Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
REALM Infographic: Patient Group Representative Perspectives on the Oversight of AI in Healthcare
0
Zitationen
3
Autoren
2026
Jahr
Abstract
This infographic summarises key findings from the REALM Stakeholder Workshop: Patient Group and Medical Charity Perspectives on the Evaluation of Medical AI, held on 19 March 2026 in Brussels, Belgium. The workshop formed part of the Horizon Europe project REALM (Real-World Data and Evidence for Regulatory Assessment of Artificial Intelligence–Based Medical Devices), which brings together academic, clinical, regulatory, and industry partners across Europe to develop improved approaches for the evaluation, benchmarking, and governance of AI-based medical software. The event brought together 9 representatives of patient organisations and medical research charities from 5 European countries. Participants engaged with presentations on two REALM demonstrator cases — PGx2P (pharmacogenomics-informed prescribing support) and COPowereD (prediction of COPD exacerbations) — before taking part in structured group discussions on: trust in medical AI, access and regulatory oversight, and what should matter in evaluating medical AI systems. The workshop highlighted several key themes: trust in medical AI must be actively earned through transparency, explainability, and meaningful human oversight; regulation should be both robust and adaptive to technological innovation; evaluation frameworks should move beyond technical performance to include quality of life, usability, fairness, representativeness, and real-world patient experience; and patient perspectives should play a central role in shaping future AI governance and evaluation practices. The findings from the workshop will contribute to the development of REALM’s best practice guidelines and living-lab activities. For more information on the REALM project, see:https://realm-ai.eu/
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.740 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.649 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 8.202 Zit.
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
2019 · 6.886 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.781 Zit.