ARTIFICIAL INTELLIGENCE AND THE DIAGNOSTIC INVISIBILITY OF WOMEN’S METABOLIC DISORDERS: A NARRATIVE REVIEW OF SOCIO-TECHNICAL CHALLENGES IN PCOS/PMOS RECOGNITION

The narrative review discusses the proposal to rename Polycystic Ovary Syndrome (PCOS) to Polyendocrine Metabolic Ovarian Syndrome (PMOS), highlighting a shift towards a systemic understanding of the disorder. It identifies diagnostic delays as socio-technical issues influenced by fragmented care and gender biases in electronic health records (EHRs). The review suggests that AI-supported EHR analysis could improve early recognition of PCOS/PMOS by integrating data from various medical specialties. However, it warns that AI systems might perpetuate existing biases if trained on incomplete datasets. The study concludes that AI should be seen as a socio-technical tool, emphasizing the importance of transparency, explainability, and bias-aware validation in its application.