Predicting the outcomes of in vitro fertilization programs using neural networks

DOI

https://dx.doi.org/10.18565/aig.2026.198

Portnov I.G., Lisovskaya T.V., Rezaikin A.V., Mayasina E.N., Salimov D.F., Ivanov A.V., Gzgzyan A.M., Levin V.A.
1) Clinical Institute of Reproductive Medicine, Yekaterinburg, Russia; 2) Ural State Medical University, Ministry of Health of Russia, Yekaterinburg, Russia; 3) Mariinsky City Hospital, St. Petersburg, Russia; 4) St. Petersburg State University, St. Petersburg, Russia; 5) Ornament Health AG, Lucerne, Switzerland

Background. The aim of any cycle of assisted reproductive technology (ART) treatment is the birth a healthy child. In this context, the development and use of mathematical models based on artificial neural networks in reproductive medicine are of particular importance, as they enable the prediction of program outcomes.
Objective. To improve the effectiveness of in vitro fertilization (IVF) programs by using predictive data based on neural networks.
Materials and methods. There were three interrelated predictive mathematical models describing the main stages of the IVF protocol: controlled ovarian stimulation and follicular aspiration, fertilization and embryo culture, and embryo transfer into the uterine cavity. Each of them was implemented as an artificial neural network with a specified architecture and was pre-trained on a training dataset comprising data from 5,200 IVF protocols. The training of neural networks was carried out using the Python programming language and standard libraries, in the form of a program for a personal computer.
Results. The analysis of the effectiveness of controlled ovarian stimulation revealed a statistically significant increase in the number of cycles with 6 to 19 oocytes retrieved when using the predictive model (114/152 (75.0%) cycles) compared with the control group (93/152 (61.19%) cycles, p=0.033). The analysis of the criterion variable for the second stage revealed differences between the groups when comparing the number of cycles in the first category (zero embryos obtained), the second category (one to two good-quality embryos) and the third category (three or more good-quality embryos). The differences between the groups in the number of embryos suitable for transfer into the uterine cavity were statistically significant, p=0.032. The analysis of the criterion variable for stage 3 revealed that pregnancy occurred in 70 out of 152 (46.05%) cases, compared with 53 out of 152 (34.76%) in the control group, p=0.047. The clinical case of infertility treatment using assisted reproductive technology (ART) is presented as an example of the application of the developed mathematical model based on neural networks; this demonstrated rather high accuracy in predicting the outcomes of each stage of the IVF protocol: 85.6%, 82.9% and 71.2% respectively.
Conclusion. Choosing the protocol for each stage of the IVF program based on the predicted outcomes, using artificial neural networks, makes it possible to adjust the treatment during the course of the program or to discontinue it in proper time.

Authors’ contributions. Portnov I.G., Rezaikin A.V. – developing the general concept; Lisovskaya T.V., Rezaikin A.V. – developing the study design; Ivanov A.V., Mayasina E.N., Gzgzyan A.M. – collecting and processing the material; Rezaikin A.V. – developing a neural network model; Salimov D.F., Rezaikin A.V. – statistical data analysis; Lisovskaya T.V. – writing the text; Levin V.A. – editing the article.
Conflicts of interest. The authors declare that there are no conflicts of interest.
Funding. No funding from external organizations or public funds was used in preparing this article.
Ethical Approval. The clinical study protocol was approved by the local ethics committee of Clinical Institute of Reproductive Medicine (Protocol No. 09 dated 15 December 2024).
Generative Artificial Intelligence. Artificial intelligence was not used in preparing and submitting the manuscript.
Patient Consent for Publication. The patients signed informed consent for the publication of their data.
Authors' Data Sharing Statement. The data supporting the findings of this study are available on request from the corresponding author after approval from the principal investigator.
For citation: Portnov I.G., Lisovskaya T.V., Rezaikin A.V., Mayasina E.N., Salimov D.F., Ivanov A.V., Gzgzyan A.M., Levin V.A. Predicting the outcomes of in vitro fertilization programs using neural networks.
Akusherstvo i Ginekologiya/Obstetrics and Gynecology. 2026; (7): 160-169 (in Russian)
https://dx.doi.org/10.18565/aig.2026.198

artificial intelligence in IVF

neural networks

the effectiveness of IVF programs

predicting the outcome of IVF programs

Full text (in Russian)

