Virtual Multi-Omics: Decoding the Tumour Microenvironment from H&E Slides using Spatially-Aware AI
Supervisors
Prof John Le Quesne, School of Cancer Sciences, University of Glasgow
Dr Ke Yuan, School of Cancer Sciences, University of Glasgow
Prof David Chang, School of Cancer Sciences, University of Glasgow
Collaborator: TileBio
Summary
Deep learning AI for spatial biomarker discovery in cancer The cellular architecture of a tumour dictates disease progression, but much of the information available on a glass slide is wasted. This project will develop an AI model that infers detailed spatial-omic data directly from routine haematoxylin and eosin images. The core of the project is a transformer model that treats tissue architecture as a language. By learning the contextual relationships between cells, or the “grammar” of tissue organisation, the model moves beyond single cell analysis and understands the whole tumour ecosystem. Development will use very large collections of multimodal pathology data for self-supervised pre-training, followed by fine-tuning for multi-omic prediction from H&E alone. The primary outcome will be a validated virtual multi-omics platform for analysing large clinical cohorts. It will be applied to discover and validate spatial biomarkers, such as immunosuppressive niches and morphological signatures, that predict outcome and response to immunotherapy in lung cancer and mesothelioma. Because the model needs only a routine H&E image, it brings the benefits of spatial biology within reach to all patients in Scotland, where lung cancer incidence in the poorest is three times that of the richest. The work sits at the interface of machine learning, cancer biology and therapeutic development, and offers advanced training in a highly translational environment with strong industrial links.
Supervisor(s) for each project will be holding a 30-minute Q&A session during the first two weeks of December.
If you have any questions about a project, you are welcome to attend the relevant session.
If you would like to receive the webinar details, please submit a request via the Webinar Link Request Form.