About the institute: The Institute of AI for Health (AIH) at Helmholtz Munich is part of the Computational Health Center, dedicated to advancing artificial intelligence and machine learning for medical applications. Its research focuses on modeling disease dynamics, developing new therapies, biomedical image analysis, and integrating multi-omics data for precision medicine. AIH’s mission is to translate biomedical discoveries into transformative medical solutions, shaping the future of healthcare.
Principal investigator : Peter Horvath, PhD, DSc
Persons involved:
- Alexey Surnov, PhD
- David Csikos
Role within Consortium:
Lead of WP3 “AI-based digital pathology to assess tumor heterogeneity at SCP/clonal level composition”.
Task Description:
- WP3: To harness AI in digital pathology for analyzing tumor heterogeneity at both single-cell and clonal levels. This includes creating an AI-driven workflow for digital pathology and introducing a novel approach for single-cell and clone isolation, further enhanced with ultra-sensitive proteomics via advanced Deep Visual Proteomics. The objective is to complement bulk proteomics data with precise insights into protein biomarkers in their spatial context, together with information on tissue architecture, cell organization, and the microenvironment derived from digital pathology images. A key challenge will be designing deep learning methods to process high-resolution tissue slides, including the development of a deep convolutional neural network (DCNN) to evaluate tissue region similarity (TRS) using state-of-the-art self-supervised and weakly supervised techniques with visual transformers. Special emphasis will be placed on tailoring augmentation strategies for different staining protocols.