Senior Research Associate in Machine Learning for Medical Imaging and Molecular Prediction

The role

We are seeking a talented and enthusiastic postdoctoral (or working towards a PhD) scientist with experience and a track record in machine learning, particularly for applications of deep learning for medical imaging and/or molecular biomarker development, for a maternity cover role.

The successful applicant will join our world-leading and highly collaborative multi-disciplinary team of cancer population research scientists at the University of Bristol, based within our Cancer Research UK-funded Obesity-related Cancer Epidemiology Programme (OCEP). Our previously CRUK-funded programmes (2015 – 2025) substantially increased understanding of obesity’s importance in cancer aetiology, identifying complex links between the anatomical distribution of adipose tissue, metabolic dysfunction, and cancer risk. We demonstrated an urgent need to go ‘beyond BMI’ to investigate how unhealthy adipose distribution and its metabolic sequelae increase risk of obesity-related cancers and develop intervention strategies that target those mechanisms.

Hybrid working is available, ideally with at least one day per week on campus; however, this is negotiable.

What will you be doing?

This role will be based within Work Package 2, ���Risk stratification’, and will develop multi-omic models of obesity-related cancer risk from molecular models of cancer-related risk factors including imaging-derived adiposity traits. Risk models will be evaluated for their capacity to inform targeted interventions.

You should apply if

you have:

Detailed knowledge of population-based statistical methods to analyse large, multidimensional datasets 
Expertise in the use of machine learning methods for deriving and evaluating predictive models from large datasets, including using deep learning models for extracting features from medical imaging
Experience accessing and analysing large datasets within high-performance computing and/or cloud compute environments
Strong track record of academic publications
Experience of collaborating and corresponding with multiple studies