Klaus Hermann Maier-Hein
Affiliated Professor of Computer Vision
Research Interests
Professor Maier-Hein's teaching and research interests span methodological advances in medical image computing for quantitative analysis of biomedical imaging data, particularly in oncology. He develops machine learning methods for semantic segmentation and object detection in volumetric images, complemented by representation learning and multimodal modeling across imaging and clinical data. Professor Maier-Hein's central goal is to bridge algorithmic innovation and clinical imaging research. Email
In addition to his position at MBZUAI, Professor Maier-Hein is a Full Professor (W3) at Heidelberg University, Head of the Division of Medical Image Computing at the German Cancer Research Center (DKFZ), and Head of the Section for Pattern Analysis and Learning at Heidelberg University Hospital. At DKFZ, he also is one of the Managing Directors of DKFZ Digital Oncology, overseeing research strategy across 14 divisions with more than 400 scientists.
- Habilitation, Ruprecht-Karls-University Heidelberg, Germany
- Doctor of Science (Dr. rer. nat.), Ruprecht-Karls-University Heidelberg, Germany
- Diplom-Informatiker, Karlsruhe Institute of Technology, Germany
- Master in Science, Federal University of Santa Catarina, Brazil
- Minor in Applied Studies of Culture and Society, Karlsruhe Institute of Technology, Germany
- Winner or top-ranked team in more than 30 international medical image analysis challenges (MICCAI, CVPR, Kaggle), including multiple first-place finishes, 2023–2025
- Rising Stars of Science Award (Research.com, Rank 6 in Germany), 2025
- BVM CHILI Award for Best Ph.D. Thesis (as supervisor of Fabian Isensee), 2021
- Johann Peter Süssmilch Medal (GMDS), 2015
- German High Tech Champions Award in Medical Imaging (Federal Ministry of Education and Research / RSNA), 2014
- Best Scientific Thesis of the Year Award (BVM Conference), 2011
- Young Academics Award (NeuroWiss), 2010
- Disch et al.: “CRONOS: Continuous Time Reconstruction for 4D Medical Longitudinal Series", ICCV, 2026.
- Rokuss et al.: “LesionLocator: Zero-Shot Universal Tumor Segmentation and Tracking in 3D Whole-Body Imaging”, CVPR, 2025.
- Antonelli et al.: “The Medical Segmentation Decathlon”, Nature Communications, 2022.
- Isensee et al.: “nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation”, Nature Methods, 2021.
- Baumgartner et al.: “nnDetection: A Self-configuring Method for Medical Object Detection”, MICCAI, 2021.
- Selected recent works on foundation models, large-scale learning, and benchmarking in medical imaging (CVPR, MICCAI 2023–2025), focusing on scalable, general-purpose AI systems for 3D and longitudinal data.