Postdoctoral Research Assistant in Multimodal Analysis to understand medical procedures.

University of Oxford; Oxford; On-site; GBP 38,000 - 48,000; Full time

Job summary

The University of Oxford invites applications for a postdoctoral researcher to join the Turing AI World-Leading Fellowship led by Professor Alison Noble. The role focuses on bridging theory and application of medical video/multimodal analysis and medical procedure modelling at the Institute of Biomedical Engineering, Headington.

You will design and pilot ML-based automated analysis models, work with clinical collaborators from first principles, and evaluate models in simulated and real clinical

Qualifications

• PhD or near completion in medical image/video analysis, computer-assisted intervention, or related field.
• Publications in relevant areas.
• Interest in blue-sky research at the interface of ML and clinical needs.

Responsibilities

• Design and pilot test ML-based medical video/multimodal analysis models.
• Collaborate with clinicians to evaluate models in simulated and real clinical scenarios.
• Publish findings and contribute to the Noble group's research goals.

Skills

Medical video analysis

Multimodal analysis

Machine learning research

Collaborative research

Education

PhD / DPhil in related field

Job description

We are seeking a creative and highly motivated postdoctoral researcher to join the Turing AI World-Leading Fellowship research programme lead by Professor Alison Noble. Topics in the research programme include single- and multi-modal video-based human-AI collaboration, federated learning for healthcare collaboration and partnership, and applied research to understand human skill in healthcare settings and AI to support healthcare professional skills training. The appointee will be a technical researcher joining the AI to understand human skills strand of the research programme. This post is focused on bridging theory and application of medical video/multimodal analysis and medical procedure modelling. The research will involve building video/multimodal models that can analyse medical procedures and evaluate how usable they are. Evaluation may involve quantitative studies (model performance) and qualitative studies (human factors assessment). The appointee will be part of the Noble research group at the Institute of Biomedical Engineering, based at the Old Road Campus in Headington. The post is funded by UKRI and is fixed-term for 18 months.

You will be working in a small, highly motivated interdisciplinary team working towards a shared goal. You will be responsible for the design and pilot testing of machine learning-based automated medical video/multimodal analysis models. The research will include work that aims to understand medical procedures using video/multimodal analysis. This requires solid technical experience in medical video/multimodal analysis and building efficient learning approaches from sparse data, as well as open-mindedness to co-develop blue-sky research questions with clinical collaborators from first principles. The work will involve both computational methods research and collaboration with experts in clinical medicine and human factors to design and evaluate models in simulated and real clinical scenarios. Evaluation may involve quantitative studies (model performance) and qualitative studies (human factors assessment).

You should hold a relevant PhD/DPhil (or near completion) with publications in medical image/video analysis, computer-assisted intervention, or video analysis in computer vision. Knowledge of a specific medical procedure is not a requirement but a genuine interest in research at the interface of machine learning with clinical demand and the curiosity to develop blue-sky research questions is.

Institute of Biomedical Engineering, Old Road Campus, Headington, OX3 7DQ