Fellowship Bio

I am a specialty registrar in radiology in Cambridge, subspecialising in neuroradiology. I previously studied medicine, surgery, and neuroscience at Cambridge, and was awarded a PhD for research in molecular neurology.

Fellowship Project

Real-world clinical evaluation of commercially-available artificial intelligence tools in radiology

Cambridge University Hospitals NHS Foundation Trust

My project has focused on the real-world clinical evaluation of commercially-available artificial intelligence tools in radiology. I aimed to evaluate specific artificial intelligence tools that had recently been deployed in our department for clinical or research purposes, as well as develop a general framework for the clinical evaluation of any future tool within radiology. This work has been led by my supervisor, Dr Tilak Das, a consultant neuroradiologist and the clinical AI lead within our department. I have been responsible for planning and performing the clinical evaluation of an AI tool aiding the longitudinal analysis of demyelinating lesions on MRI brain studies in patients with multiple sclerosis, and AI tools aiding the detection of large vessel occlusion on CT head angiogram studies in patients with suspected ischaemic stroke. Some of this work has been performed and results presented at the NHS/RCR Global AI Conference 2025, and some of this work is ongoing. I have also prepared a detailed evaluation framework document and summary checklist of key requirements that need to be met when implementing AI tools, to ensure that they are safe and effective.

Fellowship Testimonial

It has been a privilege to be part of the NHS Fellowship in Clinical AI. The organised lectures and workshops, self-directed learning, and project work have provided exposure across all domains of clinical AI, including technical, clinical, and regulatory. AI is expected to play an important role in radiology in the future, and I hope that the Fellowship has equipped me with the necessary knowledge, skills, and experience to ensure its safe and effective implementation.