An Artificial Intelligence-Based Model to Predict Pregnancy After Intrauterine Insemination: A Retrospective Analysis...

Area of Science:

• Reproductive Medicine
• Artificial Intelligence in Healthcare
• Machine Learning in Clinical Applications

Background:

• Intrauterine insemination (IUI) is a primary treatment for infertility, particularly unexplained cases.
• Current IUI success rates can be improved with predictive tools for better patient management.
• Developing AI-based support is crucial for optimizing IUI cycle outcomes.

Purpose of the Study:

• To develop a precise machine learning model for predicting pregnancy outcomes in Intrauterine Insemination (IUI) cycles.
• To identify key clinical and laboratory parameters that influence IUI success rates.
• To enhance clinical decision-making for couples undergoing fertility treatments.

Main Methods:

• A retrospective study analyzed 9501 IUI cycles from 3535 couples (aged 18-43).
• Twenty-one clinical and laboratory parameters were used to train and evaluate various machine learning algorithms.
• Model performance was assessed using Area Under the Curve (AUC) analysis.

Main Results:

• A Linear Support Vector Machine (SVM) model demonstrated superior performance compared to other classifiers.
• Key predictors for positive pregnancy outcomes included pre-wash sperm concentration, ovarian stimulation protocol, cycle length, and maternal age (AUC = 0.78).
• Paternal age was identified as the least effective predictor.

Conclusions:

• The developed Linear SVM model effectively predicts positive pregnancy outcomes following IUI.
• This AI tool offers potential benefits for clinical management and patient decision-making in fertility treatment.
• Further validation with independent datasets is necessary before widespread clinical implementation.