AI Integration of Pathology and Radiotherapy for Survival Prediction in Oesophageal Cancer (23041)

Four Year Funded Studentship. Application Deadline: 15 March 2025

Multimodal AI-Driven Integration of Digital Pathology and Radiotherapy for Precision Risk Stratification and Survival Prediction in Oesophageal Cancer.

Primary supervisor: Dr Charles-Antoine Collins-Fekete

Co-Supervisor: Prof Maria Hawkins

Introduction

A funded PhD studentship is available in the UCL Department of Medical Physics and Biomedical Engineering. Funding will be at least the UCL minimum stipend rate. Details can be found here.

The project is embedded within the UCL Cancer Collaboratorium, a highly multi-disciplinary team of renowned cancer radiotherapy and imaging experts. This project is unique as it ties together world-class expertise from national/international research centres: University College London, University College London Hospital, UCL Partners Network, and the CRUK City of London. The available infrastructure (GPU) and expertise within this project will provide a strong basis for scientific discoveries at the peak of the current state-of-the-art.

As a member of UCL, the successful candidate will have access to a wealth of advanced courses for facilitating research and soft skills, allowing them to advance as an independent researcher. The successful candidate will join our Medical Physics and Biomedical Engineering MPhil/PhD and benefit from the activities and events organised by the department, whilst also being part of the UCL Cancer Collaboratorium. 

Project Background

Oesophageal cancer is a highly aggressive malignancy with a poor prognosis, ranking as the sixth leading cause of cancer-related deaths worldwide. The 5-year survival rate remains below 20%, largely due to late-stage diagnosis and the lack of robust predictive tools for personalised treatment planning. Current approaches in oncology rely on single-modality data, such as histopathology or radiology, which fail to capture the full complexity of the disease. Integrating multimodal data—combining digital pathology, radiology and radiotherapy, genomics, and clinical outcomes—offers a transformative opportunity to improve risk stratification and survival prediction. This project leverages unique datasets, including in-house cohorts, the multimodal SCOPE 2 clinical trial data, and pathology data from partner organisations, totalling approximately 1000 patients at the outset, with ongoing accrual.

The research builds on the expertise of an interdisciplinary team of AI researchers and clinicians who have delivered proven results in colorectal cancer, where their multimodal AI frameworks achieved state-of-the-art performance in predicting treatment response and survival outcomes. By adapting and refining these methodologies for oesophageal cancer, the team aims to address critical gaps in this understudied malignancy. Their prior work in cancer, published in leading journals, demonstrates the expertise of the team in integrating heterogeneous data types into clinically actionable tools. This project will extend these successes, combining cutting-edge AI techniques with robust clinical validation to deliver scalable solutions for personalised oncology.

Research aims

This project aims to develop a multimodal AI framework, including foundational models, to independently and jointly evaluate risk across digital pathology, radiology/radiotherapy, genomics and clinical data in oesophageal cancer. By integrating these modalities, we seek to create a comprehensive predictive tool for survival outcomes and treatment response. The research will focus on building interpretable, robust models that can stratify patients into risk categories, enabling personalised therapeutic strategies.  

The project holds significant commercial potential, with applications in diagnostic tools, decision-support systems, and predictive analytics for healthcare providers and pharmaceutical companies.

Essential criteria

• Fee Status: This studentship is fully funded. 
• Academic background: The PhD position would suit applicants with a first or upper-second-class UK Bachelor’s degree (or international equivalent) in Computer Science, Biomedical Engineering, Medical Physics, Data Science, or a related quantitative discipline.
• Technical Skills: Proficiency in Python programming, with experience in deep learning frameworks (e.g., PyTorch, TensorFlow).
• Research Interest: Demonstrable interest in AI applications for healthcare, particularly in cancer prognosis or treatment optimisation.
• Data Handling: Ability to manage and curate large, multimodal datasets (e.g., medical imaging, genomics, clinical records).

Desirable criteria 

• Advanced Degrees: A Master’s degree (or equivalent research experience) in a relevant fieldDomain Knowledge: Familiarity with medical imaging analysis (radiology/pathology) or cancer biology.
• Translational Research: Experience working with clinical datasets or collaborating with healthcare professionals.
• Software/MLOps: Exposure to tools for reproducible AI (e.g., Docker, MLflow) or cloud computing (AWS, Google Cloud).
• Communication Skills: Ability to present complex technical concepts to interdisciplinary audiences.

Funding 

This is a full studentship available to Home fee-paying and Overseas applicants. 

The successful student will receive a stipend starting from at least the UCL minimum (£21,237 in 2024/25) as well as the cost of tuition fees for [e.g., “Home”] fee students (£6,035 in 2024/25). The stipends awarded to PhD students at UCL are tax free and incur no income tax or national insurance contributions. The amount received increases each year over the duration of the studentship.

Read more about UCL’s fee eligibility criteria 

How to apply

The deadline for this application is 15 March 2025, to start by 1 May 2025, but applications will be assessed on a rolling basis so please do apply as early as possible.

Please complete the following steps to apply:

• Send an expression of interest and current CV to: c.fekete@ucl.ac.uk and medphys.pgr@ucl.ac.uk
• Please quote Project Code: 23041 in the email subject line. 
• Make a formal application via the UCL application portal. Please select the programme code RRDMPHSING01 (Research Degree: Medical Physics) and enter Project Code 23041 under ‘Name of Award 1’  
• If shortlisted, candidates will be invited for an interview, with more details about this to be provided when being invited

Application timeline

After the deadline, all applicants who expressed their interests and specified Project 23041 in their Portico application will be considered for interview. Candidates will normally be invited for an interview within three weeks of the deadline. If you have not been contacted within this time period, you have unfortunately not been successful in being shortlisted.

The interview panel will normally consist of the supervision team for the project. Please note that applications without specifying the project they are applying for and/or making a formal Portico application will be automatically rejected.

If you are offered and accept a studentship position, a formal UCL Offer of Admission will be sent to you in addition to an offer of studentship funding.