Explainability, Uncertainty, and Clinical Trust in Artificial Intelligence for Endometriosis: A Scoping Review

This scoping review will map and synthesise recent literature on artificial intelligence and machine learning models developed for the diagnosis and clinical decision support of endometriosis and adenomyosis. The project focuses particularly on whether these systems are explainable, uncertainty-aware, and clinically trustworthy, rather than on performance alone. In addition to identifying the clinical tasks, data modalities, and model families used, the review will examine how studies report data and model design choices, validation strategies, explainability methods, uncertainty handling, ethics, bias, data governance, and regulatory or oversight considerations. It will also explore how issues of clinician and patient trust, translational readiness, and safe clinical adoption are described in the literature. The expected outcome is a structured evidence map of the current field, highlighting methodological strengths, reporting gaps, and key barriers to the development of trustworthy and clinically deployable AI for endometriosis and adenomyosis.