Sign in Subscribe
Machine learning

Artificial Intelligence in Predicting Preconception Outcomes

Strategic Insights

— 1 min read

Data from a recent study have shed new light on the capabilities of artificial intelligence (AI) in predicting preconception outcomes. Researchers from Army Medical University have used five machine-learning algorithms to analyze data from a cohort of 4,097 couples, recruited from the Chongqing Health Center for Women and Children, to identify factors associated with preconception outcomes. The study's findings suggest that AI can be a valuable tool in predicting fertility outcomes, with the gradient boosting machine performing best in predicting associations between pre-pregnancy socio-psycho-behavioral exposures and preconception outcomes.

Key Takeaways:

  • The study used five machine-learning algorithms (Logistic Regression, Naive Bayes, Random Forest, Gradient Boosting Machine, and Support Vector Machine) to predict preconception outcomes based on data from 4,097 couples.
  • The gradient boosting machine performed best, with a relatively high AUC value (0.651) and better F1 score (0.61) in feature Set 1.
  • The study identified 24 variables associated with preconception outcomes in both spouses, including female age, male age, no pregnancy within 1 year without contraception, female pregnancy history, total sperm vitality, and use of contraceptive measures before enrollment.
  • The results indicated that AI can be a useful tool in predicting fertility outcomes, with the simplified model excluding semen parameters showing comparable performance to the more complex model.
  • The study's findings have implications for the use of AI in predicting preconception outcomes, particularly in the context of emerging technologies and psychological factors.

Statistics:

  • 4,097 couples were recruited from the Chongqing Health Center for Women and Children.
  • 5 machine-learning algorithms were used to predict preconception outcomes.
  • The gradient boosting machine performed best, with an AUC value of 0.651 and F1 score of 0.61.
  • 24 variables were identified as associated with preconception outcomes in both spouses.
  • The study's findings suggest that AI can be a useful tool in predicting fertility outcomes.

Sources:

  • Pregnancy probability prediction models based on 5 machine learning algorithms and comparison of their performance (lujunjunyidaxuexuebao, 2025, 47(12):1376-1387).
  • Editorial Office of Journal of Army Medical University.
  • Ren, C., Yang, H., Zhou, N. (2025). Research from Army Medical University Has Provided New Data on Machine Learning (Pregnancy probability prediction models based on 5 machine learning algorithms and comparison of their performance). Health & Medicine Week. July 18, 2025; p 4036.

Read more