Кропотов Д.А. Алгоритмы выбора моделей и синтеза коллективных решений в задачах классификации, основанные на принципе устойчивости. М.: URSS, 2006.

[11] Prokopchina SV, Fedichkin AI. The use of Bayesian intelligent technologies for the assessment of integral indicators. Collection of reports International Conference on Soft Computing and Measurements. 2006. Р. 20–22. (In Russ.)

[12] Прокопчина С.В., Федичкин А.И. Применение байесовских интеллектуальных технологий для оценки интегральных показателей // Сборник докладов Международной конференции по мягким вычислениям и измерениям. 2006. С. 20–22.

RIGHTS & PERMISSIONS

Levan'kov B.V., Vyborov E.M., Yakovenko N.I.