Abdullah K.A.L., Atazhanova T., Chavez-Badiola A., Shivhare S.B. Automation in ART: paving the way for the future of infertility treatment. Reprod. Sci. 2023; 30(4): 1006-16. https://dx.doi.org/10.1007/s43032-022-00941-y
Корсак В.С., Смирнова А.А., Шурыгина О.В. Регистр ВРТ Общероссийской общественной организации «Российская Aссоциация Репродукции Человека». Отчет за 2020 год. Проблемы репродукции. 2022; 28(6): 12-27. https://dx.doi.org/10.17116/repro20222806112 [Korsak V.S., Smirnova A.A., Shurygina O.V. ART Registry of the All-Russian Public Organization «Russian Association of Human Reproduction». Report for 2020. Russian Journal of Human Reproduction. 2022; 28(6): 12-27 (in Russian). https://dx.doi.org/10.17116/repro20222806112].
ESHRE Special Interest Group of Embryology; Alpha Scientists in Reproductive Medicine. The Vienna consensus: Report of an expert meeting on the development of ART laboratory performance indicators. Reprod. BioMed. Online. 2017; 35(5): 494-510. https://dx.doi.org/10.1016/j.rbmo.2017.06.015
ESHRE Clinic PI Working Group; Vlaisavljevic V., Apter S., Capalbo A., D'Angelo A., Gianaroli L., Griesinger G. et al. The Maribor consensus: report of an expert meeting on the development of performance indicators for clinical practice in ART. Hum. Reprod. Open. 2021; 2021(3): hoab022. https://dx.doi.org/10.1093/hropen/hoab022
Yuan G., Lv B., Hao C. Application of artificial neural networks in reproductive medicine. Hum. Fertil. (Camb). 2023; 26(5): 1195-201. https://dx.doi.org/10.1080/14647273.2022.2156301
Jiang V.S., Bormann C.L. Artificial intelligence in the in vitro fertilization laboratory: a review of advancements over the last decade. Fertil. Steril. 2023; 120(1): 17-23. https://dx.doi.org/10.1016/j.fertnstert.2023.05.149
Salih M., Austin C., Warty R.R., Tiktin C., Rolnik D.L., Momeni M. et al. Embryo selection through artificial intelligence versus embryologists: a systematic review. Hum. Reprod. Open. 2023; 2023(3): hoad031. https://dx.doi.org/10.1093/hropen/hoad031
Riegler M.A., Stensen M.H., Witczak O., Andersen J.M., Hicks S.A., Hammer H.L. et al. Artificial intelligence in the fertility clinic: status, pitfalls and possibilities. Hum. Reprod. 2021; 36(9): 2429-42. https://dx.doi.org/10.1093/humrep/deab168
Curchoe C.L., Flores-Saiffe Farias A., Mendizabal-Ruiz G., Chavez-Badiola A. Evaluating predictive models in reproductive medicine. Fertil. Steril. 2020; 114(5): 921-6. https://dx.doi.org/10.1016/j.fertnstert.2020.09.159
Патент на изобретение № 2809429 Российская Федерация, МПК A61B 8/00 (2006.01),G01N 33/68 (2006.01), G01N 33/48 (2006.01). Способ прогнозирования эффективности программ вспомогательных репродуктивных технологий на основе нейронных сетей/ Портнов И.Г., Лисовская Т.В., Резайкин А.В., Маясина Е.Н., Салимов Д.Ф., Коган И.Ю., Иванов А.В., Гзгзян А.М., Лисовский С.М., Аскеров Р.А., заявитель Общество с ограниченной ответственностью «ИЛАРМ-ЕК» (ООО «ИЛАРМ-ЕК») (RU).- № 2022129790; заявл. 17 ноября 2022 г.; опубл. 11 декабря 2023 г. [Patent for invention No. 2809429 Russian Federation, IPC A61B 8/00 (2006.01), G01N 33/68 (2006.01), G01N 33/48 (2006.01). Method for predicting the effectiveness of assisted reproductive technology programs based on neural networks / Portnov I.G., Lisovskaya T.V., Rezaikin A.V., Mayasina E.N., Salimov D.F., Kogan I.Yu., Ivanov A.V., Gzgzyan A.M., Lisovsky S.M., Askerov R.A., applicant Limited Liability Company "ILARM-EK" (OOO "ILARM-EK") (RU). - No. 2022129790; declared November 17, 2022; published December 11, 2023 (in Russian)].
