Brendan Delaney

Brendan Delaney is a Professor and Chair in Medical Informatics and Decision Making at Imperial College London's Department of Surgery & Cancer, within the Faculty of Medicine. He previously held the title of Wolfson Professor of General Practice at King's College London. His roles include chairing the Faculty of Medicine's AI Research Committee and contributing to national initiatives like the Medical Research Council Data Science Strategic Advisory Group and NIHR MRC Better Methods Better Research oversight group. He is also a GP Partner at Albion Street Group Practice.

Educations:

• MD, University of Birmingham (1993–1997)
• MB ChB, University of Oxford (1984–1987)
• MA, University of Cambridge (1981–1984)

Research Interests: Delaney's work focuses on AI-driven medical diagnosis, learning health systems, and cancer early detection in primary care. He emphasizes computable clinical guidelines, explainable AI, and data integration for equitable healthcare delivery. His research also addresses long COVID's clinical and societal impacts, including productivity loss, symptom fluctuations, and guideline development. Key areas include decision support systems, data semantics, and reducing bias in clinical documentation.

Scientific Awards & Fellowships:

• Founding Fellow of the UK Faculty of Clinical Informatics (2017–2024)
• Fellow of the British Computer Society (2024–present)

Advising & Grants: Delaney supervises three PhD students in AI for healthcare. He leads a £2.5m NIHR i4i grant with SOAP Health to improve cancer early detection via AI. His research is supported by the EPSRC (ROAD2H project) and Imperial College NHS Trust Biomedical Research Centre.

Labs & Teams: He collaborates with Imperial's Institute of Global Health Innovation, AI for Healthcare Network, and Cancer Technology Network. His LOCOMOTION consortium studies long COVID's impact on health services. He also contributes to the RECAP initiative for remote assessment tools and the TRANSFoRm project for primary care data interoperability.