D.S. Ripka1, E.A. Semenova2
1,2 Saint-Petersburg Electrotechnical University «LETI» (St. Petersburg, Russia)
2 easemenova@etu.ru
Currently, assisted reproductive technologies (ART) are one of the key methods of infertility treatment. The success of ART procedures is determined by both the medical and individual characteristics of the patients, which makes the task of predicting outcomes extremely difficult. In this regard, an important step is the identification and ranking of diagnostically significant indicators (DP) that can be used to improve the accuracy of forecasts and personalize treatment.
Aim – formation of a ranked set of diagnostically significant indicators based on an expert assessment of the importance of various signs in planning procedures for assisted reproductive technologies and pregnancy, necessary to increase the accuracy of predictive models predicting ART outcomes.
As part of the study, a survey of reproductive doctors was conducted, who assessed the importance of various signs used in planning ART and pregnancy procedures. Each feature was evaluated on a scale from 1 to 100, after which the average value, the spread (standard deviation), the weight and the magnitude of the error were calculated for each of them. These data allowed us to form a ranked set of diagnostically significant indicators that will be used to improve data quality and improve the accuracy of predictive models.
The results of the study showed that the most significant factors influencing the success of assisted reproductive technologies are the level of ovulation, the presence of gynecological diseases and hormone levels. These indicators have received high ratings from experts (average values above 95 points) and are characterized by a low spread of opinions (standard deviation less than 10). This indicates that these factors play a key role in predicting ART outcomes and should be taken into account when developing personalized treatment approaches.
The results open up new opportunities for developing personalized approaches to infertility treatment and improving the quality of predictive models. Using a ranked set of diagnostically significant indicators will allow doctors to more accurately predict the outcomes of ART and make informed decisions when planning procedures.
Ripka D.S., Semenova E.A. Formation of a set of diagnostically significant indicators for predicting the outcomes of assisted reproductive technologies based on expert assessments. Biomedicine Radioengineering. 2025. V. 28. № 2. P. 63–69. DOI: https:// doi.org/10.18127/j15604136-202502-10 (In Russian)
- Ducreux B., Patrat C., Trasler J., Fauque P. Transcriptomic integrity of human oocytes used in ARTs: technical and intrinsic factor effects. Human reproduction update. 2023. DOI: 10.1093/humupd/dmad025.
- Jiang M. et al. Factors affecting success rate of atraumatic restorative treatment (ART) restorations in children: A systematic review and meta-analysis. Journal of dentistry. 103526. 2020. DOI: j.jdent.2020.103526.
- Ripka D.S., Semenova E.A. Sistema prognozirovaniya effektivnosti provedeniya procedury VRT: analiz sistemy faktorov. Biotekhnosfera. 2025. № 1. S. 34–37 (In Russian).
- Peng J. et al. Machine learning algorithms in constructing prediction models for assisted reproductive technology (ART) related live birth outcomes. Sci Rep 14. 32083. 2024. DOI: 10.1038/s41598-024-83781-x.