EAU response to the AI in science call for evidence

The European Association of Urology (EAU) welcomes the opportunity to contribute to the European Strategy for AI in science. As a leading authority in urological research and clinical practice, we are committed to advancing healthcare through innovative data initiatives and evidence-based approaches.

One of our flagship initiatives is the UroEvidenceHub, a comprehensive platform that uses the power of big data, artificial intelligence (AI), and real-world evidence (RWE) to improve patient care across all urological conditions. The UroEvidenceHub aims to bridge the gap between clinical practice guidelines and individual patient characteristics, ensuring personalised and effective treatment options. The UroEvidenceHub will integrate data from various sources, including clinical trials, electronic health records, and patient-reported outcomes (PROMS).

By analysing this data, we can generate insights that inform clinical guidelines and decision support tools, ultimately improving the diagnosis, treatment, and management of urological diseases. For example, the Innovative Medicines Initiative (IMI) -funded PIONEER project, a key component of the UroEvidenceHub, focuses on prostate cancer and has successfully standardised and integrated data from over 3.5 million patients. Our commitment to data-driven healthcare is further shown by our collaboration with other IMI projects, such as OPTIMA. These projects have significantly contributed to the development of the UroEvidenceHub, enabling us to provide personalised healthcare recommendations and optimise treatment strategies for urological cancers (around one third of all cancers), bladder disorders including incontinence, and kidney diseases.

Recommendations to the European Strategy for AI in science:

• Health is a critical sector where there are many potential opportunities for scientific research and funding. Currently missing from the call for evidence, health must be a key priority. A central pillar of the future AI research council should be AI research in health.
• Research and funding for AI science in health should be maintained and increased during the next multi-annual financial framework. Dedicated testing and experimentation infrastructures in health, including sandboxes, to facilitate real-world AI validation require investment as well as federated data infrastructures, aligning with EHDS to support cross-border AI research in health.
• The AI Research Council should work closely with the AI Scientific Panel and Stakeholder Forum to ensure clinical and patient expertise is embedded in governance. Healthcare professionals and patients must be at the centre of the development and implementation of AI in health. The AI Research Council should also encourage their involvement, not only focusing on research institutions and universities, but also medical societies and patient coalitions as drivers of scientific change in medical fields.
• An AI research council should also assist with ensuring a clear understanding of the legal framework governing AI science in health. There are multiple regulations impacting the health sector, including horizontal legislation such as the AI Act and General Data Protection Regulation (GDPR) and sectoral legislation such as the Clinical Trials Regulation, the Health Technology Assessment Regulation, the Pharmaceutical Package, the European Health Data Space, the Medical Devices Regulation and IVD Regulation.

This layering of Regulatory approaches means that in practice there can be a much confusion amongst researchers and practitioners, and sometimes duplicative and contradictory messages from implementing authorities. We believe the AI research council could play a role in ensuring that implementing authorities work together, building collective expertise and issuing joint guidance for researchers and research institutions.