Digital Biomarkers

Overview

Digital biomarkers are quantifiable parameters of patients that enable computer-aided targeted diagnostics and personalized medicine.
We develop AI tools to analyze pathology images and develop novel biomarkers in an experienced team with diverse scientific, clinical, and technical expertise.
We collaborate with a wide network of clinical, industry and academic partners in joint research and contract R&D projects in biomarker research and digital pathology.
Our software solutions aid in disease diagnostics, reducing the workload of pathologists, enabling more effective treatment, and ultimately reducing healthcare costs.
We support the entire development cycle comprising exploratory scientific research, development of usable prototypical software tools, and translation into marketable IVDs.

Digital Biomarker R&D. We support clinical and pharmaceutical research studies in data-driven identification of novel biomarkers based on images and other clinical data.
Image & Data Analysis. We develop algorithms for robust qualitative and quantitative analysis of microscopic images and clinically relevant data.
Software IVD Development. We develop practical software solutions and help translate them into marketable in-vitro diagnostics.
Image Registration. We enable you to combine information from multiple serial sections with award-winning accuracy and performance.
→ HistokatFusion

Meaningful Digital Biomarkers. We develop algorithms to quantify morphology and molecular information in its spatial context to explore clinically meaningful biomarkers.
→ Meaningful Digital Biomarkers

Foundation Models—Learning from Small Datasets.
Foundation models integrate multi-center knowledge for AI in personalized medicine.
→ Small Datasets in Personalized Medicine

Hemato-Oncology. We develop AI-driven algorithms for the analysis of blood and bone marrow cells to diagnose hemato-onlological diseases.
→ Hemato-oncology

Image Registration. We integrate multi-stain information from multiple slides using image registration for histopathology image analysis.
→ HistokatFusion

Federated Learning. We develop privacy-preserving federated AI models, enabling collaborative learning across multiple labs for joint research projects.

Data Acquisition and Curation. We have expertise in selecting representative and unbiased collections for AI development and data analysis in pathology.

Develop Software IVDs. We have many years of experience in the development of software solutions for clinical practice.
→ Develop Software IVDs

Data-Driven Discovery of Immune Contexture Biomarkers. Schwen et al. 2018. article
Artificial Intelligence in Pathology: From Prototype to Product. Homeyer et al. 2022. article
Recommendations on compiling test datasets for evaluating artificial intelligence solutions in pathology. Homeyer et al. 2022. article
The NCI Imaging Data Commons as a platform for reproducible research in computational pathology. Schacherer et al. 2023. article
Overcoming data scarcity in biomedical imaging with a foundational multi-task model. Schaefer et al. 2024. article

1st place in the UNICORN MICCAI 2025 Lighthouse Challenge for the best overall medical foundation model as well as in the sub-categories pathology and radiology. https://unicorn.grand-challenge.org
1st place in the ANHIR challenge on histological image registration. https://anhir.grand-challenge.org
2nd place in the LEOPARD challenge for prostate cancer risk prediction. https://leopard.grand-challenge.org/test-results/