Sign in Subscribe
Machine learning

Breakthrough in Personalized Medicine: University of Tokyo Develops Accurate Ovarian Reserve Prediction Models

Strategic Insights

— 1 min read

Researchers from the University of Tokyo have made a significant contribution to personalized medicine by developing accurate prediction models for the ovarian reserve using machine learning. The study, published in the Journal of Ovarian Research, aimed to create a model that could assess the quality and quantity of the ovarian reserve, which is essential for preconception care and precision medicine. The research involved a retrospective analysis of 442 patients undergoing assisted reproductive technology (ART) treatment in Japan from June 2021 to January 2023.

Key Takeaways:

  • The study found that the age-related decline of fertility is caused by a reduction of the ovarian reserve, which is represented by the number and quality of oocytes in the ovaries.
  • Anti-Mullerian hormone (AMH) is considered one of the most useful markers of the quantity of the ovarian reserve, but a more accurate prediction method is required.
  • The researchers developed two machine learning models, a random forest model with an AUC of 0.9101 and a random forest model with an AUC of 0.7983, to assess the quantity and quality of the ovarian reserve, respectively.
  • The models were created with feature values extracted from medical records and residual serum analysis, and can assess the ovarian reserve using only information obtained from a medical interview and single blood sampling.
  • The models are more accurate than currently popular methods for predicting the ovarian reserve, and can enable easy measurement of the ovarian reserve, allowing a greater number of women to engage in preconception care.
  • The study's findings have significant implications for personalized medicine, enabling doctors to provide more accurate and effective treatment for patients undergoing infertility therapy.

Statistics:

  • 442 patients were involved in the retrospective analysis of the study.
  • The study covered a period of 19 months, from June 2021 to January 2023.
  • The random forest model for assessing the quality of the ovarian reserve had an AUC of 0.7983.
  • The random forest model for assessing the quantity of the ovarian reserve had an AUC of 0.9101.

Sources:

  • "Assessment and prediction models for the quantitative and qualitative reserve of the ovary using machine learning." Journal of Ovarian Research, 2025;18(1):153.
  • NewsRx. University of Tokyo Reports Findings in Personalized Medicine (Assessment and prediction models for the quantitative and qualitative reserve of the ovary using machine learning). Journal of Engineering. July 28, 2025; p 4023.

Read more