Over the past two decades, technological advances have enabled efficient digitization of histological tissue sections, thus helping to simplify workflows in pathology laboratories around the world. This digitization has enabled the creation of large digital slide libraries, the best known of which is probably the Cancer Genome Atlas (National Institutes of Health, USA), which gives researchers around the world free access to a richly curated and annotated dataset of pathology images.

Digital tissue sections, so-called Whole Slide Images, allow the application of computer algorithms to analyze features of the tissue that are of interest to us. Factors such as the spatial proximity or orientation of certain cell types to each other play a decisive role here. Our goal is to differentiate many different tissue types within a slide, for example, to quantify the fragmentation of tumor tissue after a specific treatment. To this end, we use advanced techniques such as deep learning, a subdiscipline of artificial intelligence, to automatically classify tissues or precisely segment cell nuclei to gain detailed insights into tissue structure. Linking these data with clinical, outcome, and genomic information enables extensive research activities in the field of precision medicine. Various methods and technologies are used, including:

• Tissue Microarray (TMA)
• Whole-Slide Imaging (WSI)
• Image segmentation and classification
• Artificial Intelligence and Machine Learning
• Integrative analysis of multiple data sets from different technology platforms (integromics)