Женское бесплодие. Клинические рекомендации. 2024. Акушерство и гинекология. 2025; 10(Приложение). https://dx.doi.org/10.18565/aig.2025.297 [Female infertility. Clinical guidelines. 2024. Obstetrics and Gynecology. 2025; 10(Suppl.) (in Russian). https://dx.doi.org/10.18565/aig.2025.297].
Bourdon M., Garnier A., Maignien Ch., Marcellin L., Dulioust E., Sogni P. et al. Assisted reproductive technology outcomes in women with a chronic viral disease. AIDS. 2021; 35(7): 1073-81. https://dx.doi.org/10.1097/QAD.0000000000002859
Zhang L., Zhang F., Ma Z., Jin J. Hepatitis B virus infection, infertility, and assisted reproduction. J. Zhejiang Univ. Sci. B. 2024; 25(8): 672-85. https://dx.doi.org/10.1631/jzus.B2300261
Shaw-Jackson C., Capraro M., Ameye L., Vandromme J., Manigart Y., Rozenberg S. et al. In vitro fertilization for women infected by hepatitis C virus: a matched case-control study and a systematic literature review. J. Assist. Reprod. Genet. 2017; 34(5): 587-97. https://dx.doi.org/10.1007/s10815-017-0892-8
Wei N., Wang K., Cheng F., Chang Z., Ren X., Liu Z. et al. Prevalence of human papillomavirus infection in the female partner of infertile couples undergoing IVF/ICSI-ET and subsequent reproductive outcomes. J. Clin. Med. 2022; 11(23): 7185. https://dx.doi.org/10.3390/jcm11237185
George J.S., Srouji S.S., Little S.E., Ginsburg E.S., Lanes A. The impact of increasing body mass index on in vitro fertilization treatment, obstetrical, and neonatal outcomes. J. Obstet. Gynecol. 2024; 230(2): 239.e1-e14. https://dx.doi.org/10.1016/j.Azg.2023.10.018
Qin X., Du J., He R., Li Y. , Zhu Q., Li Y. et al. Adverse effects of type 2 diabetes mellitus on ovarian reserve and pregnancy outcomes during the assisted reproductive technology process. Front. Endocrinol. (Lausanne). 2023: 14: 1274327. https://dx.doi.org/10.3389/fendo.2023.1274327
Yuan R.Y., Li S., Feng X., Li X.L., Lin X., Gao F. et al. Comparison of embryo quality and pregnancy outcomes for patients with low ovarian reserve in natural cycles and mildly stimulated cycles: a cohort study. J. Obstet. Gynaecol. 2024; 44(1): 2303693. https://dx.doi.org/10.1080/01443615.2024.2303693
Harvey A.J., Willson B.E., Surrey E.S., Gardner D.K. Ovarian stimulation protocols: impact on oocyte and endometrial quality and function. Fertil. Steril. 2025; 123(Suppl.1): 10-21. https://dx.doi.org/10.1016/j.fertnstert.2024.08.340
Braam S.C., de Bruin J.P., Mol B.W.J., van Wely M. The perspective of women with an increased risk of OHSS regarding the safety and burden of IVF: a discrete choice experiment. Hum. Reprod. Open. 2020; 2020(2): hoz034. https://dx.doi.org/10.1093/hropen/hoz034
Sun B., Ma Y., Li L., Hu L., Wang F., Zhang Y. et al. Factors associated with ovarian hyperstimulation syndrome (OHSS) severity in women with polycystic ovary syndrome undergoing IVF/ICSI. Front. Endocrinol. (Lausanne). 2021; 11: 615957. https://dx.doi.org/10.3389/fendo.2020.615957
Tran D., Cooke S., Illingworth P.J., Gardner D.K. Deep learning as a predictive tool for fetal heart pregnancy following time-lapse incubation and blastocyst transfer. Hum. Reprod. 2019; 34(6): 1011-8. https://dx.doi.org/10.1093/humrep/dez064
Mathew A., Amudha P., Sivakumari S. Deep learning techniques: an overview. In: Hassanien A., Bhatnagar R., Darwish A., eds. Advanced machine learning technologies and applications. AMLTA 2020. Singapore: Springer; 2020: 599-608. https://dx.doi.org/10.1007/978-981-15-3383-9_54
Matt T. "Stochastic Gradient Descent". Business Data Science: Combining Machine Learning and Economics to Optimize, Automate, and Accelerate Business Decisions. New York: McGraw-Hill; 2019: 303-7.

