Аннотация

Цель
To conduct a comprehensive analysis of the current state, challenges and prospects for the application of artificial intelligence technologies in the postgraduate medical education system with an emphasis on quantitative assessment of effectiveness and practical aspects of implementation.
Материал и методы
A systematic literature review was performed in the PubMed database using MeSH terms: «Artificial Intelligence»[MeSH], «Machine Learning»[MeSH], «Deep Learning»[MeSH], «Natural Language Processing»[MeSH] in combination with «Education, Medical, Graduate»[MeSH], «Internship and Residency»[MeSH], «Clinical Competence»[MeSH], «Education, Medical, Continuing»[MeSH]. The search covered publications from September 2020 to September 2025. Of the initially found 1405 publications, after applying inclusion and exclusion criteria, 35 most relevant studies were included in the final analysis, including cohort studies and systematic reviews.
Результаты
Deep learning systems in radiological diagnostics demonstrate significant improvement in key indicators: reduction of image post-processing time by 90.8%, interpretation time by 37.2%, increase in overall diagnostic sensitivity from 73% to 88.46%, especially for microaneurysms 1-3 mm in size (15.8% improvement). Large language models (GPT-4) successfully pass medical licensing exams with results of 71-87%, exceeding passing scores on USMLE (85-87%), MRCS Part A (74%). In comparative studies, ChatGPT-4 surpasses expert specialists in neonatology in 60% of cases for factual accuracy and completeness of answers. Simulation training with real-time AI feedback improves surgical skill acquisition by 67%, reduces time to competency from 45-50 to 20-25 hours, provides skill retention after 3 months at 82% versus 45% with traditional training. Regional experience of Tajikistan shows the effectiveness of implementing automated systems: reduction of ambulance arrival time by 60.2%, decrease in mortality from 2.3% to 1.4%. Economic analysis demonstrates return on investment of 180% by the fifth year of implementation.
Заключение
Artificial intelligence fundamentally transforms postgraduate medical education, providing personalized learning, objective assessment of competencies and safe mastery of critical skills. The main barriers to widespread implementation remain the phenomenon of language model «hallucinations» (15-20% of responses contain inaccuracies), ethical dilemmas of responsibility, regulatory uncertainty and paradoxical negative attitudes towards AI in developed countries (58% in North America) despite objective improvement in learning outcomes across all groups. By 2030, the use of personalized AI trajectories is projected in 85% of postgraduate education programs.

Ключевые слова

artificial intelligence machine learning deep learning postgraduate medical education clinical competence simulation training telemedicine

