Development of a deep learning-based model to predict the mutational landscape of colorectal polyps from haematoxylin and eosin slides
Supervisors
Prof Joanne Edwards, School of Cancer Sciences, University of Glasgow
Dr Stephen McSorley, School of Cancer Sciences, University of Glasgow
Dr Noori Maka
Industry Partner: TileBio
Summary
Colorectal cancer (CRC) is a leading cause of cancer death in the UK, often developing from benign polyps over many years. While polypectomy can prevent CRC, not all patients require ongoing surveillance, and current guidelines place a significant burden on NHS endoscopy services. The INCISE project (INtegrated TeChnologies for Improved Polyp SurveillancE) aims to transform how we predict which patients are at risk of developing further (metachronous) polyps.
This PhD studentship offers the opportunity to work with a unique cohort of 2,642 patients, including digitised H&E slides and matched transcriptomic and mutational data. The project will develop a deep learning model to predict molecular biomarkers directly from pathology images, enabling personalised risk stratification based on routine diagnostic slides.
The student will gain interdisciplinary training in colorectal cancer biology, histopathology, bioinformatics, and AI-based image analysis. This work will contribute to more targeted surveillance strategies, reducing unnecessary procedures while improving early detection of high-risk patients.
This is an exciting opportunity to contribute to cutting-edge translational research with real-world clinical impact
There will be a Q&A Zoom call with Joanne about this Project.
Date: 10 Dec 2025
Time: 2:30 pm (UK Time)
If you wish to receive the Zoom Call information please email precisionmedicine-dtp@glasgow.ac.uk
Remembering to put in the Subject line, the name of the Supervisor, this will allow us to send you the correct Zoom call information.