DrStephen Auger

Dr Stephen Auger is a Post-CCT NIHR Clinical Lecturer in Neurology in the Department of Brain Sciences at Imperial College London and Imperial College Healthcare NHS Trust.

His work focuses on clinical AI safety, evaluation methodology, and synthetic simulation. Current medical AI benchmarks frequently fail to test algorithms against atypical presentations and rare failure modes. Dr Auger builds simulation environments that generate detailed, logic-grounded synthetic patient-doctor interactions filtered through realistically "messy" natural language interaction. These environments enable automated, large-scale counterfactual stress-testing of Large Language Models (LLMs) and clinical decision algorithms before deployment in live healthcare environments. [Paper in the context of Headache], [Paper in the context of Multiple Sclerosis]

He completed his MBPhD at University College London (UCL) at the Wellcome Centre for Human Neuroimaging under Prof Eleanor Maguire, using fMRI, virtual reality simulations and machine learning to investigate the neural mechanisms of human memory and navigation. He subsequently conducted postdoctoral research in Parkinson's disease risk prediction algorithms (PREDICT-PD) at Queen Mary University of London with Prof Alastair Noyce.

Other recent work:

• Automated NHS Clinical On-Call Rota Engine: Departmental on-call scheduling in acute hospital settings is a high-dimensional constraint optimisation problem (balancing job plans, leave, multi-site specialty coverage). It historically consumed days of senior clinician time each rota cycle. I developed a dedicated constraint-solving application built in Python/Streamlit that automates conflict-free rota generation, now used across multiple Hospitals, compressing scheduling work from days to minutes while enforcing a fair rota. 
• Quantifying Individual Treatment Effects: Developing a framework and companion web application to help clinicians move from intuitive guesswork to useful data for better clinical decisions. [Paper] | [Code]
• Machine Learning to Uncover Hidden Treatment Responses: Demonstrating how ML can detect complex patterns in clinical trial data that traditional analyses miss, and understanding the data requirements and collection practices necessary for this to be possible. [Paper] | [Code]