I am Director of Research and Lecturer for the Centre of Excellence for Data Science, AI, and Modelling (DAIM), here at the University of Hull. I also hold an Honorary Contract with Hull University Teaching Hospitals (HUTH), NHS, as a Researcher in Medical Oncology. In both roles, I specialise in the development and evaluation of AI as a medical device (AIaMD) for disease diagnosis and prediction.
I am the AI lead for the ~£11 million NIHR Centre for Addiction and Mental Health Research at Hull and a co-investigator on a ~£600,000 EPSRC project applying Bayes’ Theorem to optimise trial designs for medical diagnostics. I also involved in many other AI and healthcare projects. Finally, I am the Digital, AI and Data Science lead for Humber and North Yorkshire Cancer Alliance Research Collaborative.
Before joining Hull in 2023, I spent approx. six years as a Senior NHS Healthcare Scientist at the Royal Victoria Infirmary, Newcastle, while also working as a Senior Methodologist and Manager at the NIHR Newcastle In Vitro Diagnostics Co-operative. There, I co-led a team of NHS and university scientists focused on evaluating and validating new medical diagnostics, including AI technologies. Over my career, I have contributed to over 20 medical test development and evaluation projects, securing around £20 million in research and commercial funding.
I have published on clinical studies, medical statistics, health economics, systematic reviews for medical tests, qualitative medical research, evidence generation methodologies for AI technologies, and AIaMDs generally. My work aims to bridge the gap between technical innovation and clinical application, ensuring that AIaMDs are optimally developed to meet real clinical needs.
Research Interests
I specialise in the technical development, evaluation and implementation of Artificial Intelligence as a Medical Device (AIaMD) for disease diagnosis and prediction, with a primary focus on oncology, but am also involved in broader applications across other disease areas. My technical focus is on Support Vector Machines (SVM) and Bayesian methods. SVMs identify the optimal hyperplane, which perceptron-based and several other methods do not, and can also provide measures of difficulty of classification, both of which are important for AIaMDs. I am also deeply interested in Bayes, as it emphasises the importance of priors (prior information) into analysis, which frequentist methods do not. I am currently working on integrating Bayesian and SVM models in various ways.
Teaching and Learning
MSc Artificial Intelligence for Healthcare
MSc Artificial Intelligence and Data Science
PhD Supervision Availability
Yes
PhD Topics
I can supervise Masters and PhDs on the topic of AI in healthcare, and philosophy and AI.
Current students I am leading on:
Taofiq Adeyemo, ‘Predicting bleeds in Glioma patients on anticoagulants using AI’.
Supervisors: Dr Will Jones, Eva Sousa, Prof. Antony Maravyas, Dr Farzana Hague, Prof. Anthony Beavis.
