AUTOMATIC CLASSIFICATION OF STAGES OF EMBRYO DEVELOPMENT BASED ON MACHINE LEARNING METHODS

This article proposes a structural scheme of information flow management in the information and analytical decision support system (IADSS) for embryologists and reproductive medicine physicians based on artificial intelligence (AI), which was developed during research for a doctoral dissertation entitled "Information System for Diagnosing Pathologies in Reproductive Medicine". This system will optimize the work of physicians in reproductive centers and improve the efficiency of programs using assisted reproductive technologies (ART) by providing recommendations for implementing/adjusting protocols, automatic assessment of embryos according to the Gardner scale, and recommendations for selecting embryos taking into account genetic abnormalities. A literature review of works by authors who previously developed systems using AI in reproductive medicine was conducted. The review considered the methods and models used for various tasks, such as assessing embryo ploidy, determining the most viable embryo, assessing the implantation potential of the embryo, predicting the onset of pregnancy and live birth. A prototype of one of the modules of the IADSS for the analysis of embryonic images for the automated classification of the development stage of embryos based on their visual characteristics has been developed. In the future, it is planned to supplement other modules of the system, process large volumes of data and conduct testing in real clinical practice.

1. Lokshin V. Assisted Reproductive Technologies in the Republic of Kazakhstan: A 6-Year Trend Analysis from Efficacy to Availability / V. Lokshin, M. Omar, S. Karibaeva // J.Reproduction & Infertility. – 2022. – № 23. – R. 61-66. https://www.jri.ir/article/120131. (In English).

2. Petrova E.V. Osnovnye ehtapy preodoleniya besplodiya s ispol'zovaniem programmy vspomogatel'nykh reproduktivnykh tekhnologii i ee modifikatsii / E.V. Petrova, S.I. Mikhalevich // Meditsinskie novosti. – 2012. – № 2. – R. 45. (In Russian).

3. An artificial intelligence model (euploid prediction algorithm) can predict embryo ploidy status based on time-lapse data / V. Huang et al // Reproductive Biology and Endocrinology. – 2021. – № 19. – R. 185. https://doi.org/10.1186/s12958-021-00864-4. (In English).

4. Can the combination of time-lapse parameters and clinical features predict embryonic ploidy status or implantation? / Y. Zou et al // Reprod Biomed Online. – 2022. – № 45(4). – R. 643-651. https://doi.org/10.1016/j.rbmo.2022.06.007. (In English).

5. Artificial Intelligence-Based Detection of Human Embryo Components for Assisted Reproduction by In Vitro Fertilization / A. Mushtaq et al // Sensors. – 2022. – № 22. – R. 7418. https://doi.org/10.3390/s22197418. (In English).

6. Feasibility of artificial intelligence for predicting live birth without aneuploidy from a blastocyst image / M. Yasunari et al // Reproductive medicine and biology. – 2019. https://doi.org/10.1002/rmb2.12267. (In English).

7. Machine learning for prediction of euploidy in human embryos: in search of the best-performing model and predictive features / S. De Gheselle et al // Fertility and Sterility. – 2022. – № 117(4). – R. 0015-0282. https://doi.org/10.1016/j.fertnstert.2021.11.029. (In English).

8. Towards the automation of early-stage human embryo development detection / V. Raudonis et al // BioMedical Engineering OnLine. – 2019. – № 18. – R. 120. https://doi.org/10.1186/s12938-019-0738-y. (In English).

9. A non-invasive artificial intelligence approach for the prediction of human blastocyst ploidy: a retrospective model development and validation study / J. Barnes, et al // The Lancet Digital Health. – 2023. – № 5(1). – R. e28-e40. https://doi.org/10.1016/S2589-7500(22)00213-8. (In English).

10. Development of an artificial intelligence model for predicting the likelihood of human embryo euploidy based on blastocyst images from multiple imaging systems during IVF / S. Diakiw et al // Human Reproduction. – 2022. – Vol 30, № 37(8). – R. 1746-1759. https://doi.org/10.1093/humrep/deac131. (In English).

11. The use of voting ensembles to improve the accuracy of deep neural networks as a non-invasive method to predict embryo ploidy status / V. Jiang et al // Journal of Assisted Reproduction and Genetics. – 2023. – № 40. – R. 301-308. https://doi.org/10.1007/s10815-022-02707-6. (In English).

12. Development of an artificial intelligence based model for predicting the euploidy of blastocysts in PGT-A treatments / Z. Yuan et al // Scientific Reports. – 2023. – № 13. – R. 2322. https://doi.org/10.1038/s41598-023-29319-z. (In English).

13. Using Deep Learning with Large Dataset of Microscope Images to Develop an Automated Embryo Grading System / Chen et al // Fertility & Reproduction. – 2019. – № 01(01). – R. 1-6. https://doi.org/10.1142/S2661318219500051. (In English).

14. Embryo Ranking Intelligent Classification Algorithm (ERICA): artificial intelligence clinical assistant predicting embryo ploidy and implantation / A. Chavez-Badiola et al // Reprod Biomed Online. – 2020. – № 41(4). – R. 585-593. https://doi.org/10.1016/j.rbmo.2020.07.003. (In English).

15. Iskusstvennyi intellekt – klyuch k razvitiyu ehmbriologicheskoi laboratorii / A. Begimbaeva i dr. // Reproduktivnaya meditsina (Tsentral'naya Aziya). – 2024. – № 3. – R. 42-49. https://doi.org/10.37800/RM.3.2024.42-49. (In English).