I.G. Bakulin1, I.A. Rasmagina2, G.A. Mashevskiy3, N.M. Shelyakina4, G.F. Arutyunyan5

1, 2, 4, 5 North-Western State Medical university n.a. I. Mechnikov (St. Petersburg, Russia)
3 St. Petersburg State Electrotechnical University “LETI” (St. Petersburg, Russia)
1 igbakulin@yandex.ru, 2 irenerasmagina@gmail.com, 3 Aniket@list.ru, 4 n.sheliakina@gmail.com, 5 grant.arutyunyan117@yandex.ru

Inflammatory bowel diseases (IBD) are severe pathologies that often lead to a significant decline in quality of life and disability. Diagnosis of IBD is frequently delayed, resulting in postponed initiation of targeted therapy and worse outcomes.

The aim of the study was to develop a clinical decision support system (CDSS) for the diagnosis and differential diagnosis of IBD based on the analysis of clinical, laboratory, endoscopic, and morphological data.

Building upon previously developed individual artificial neural networks (ANNs) for IBD detection and differentiation, integrated ANNs were developed using a Bayesian belief network. Model #1 differentiated IBD by simultaneously considering the outputs of all ANNs, while Model #2 employed a two-stage approach: initial IBD detection based on clinical and laboratory data, followed by endoscopic and morphological analysis if IBD was suspected. The models were tested on 85 patients (24 (28.2%) with irritable bowel syndrome (IBS), 25 (29.4%) with Crohn’s disease of the colon, and 36 (42.4%) with ulcerative colitis, all of whom had complete clinical and laboratory data, digital video colonoscopy images, and morphological examination results. Validation showed that Model #1 achieved an accuracy of 97%, while Model #2 reached 86%.

The practical significance: the developed models can serve as the foundation for general practitioner`s and gastroenterologist’s decision-support system, improving the speed and accuracy of IBD verification.

Bakulin I.G., Rasmagina I.A., Mashevsky G.A., Shelyakina N.M., Arutyunyan G.F. Differential diagnostics of inflammatory bowel diseases using artificial intelligence: analysis of multimodal medical data. Biomedicine Radioengineering. 2025. V. 28. № 5. P. 69–73. DOI: https:// doi.org/10.18127/j15604136-202505-14 (In Russian)

1. Bolezn' Krona. Klinicheskie rekomendacii RF, 2024 (Rossiya). Tekst: elektronnyj. Med Element. URL: https://diseases. medelement.com/disease/%D0%B1%D0%BE%D0%BB%D0%B5%D0%B7%D0%BD%D1%8C-%D0%BA%D1%80%D0%BE% D0%BD%D0%B0-%D0%BA%D1%80-%D1%80%D1%84-2024/18085. Data obrashcheniya: 31.07.2025 (In Russian).
2. Yazvennyj kolit. Klinicheskie rekomendacii RF, 2024 (Rossiya). Tekst: elektronnyj. Med Element. URL: https://diseases. medelement.com/disease/%D1%8F%D0%B7%D0%B2%D0%B5%D0%BD%D0%BD%D1%8B%D0%B9%D0%BA%D0%BE%D 0%BB%D0%B8%D1%82%D0%BA%D0%BFD1%80%D1%842024/18016. Data obrashcheniya: 31.07.2025 (In Russian).
3. Lv H., Li H.Y., Zhang H.N., Liu Y. Delayed diagnosis in inflammatory bowel disease: Time to consider solutions. World J. Gastroenterol. 2024. V. 30(35). P. 3954–3958. doi:10.3748/wjg.v30.i35.3954.
4. Bakulin I.G., Skalinskaya M.I., Skazyvaeva E.V. Severo-Zapadnyj registr pacientov s vospalitel'nymi zabolevaniyami kishechnika: dostizheniya i uroki. Koloproktologiya. 2022. № 21(1(79)). S. 37–49. DOI:10.33878/2073-7556-2022-21-1-37-49 (In Russian).
5. Burisch J., Pedersen N., Čuković-Čavka S. et al. East-West gradient in the incidence of inflammatory bowel disease in Europe: the ECCO-EpiCom inception cohort. Gut. 2014. V. 63. № 4. P. 588–597. DOI: 10.1136/gutjnl-2013-304636.
6. Kawamoto A., Takenaka K., Okamoto R., Watanabe M., Ohtsuka K. Systematic review of artificial intelligence-based image diagnosis for inflammatory bowel disease. Dig Endosc off J. Jpn. Gastroenterol. Endosc Soc. 2022. V. 34(7). P. 1311–1319. DOI:10.1111/den.14334.
7. Bakulin I.G., Rasmagina I.A., Skalinskaya M.I., Mashevskij G.A., Shelyakina N.M. Primenenie iskusstvennogo intellekta dlya analiza endoskopicheskih izobrazhenij pri vospalitel'nyh zabolevaniyah kishechnika. Terapiya. 2022. № 8(7(59)). S. 7–14. DOI:10.18565/therapy.2022.7.7-14 (In Russian).
8. Rasmagina I.A., Bakulin I.G., Stambolcyan V.Sh., Mashevskij G.A., Shelyakina N.M. Kliniko-laboratornye markery predtestovoj veroyatnosti vospalitel'nyh zabolevanij kishechnika. Koloproktologiya. 2023. № 22(1(83)). S. 91–98. DOI:10.33878/2073-7556-2023-22-1-91-98 (In Russian).
9. Bakulin I.G., Rasmagina I.A., Deev R.V., Mashevskij G.A., Shelyakina N.M. Analiz gistologicheskih izobrazhenij bioptatov slizistoj tolstoj kishki u pacientov s vospalitel'nymi zabolevaniyami kishechnika pri pomoshchi iskusstvennoj nejronnoj seti. Medicinskij Vestnik MVD. 2023. № 124(3(124)). S. 21–24. DOI:10.52341/20738080_2023_124_3_21 (In Russian).
10. Bakulin I.G., Rasmagina I.A., Mashevskij G.A., Shelyakina N.M., Arutyunyan G.F. Iskusstvennyj intellekt v diagnostike vospalitel'nyh zabolevanij kishechnika: fokus na kliniko-laboratornye pokazateli. Koloproktologiya. 2025. № 24(2(92)). S. 33–41. DOI:10.33878/2073-7556-2025-24-2-33-41 (In Russian).
11. Lacy B.E., Patel N.K. Rome Criteria and a Diagnostic Approach to Irritable Bowel Syndrome. J. Clin. Med. 2017. V. 6(11). P. 99. DOI:10.3390/jcm6110099.