Enhanced predictive performance of artificial intelligence in individualized ovarian stimulation of in vitro fertilization: a retrospective cohort study - PubMed
a day ago
- #In Vitro Fertilization
- #Personalized Medicine
- #Artificial Intelligence
- Study investigates AI models for improving IVF success through controlled ovarian stimulation (COS).
- Over 2.5 million IVF cycles are conducted annually, with numbers expected to rise.
- AI models outperformed traditional clinical practices in predictive accuracy for COS optimization.
- A four-submodel system was developed, showing superior discrimination compared to conventional methods.
- Key COS components impacting response include protocol, recombinant FSH use, and starting dose.
- New potential biomarkers were identified for low and hyper ovarian responses.
- AI system demonstrated high precision in identifying effective COS strategies (95.5% and 98.4%).
- The study offers new insights for personalized reproductive medicine and IVF optimization.