MIC-DKFZ
MIC-DKFZ is the Division of Medical Image Computing at the German Cancer Research Center. It develops machine-learning and information-processing methods that structure and quantify medical images, link imaging with clinical and biological parameters, and translate reliable AI into improved cancer diagnosis, treatment and patient care.
At a Glance
- Cancer research institutes and biomedical researchers
- Hospitals and clinical imaging groups
- Radiology, radiation oncology and pathology researchers
- Academic and industrial medical-AI developers
- +1 more
AI Tools by MIC-DKFZ
(1)
nnU-Net
Self Configuring Segmentation Framework
Discussions
No discussions yet
Be the first to start a discussion about MIC-DKFZ
Latest News
AI solutions from the DKFZ set new standards in medical image processing; DKFZ researchers won seven of eight MICCAI 2025 competitions in which they participated.
MIC-DKFZ awards page records wins in the RSNA Intracranial Aneurysm Detection, autoPET, PANTHER, ODELIA, Trials NCCT, Foundation Model and MAMA-MIA challenges at MICCAI 2025.
Fabian Isensee received the Leopoldina Award for Young Scientists; the awards page also records the Extraordinary Contributions Award for nnInteractive.
MIC-DKFZ contributors received an RSNA Certificate of Merit for the MITK AI-powered medical-image annotation workflow exhibit.
Products & Services
An open-source, self-configuring framework for supervised biomedical image segmentation. It fingerprints a dataset, configures suitable U-Net variants, and provides an end-to-end workflow from preprocessing through training, model selection and inference for 2D and 3D data.
A self-configuring framework for 3D volumetric medical object detection that can be applied to new datasets without manual intervention; it includes guides for multiple medical-imaging datasets.
A Python framework for data augmentation for 2D and 3D image classification and segmentation, including multi-modal and multi-channel augmentation.
An open-source tool for automatic white-matter bundle segmentation.
Market Position
MIC-DKFZ is positioned as an academic, open-science leader in generalizable medical imaging AI rather than as a conventional software vendor. Its differentiation is self-configuring, reproducible open-source systems such as nnU-Net and nnDetection, strong challenge benchmarking, and direct translation into clinical and federated research settings; it competes for influence with university medical-AI labs, Helmholtz/European imaging groups and commercial medical-imaging AI developers.
Leadership
Executive Team
Klaus Maier-Hein
Head of Division
Professor and head of MIC-DKFZ; his division leads research in machine learning, information processing and clinical medical-image analytics and co-initiates/co-coordinates the Helmholtz Imaging Platform.
Corinna Weiler
Scientific Manager
Scientific manager listed in MIC-DKFZ's current division team.
Board of Directors
Founding Story
MIC-DKFZ is a research division rather than a separately incorporated startup. Its stated vision is to improve healthcare through methodological advances in artificial intelligence and large-scale clinical implementation, with an emphasis on interpretable decisions, uncertainty, generalizability, robust representations and open science.
Business Model
Revenue Model
Public research division of DKFZ; its public materials describe grant-funded research, collaborations and open-source software rather than a commercial pricing or subscription model.
Target Markets
- Cancer research institutes and biomedical researchers
- Hospitals and clinical imaging groups
- Radiology, radiation oncology and pathology researchers
- Academic and industrial medical-AI developers
- Clinical research networks and federated data platforms
- Cancer diagnosis and screening from MRI, CT, PET/CT and endoscopy images
- Tumor, lesion, organ and anatomical-structure segmentation
- 3D medical object detection
- Treatment planning and therapy-response assessment
- Longitudinal imaging analysis and disease stratification
- Clinical research networks and federated medical-data analysis
- National Center for Tumor Diseases
- German Cancer Consortium
- Cancer Core Europe