Towards equitable AI for women’s health:accessible data as a catalyst for innovation

Preprint

Artificial intelligence (AI) is rapidly advancing across health domains, yet its integration into women’s health remains challenged, limited by under-representationin clinical literature and datasets, inconsistent data standards, and a lack of coordinated access to multimodal research-quality data resources. This research mapsthe current horizon of accessible (i.e. open and accessible on request) data thatcan contribute to AI development for women’s health. Main resources includeclinical data repositories, cancer registries, biobanks and published researchstudies. We summarise data resources related to cancers (breast, cervical,endometrial, and ovarian), chronic and acute health conditions (cardiovascular),under-diagnosed conditions (endometriosis), wearable and vital sign data fromremote health monitoring, and discuss other potential resources, such as thebroader healthcare data in community care and pharmacy data. We provide aworking definition of ”women’s health”, a table centralising key accessible datasources under the level of resources (national registry/clinical study, single/multimodality), and discuss key challenges and opportunities to advance AI researchand innovations in the field. To support accessibility and reuse, we also provide an open-access online repository of curated datasets and offer the wider community the opportunity to add to it. This paper thus offers a cornerstone forbuilding an equitable AI for women’s health: it can support future assessments ofdata completeness, demographic diversity, clinically deployability, methodological benchmarks, licensing, pharmacovigilance, and contributes to highlightingthe global AI research in the women’s health ecosystem.

Authors: Schor, B; Caolo, KC; Doan, LMT; Delfino, M; Occhipinti, A

• Publication status: Published
• Peer review status: Not peer reviewed
• Preprint server: Research Square
• Publication date: 2026-03-10
• DOI: 10.21203/rs.3.rs-8001150/v1
• EISSN: 2693-5015
• Server owner: Research Square Company
• Language: English
• Keywords: clinical AI; AI; women’s health; accessible data; open data; FAIR data; digital health; gender bias; equity