D.S. Ripka1, E.A. Semenova2

1,2 St. Petersburg Electrotechnical University “LETI” (St. Petersburg, Russia)

1 dar.stkr@gmail.com, 2 easemenova@etu.ru

The development of predictive models for the effectiveness of assisted reproductive technologies (ART) faces limitations in clinical samples, the high dimensionality of diagnostic features, and the poor interpretability of machine learning models, which hinders their clinical application. Existing approaches fail to fully formalize and utilize the expertise accumulated by reproductive physicians over decades of practical experience, especially in the context of incomplete and dynamically changing diagnostic information.

The goal of the work. To develop a decision support system for doctors in emergency situations using modern information technologies to ensure the effectiveness of solving the task of saving a person’s life.

Results. A scientific substantiation has been carried out of the ways to improve the efficiency of solving the problem of preserving human life using modern information technologies; the tasks for achieving the goal of the work, the stages and principles of building the system, the structure of the doctor’s decision-support system, the algorithm for the operation of such a system, and the architecture of its software are formulated.

Practical significance. The principles for building a decision support system for doctors in emergency situations, proposed and described in the paper, along with the stages of their implementation, the system’s structure, and the operating algorithm, enable specialists in the field of developing such systems to clearly understand the interrelationship between the components — “models — research methods — system structure — operating algorithms — software and metrological support” — and to successfully solve research tasks and achieve the goal of developing such systems.

Ripka D.S., Semenova E.A. The concept of an open system for supporting medical decision-making for predicting the effectiveness of assisted reproductive technologies in conditions of incomplete diagnostic information // Biomedical Radioelectronics. 2026. V. 29. № 6. P. 147−152. DOI: https://doi.org/10.18127/j15604136-202606-25

1. Kritsotaki N. et al. Explainable Artificial Intelligence in Assisted Reproductive Technology: Bridging Prediction and Clinical Judgment. Biomedicines. 2026. V. 14(5). P. 1024. https://doi.org/10.3390/biomedicines14051024.
2. Nawaz A. et al. The role of multimodality in clinical disease diagnosis: advances, challenges, and opportunities. Front. Public Health. 2026. V. 14. P. 1788454. DOI: 10.3389/fpubh.2026.1788454.
3. Zhou Q., Chen Zh., Peng, S. Evaluating the application of dynamic prediction models in oncological prognostic studies with repeated measurement predictors. npj Precis. Onc. 2025. V. 9. P. 376. https://doi.org/10.1038/s41698-025-01162-7.
4. Semenova E., Ripka D. Formation of a Complex of Diagnostically Significant Indicators for Predicting the Outcomes of Assisted Reproductive Technologies and Pregnancy Pathologies. STDH. 2025. 294–296. 10.1109/STDH66836.2025.11227529.
5. Ripka D.S., Semenova E.A. Gibridnaya ekspertnaya model' prognozirovaniya iskhodov VRT na osnove vzveshennogo dereva reshenij i logisticheskoj kalibrovki. Biomedicinskaya radioelektronika. 2026. T. 29. № 3. S. 119−124. DOI: https:// doi.org/10.18127/j156 04136-202603-21 (In Russian).