Полный текст

Скачать статью в PDF

Полная версия статьи в формате PDF

Список литературы

  1. Yilmaz R, Bakhaidar M, Alsayegh A, et al. Real-Time multifaceted artificial intelligence vs In-Person instruction in teaching surgical technical skills: a randomized controlled trial. Sci Rep. 2024;14:14164. doi:10.1038/s41598-024-65716-8.
  2. Zong H, Li J, Wu E, Wu R, Lu J, Shen B. Performance of ChatGPT on Chinese national medical licensing examinations: a five-year examination evaluation study for physicians, pharmacists and nurses. BMC Med Educ. 2024;24(1):143. doi:10.1186/s12909-024-05125-7.
  3. Zong H, Wu R, Cha J, et al. Large Language Models in Worldwide Medical Exams: Platform Development and Comprehensive Analysis. J Med Internet Res. 2024;26:e66114. doi:10.2196/66114.
  4. Kung TH, Cheatham M, Medenilla A, et al. Performance of ChatGPT on USMLE: potential for AI-assisted medical education using large language models. PLOS Digit Health. 2023;2(2):e0000198. doi:10.1371/journal.pdig.0000198.
  5. Yiu A, Lam K. Performance of large language models at the MRCS Part A: a tool for medical education? Ann R Coll Surg Engl. 2023. doi:10.1308/rcsann.2023.0085.
  6. Yitzhaki S, Peled N, Kaplan E, et al. Comparing ChatGPT-4 and a Paediatric Intensive Care Specialist in Responding to Medical Education Questions: A Multicenter Evaluation. J Paediatr Child Health. 2025. doi:10.1111/jpc.70080.
  7. Zheng K, Shen Z, Chen Z, et al. AI-empowered scenario-based simulation teaching method for medical interns in cardiovascular medicine: a randomized controlled study. BMC Med Educ. 2024;24:1003. doi:10.1186/s12909-024-05977-z.
  8. George LC, O’Neill R, Merchant AM. Residency Training in Robotic General Surgery: A Survey of Program Directors. Minim Invasive Surg. 2018;2018:8464298. doi:10.1155/2018/8464298.
  9. Dulan G, Rege RV, Hogg DC, Gilberg-fisher KM. Developing a comprehensive, proficiency-based training program for robotic surgery. Surgery. 2012;152(3):477–488. doi:10.1016/j.surg.2012.07.028.
  10. Dulan G, Rege RV, Hogg DC, Gilberg-fisher KM. Developing a comprehensive, proficiency-based training program for robotic surgery. Surgery. 2012;152(3):477–488. doi:10.1016/j.surg.2012.07.028.
  11. Zidoun A, Mardi A. The effect of artificial intelligence simulators compared to simulated patients on medical students’ history-taking skills: a randomized controlled trial protocol. BMC Med Educ. 2024;24:1260. doi:10.1186/s12909-024-06547-z.
  12. Zhou J, Zhang J, Wan R, et al. Integrating AI into clinical education: evaluating general practice trainees’ proficiency in distinguishing AI-generated hallucinations and impacting factors. BMC Med Educ. 2025;25:406. doi:10.1186/s12909-025-06916-2.
  13. Ziapour A, Motevaseli S, Fathi M, et al. Factors influencing medical students’ readiness for artificial intelligence: a cross-sectional study. BMC Med Educ. 2025;25:264. doi:10.1186/s12909-025-06852-1.
  14. Clusmann J, Kolbinger FR, Muti HS, Schoepf UJ. The future landscape of large Language models in medical practice: systematic review on the promising perspectives and valid concerns. Healthc (Basel). 2023;11(6):887.
  15. Masters K. Artificial intelligence in medical education. Med Teach. 2019;41(9):976–80. doi:10.1080/0142159X.2019.1595557.
  16. Gkegkes I, Mamais I, Iavazzo C. Robotics in general surgery: A systematic cost assessment. J Minim Access Surg. 2017;13(4):243–255. doi:10.4103/0972-9941.195565.
  17. Zhao C, Xiao M, Liu H, et al. Reducing the number of unnecessary biopsies of US-BI RADS 4a lesions through a deep learning method for residents-in-training: a cross-sectional study. BMJ Open. 2020;10(6):e035757. doi:10.1136/bmjopen-2019-035757.
  18. Alowais SA, Alghamdi SS, Alsuhebany N, et al. Revolutionizing healthcare: the role of artificial intelligence in clinical practice. BMC Med Educ. 2023;23:689. doi:10.1186/s12909-023-04698-z.
  19. Gulov MK, Shermatov DS, Sattorov DK, Kobilov KK. Appliance of modern informative technologies in medical postgraduate educational process. Avicenna Bulletin. 2017;19(1):42-5. doi:10.25005/2074-0581-2017-19-1-42-45.
  20. Стратегия развития искусственного интеллекта в Республике Таджикистан на период до 2040 года. постановлению Правительства Республики Таджикистан. 30 сентября 2022 года, №483. Доступно по: [http://www.portali-huquqi.tj/publicadliya/view_qonunhoview.php?showdetail=&asosi_id=26592](http://www.portali-huquqi.tj/publicadliya/view_qonunhoview.php?showdetail=&asosi_id=26592).
  21. Гулов МК, Шерматов ДС, Саторов ДК, Кобилов КК. Использование современных информационных технологий в медицинском последипломном образовательном процессе. Вестник Авиценны. 2017;19(1):42-45. doi: 10.25005/2074-0581-2017-19-1-42-45.
  22. Среднесрочная программа развития цифровой экономики в Республике Таджикистан на 2021–2025 годы от 26 октября 2021 года, №460. Доступно по: [https://www.adlia.tj/show_doc.fwx?rgn=140416](https://www.google.com/search?q=https://www.adlia.tj/show_doc.fwx%3Frgn%3D140416).
  23. Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, Elepaño C, et al. Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models. PLOS Digit Health. 2023;2(2):e0000198.
  24. Yitzhaki S, Peled N, Kaplan E, Kadmon G, Nahum E, Gendler Y, et al. Comparing ChatGPT-4 and a Paediatric Intensive Care Specialist in Responding to Medical Education Questions: A Multicenter Evaluation. J Paediatr Child Health. 2025.
  25. Zhao X, Xie P, Wang M, Li W, Pickhardt PJ, Xia W, et al. Deep learning–based fully automated detection and segmentation of lymph nodes on multiparametric-mri for rectal cancer: A multicentre study. EBioMedicine. 2020;56:102780.
  26. Zhou J, Zhang J, Wan R, Cui X, Liu X, Zhang Y, et al. Integrating AI into clinical education: evaluating general practice trainees’ proficiency in distinguishing AI-generated hallucinations and impacting factors. BMC Med Educ. 2025;25:406.
  27. Zheng K, Shen Z, Chen Z, Hu Z, Deng X, Liu X, et al. AI-empowered scenario-based simulation teaching method for medical interns in cardiovascular medicine: a randomized controlled study. BMC Med Educ. 2024;24:1003.
  28. Zheng T, Xiao H. The role of artificial intelligence curriculum in promoting medical students’ preparedness for future work with AI: a large-scale cross-sectional study. BMC Med Educ. 2025;25:1115.
  29. Zheleznyakov OV. Ethical challenges of applying artificial intelligence elements in medicine and medical research. Med Sci Educ. 2023;14(1):21–32.
  30. Стратегия развития искусственного интеллекта в Республике Таджикистан на период до 2040 года. Постановление Правительства Республики Таджикистан. 30 сентября 2022 года, №483.
  31. Zong H, Li J, Wu E, Wu R, Lu J, Shen B. Performance of ChatGPT on Chinese national medical licensing examinations: a five-year examination evaluation study for physicians, pharmacists and nurses. BMC Med Educ. 2024;24:143.