Received 23.06.2026

Accepted 10.07.2026

Igor G. Portnov, PhD, General Director, Clinical Institute of Reproductive Medicine, Yekaterinburg, Russia, info@Kirm.clinic
Tatiana V. Lisovskaya, Dr. Med. Sci., Associate Professor, Deputy Director General for Scientific, Organizational, and Methodological Work, Clinical Institute of Reproductive Medicine, Yekaterinburg, Russia, tv.lis@mail.ru, https://orcid.org/0000-0002-9747-9323
Alexey V. Rezaikin, PhD, Associate Professor, Department of Medical Physics and Digital Technologies, Ural State Medical University, Ministry of Health of Russia, Yekaterinburg, Russia, alexrez@yandex.ru, https://orcid.org/0000-0002-8665-5299
Andrey V. Ivanov, obstetrician-gynecologist, reproductologist, Head of the Assisted Reproductive Technologies Department, City Mariinsky Hospital, St. Petersburg, Russia, dr.ivanovav@gmail.com
Elena N. Mayasina, PhD, Deputy General Director for Medical Unit, Clinical Institute of Reproductive Medicine, Yekaterinburg, Russia, elena.mayasina@gmail.com,
https://orcid.org/0000-0002-3387-819X
Daniil F. Salimov, PhD, embryologist, Deputy General Director for Laboratory Work, Clinical Institute of Reproductive Medicine, Yekaterinburg, Russia, fsalimov@mail.ru, https://orcid.org/0009-0004-2972-494Х
Alexander M. Gzgzyan, Dr. Med. Sci., Professor, Professor at the Department of Obstetrics, Gynecology and Reproductive Medicine, St. Petersburg State University,
St. Petersburg, Russia; Medical Director, SKYFERT LLC, St. Petersburg, Russia, agzgzyan@mail.ru, https://orcid.org/0000-0003-3917-9493
Vitaly A. Levin, endocrinologist, specialist in health management and public health, Ornament Health AG, Lucerne, Switzerland, vitalylevin0205@gmail.com,
https://orcid.org/0009-0000-8758-0842

Predicting the outcomes of in vitro fertilization programs using neural networks

Portnov I.G., Lisovskaya T.V., Rezaikin A.V., Mayasina E.N., Salimov D.F., Ivanov A.V., Gzgzyan A.M., Levin V.A.

Keywords

References

About the Authors

Similar Articles