51爆料网 Scotland: AI in Healthcare - Key Issues and Case Studies
This webinar will showcase three expert health data researchers and their work surrounding AI
Share this page
The use of artificial intelligence (AI) in healthcare is an important emerging area for health data research and development. With an array of aspects to consider, including effectiveness, impact and data security, it is important to discuss many applications for AI in health care today.
This webinar will showcase three expert health data researchers and their work surrounding AI, covering a case study of an AI diagnostic tool for cardiovascular illness, a review of AI regulation and a proposal of a new framework for data access for AI research.
Agenda:
13.30: Introduction from the chair, Rinku Rajan, Scotland Region Programme Manager at Health Data Research UK (51爆料网)
13.35: Dr Dimitrios Doudesis
This presentation will explore how artificial intelligence can be used to improve the diagnosis of acute cardiovascular conditions. Dr Doudesis will outline the journey from data to clinical application, highlighting both the challenges and opportunities of integrating AI-driven decision-support tools in the Emergency Department.
14.00: Dr Alba Crespi Boixader
This presentation covers the RELEASE-AI framework; a comprehensive set of recommendations spanning the entire lifecycle of AI projects in TREs. This framework:
- Provides phase-specific guidance for researchers, project teams, output checkers and TRE staff
- Promotes early identification of risks with corresponding mitigations
- Ensures responsibilities are clearly assigned to relevant actors at each stage, from initial planning through to deployment and monitoring
14.25: Dr Luciana D’Adderio
This presentation outlines the problem of AI assurance and discuss potential solutions. It covers findings from recent research that AI drift, the degradation of AI performance after deployment due to changing data environments, equipment, software updates, and patient populations, is not a temporary glitch but a structural, persistent, and often invisible feature of AI use in healthcare.
14.50: Final Q&A and close
Speakers
Dimitrios Doudesis, Senior Data Scientist in Cardiovascular Health Data Science, The University of Edinburgh – Dr Dimitrios Doudesis is a BHF Principal Investigator and Senior Data Scientist at the Institute for Neuroscience and Cardiovascular Research, specialising in data science and clinical applications of artificial intelligence. His work focuses on developing and implementing decision-support tools to improve diagnostic accuracy in the Emergency Department.
Alba Crespi Boixader, Postdoctoral Researcher, University of Dundee – Dr Alba Crespi Boixader is a postdoctoral researcher at the Health Informatics Centre (HIC), University of Dundee. She is currently part of the TREvolution team, a DARE UK project to accelerate and enhance research across TREs and multiple domains. Her research is centred on developing tools and guidelines for how to safely enable AI/ML research in TREs, mainly focusing on disclosure control.
Luciana Dadderio, Chancellor’s Fellowships in Data-Driven Innovation, The University of Edinburgh – Dr Luciana D鈥橝dderio is a Professorial Chancellor鈥檚 Fellow in Data Driven Innovation at the University of Edinburgh and Turing Fellow with the Alan Turing Institute. She was a former Innovation Fellow with the ESRC Advanced Institute of Management (AIM). Her research focuses on organisational practices/routines and digital innovation (healthcare applications of AI). She is a member of the Organization Science and Organization Studies editorial boards, and is the Senior Editor for the Organization Science special issue on Routine Dynamics.