The increasing availability of data from sources such as hospital information systems, electronic health records, and health-related registries necessitates a new approach for developing AI-based decision support to aid clinicians with diagnostic decision-making. This is particularly crucial in shortening the diagnostic journey for patients with rare diseases. However, mapping European rare disease databases to facilitate ML-based screening technologies presents key challenges in organizational, FAIR (Findable, Accessible, Interoperable, and Reusable), and legal principles.
A scoping review was conducted following the PRISMA-ScR checklist, involving a primary search in three electronic databases – MEDLINE/Pubmed, Scopus, and Web of Science – and a secondary search in Google Scholar and on organization websites. This rigorous review process was independently carried out by two researchers, using a specially developed charting form to analyze relevant studies and categorize data items into organizational, FAIR, and legal domains.
The screening process yielded 73 eligible studies based on the inclusion and exclusion criteria, with over 60% of the research published in the past five years originating exclusively from EU/EEA countries. Over the last decade (2013-2022), a cycling trend was observed in the publications, with a peak of challenges reported every four years. Notably, organizational challenges dominated articles published up to 2018, while legal issues became the most frequently discussed topic from 2018 to 2022.
The data items were categorized into three domains: organizational (36 items), FAIR (15 items), and legal (33 items). In the organizational domain, data accessibility, sharing, long-term sustainability, governance, planning, and design, lack of harmonization and standardization, data collection quality, and privacy risks were the main challenges. In the FAIR domain, issues related to findability, accessibility, interoperability, and reusability were identified. The legal domain comprised challenges related to data protection, data management, ownership, research under GDPR and member state law, trust, transparency, and digitalization of health.
Despite the opportunities offered by innovation – such as automation, electronic health records, hospital-based information systems, biobanks, rare disease registries, and European Reference Networks – a diversity of challenges remain. Immediate action is needed to improve the governance of rare disease registries, implement FAIR principles, and enhance the EU legal framework.
Original Article DOI: 10.3389/fpubh.2023.1214766
Original title: Challenges in mapping European rare disease databases, relevant for ML-based screening technologies in terms of organizational, FAIR and legal principles: scoping review
This article has been prepared with the assistance of AI and reviewed by an editor. For more details, please refer to our Terms and Conditions. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author.
