Introduction

Immunotherapy has become a life-saving option for advanced cancer patients. However, only a minority of patients develop a durable response. Despite great efforts to explain the variable responses to immunotherapy and to optimize patient selection, current diagnostic tools cannot sufficiently guide clinical practice. This project combines state-of-the-art multiplexed immunofluorescence microscopy with the latest techniques in image processing and deep learning to advance the understanding of how cell interrelations in the tumor microenvironment affect the disease progression and treatment efficacy.

Starting from a large collection of acquired multispectral histology images, we aim to develop advanced interpretable AI-driven approaches for characterization of the structural organization and interrelations of different cell types, enabling reliable and explainable prediction of patient disease progression. Directions we are exploring include multi-channel image-based analysis, geometric data representations such as cell graphs and point clouds, multi-modal approaches with H&E histology, and extending the analysis to three dimensions through tissue volume reconstruction.

The project heavily relies on interdisciplinary competences and is conducted in close collaboration with Patrick Micke's group at the Department of Immunology, Genetics and Pathology (IGP) at Uppsala University.

More headers?

More info about methods/subprojects, publications, data, involved people, etc.


Funding

This research is funded by: