AI holds great promise in personalizing and improving care for patients with ovarian disease, a new systematic review and meta-analysis finds.
However, challenges remain in translating research into routine clinical care.
Promising applications of AI in ovarian care
In the analysis, the AI model showed high diagnostic accuracy for ovarian cancer by combining things such as ultrasound scans and blood test results.
Across 81 studies, the AI model correctly identified ovarian cancer in approximately 9 out of 10 cases, with a pooled specificity of 89% to 94%.
It was also highly accurate in excluding ovarian cancer even in the absence of ovarian cancer, with specificity ranging from 85 to 91%.
Additionally, the explainable AI tool effectively predicted complete surgical cytoreduction in advanced ovarian cancer, i.e., removal of all visible cancer in the operating room, with a pooled AUC of 0.87.
In the broader field of reproductive medicine, AI algorithms have helped physicians optimize ovarian stimulation protocols and predict follicular growth, reliably modeling ovarian responses in IVF (with a pooled AUC of 0.81).
Translating research into routine clinical practice
However, the researchers reported that the challenge ultimately lies in realizing the full potential of AI in ovarian care, as it is difficult to translate promising research findings into routine clinical practice.
The researchers noted substantial heterogeneity among studies caused by retrospective study designs, analysis of variable AI systems, and lack of standardized validation of AI models.
Only 22% of the studies analyzed reported prospective multicenter external validation.
Researchers undertook this rigorous validation to bridge the gap between research and daily clinical practice, and called for standardized methodologies and reporting frameworks, smooth integration with clinical workflows, and robust governance to ensure responsible and ethical use of AI.
Nevertheless, the authors conclude: “Artificial intelligence is a transformative force in the management of ovarian conditions.
“In gynecological oncology, AI will enhance every step of treatment, from early detection and accurate diagnosis to prognostic stratification and surgical planning.
“In reproductive medicine, AI will personalize ovarian stimulation and refine diagnosis of disparate endocrine diseases such as PCOS.”
reference
Yu H, Peng L. Artificial intelligence in ovarian pathophysiology and management: a systematic review and meta-analysis. J Ovary Research Institute. 2026;DOI:10.1186/s13048-026-02083-0.
Featured image: nerthuz by Adobe Stock
