Machine Learning Models for Predicting Cumulative Live Birth Rates in Women with Diminished Ovarian Reserve at Different Stages

Overview

This study developed and validated machine learning models to predict cumulative live birth rates in women with diminished ovarian reserve (DOR).

Background

Diminished ovarian reserve (DOR) affects approximately 10% of infertile women and is associated with lower cumulative live birth rates following assisted reproductive technology (ART). Current prediction tools primarily rely on static parameters.

Data Highlights

Key Findings

• The CatBoost algorithm achieved the highest test-set discrimination across all stages.
• Baseline models included five features: female age, male age, AMH, BMI, and infertility duration.
• Post-stimulation markers provided negligible incremental value to cumulative live birth prediction.
• Embryological parameters significantly enhanced prediction accuracy at the pre-transfer stage.
• SHAP analysis indicated female age and embryo quality were the most influential factors in predictions.

Clinical Implications

The findings indicate that incorporating embryological parameters into predictive models can enhance the accuracy of cumulative live birth rate predictions for women with DOR.

Conclusion

The study highlights the use of machine learning models that integrate dynamic treatment data to enhance predictive accuracy in reproductive medicine.

Related Resources & Content

1. Frontiers in Endocrinology, 2026 -- Development and validation of a clinical prediction model for poor ovarian response in assisted reproductive technology
2. Frontiers in Endocrinology, 2026 -- Development and internal validation of a post-retrieval machine learning models for OHSS risk stratification in assisted reproductive technology: an exploratory study
3. Frontiers in Endocrinology, 2026 -- Non-linear saturation threshold of gonadotropin dose on cumulative live birth rates in advanced-age women with polycystic ovary syndrome: a retrospective cohort study
4. The International Glossary on Infertility and Fertility Care, 2025 | American Society for Reproductive Medicine | ASRM
5. Clinical outcomes in Patient-Oriented Strategies Encompassing Individualized Oocyte Number (POSEIDON) low-prognosis patients receiving in vitro fertilization/intracytoplasmic sperm injection treatment: a multicentered retrospective cohort study - ScienceDirect
6. Frontiers in Reproductive Health — Development of Machine Learning-Based Predictive Models for Fertility Intentions in Patients with Crohn's Disease
7. Artificial intelligence in the in vitro fertilization laboratory: a committee opinion | American Society for Reproductive Medicine | ASRM
8. The International Glossary on Infertility and Fertility Care, 2025 | American Society for Reproductive Medicine | ASRM
9. Clinical outcomes in Patient-Oriented Strategies Encompassing Individualized Oocyte Number (POSEIDON) low-prognosis patients receiving in vitro fertilization/intracytoplasmic sperm injection treatment: a multicentered retrospective cohort study - ScienceDirect