Bjoern Menze is the newly appointed Helmut Horten Professor for Biomedical Image Analysis and Machine Learning at the UZH Department of Quantitative Biomedicine. He plans to apply neuroimaging to key applications areas, including modeling of tumor growth and the extraction and analysis of cerebrovascular networks.
Bjoern Menze comes to Zurich from the TU Munich, where he was Professor of Informatics. His research field is medical image computing, where he explores topics at the interface of machine learning, medical imaging, and image-based diagnostics.
His work aims to transform descriptive biomedical images into models that allow analysis of the underlying physiological and patho-physiological processes. He is also interested in applying such models to databases of large data numbers of patients to learn about correlations between model features and disease patterns at a population scale.
For neuroimaging, Menze plans to target key application areas of the modeling of tumor growth and the extraction and analysis of cerebrovascular networks. In recent work, he has applied tumor growth modeling to the design of personalized radiotherapy approaches for glioblastoma and the estimation of the fast-neural parameter for these tumor models.
A further research goal is the development of new computational approaches to analyzing large 3D vascular data sets. In this approach, Menze aims at segmenting and tracing vascular structures, as well as at extracting and analyzing the full vascular network of the brain for both clinical and preclinical research applications.
- Lipkova et al. 2019 – Personalized Radiotherapy Design for Glioblastoma: Integrating Mathematical Tumor Models, Multimodal Scans, and Bayesian Inference Article
- Ezhov et al. 2019 – Neural parameters estimation for brain tumor growth modeling. MICCAI 2019 pdf
- Rempfler et al. 2015 – Reconstructing Cerebrovascular Networks under Local Physiological Constraints by Integer Programming. Medical Image Analysis MICCAI 2014 pdf
- Todorov et al. 2020 Machine learning analysis of whole mouse brain vasculature. Nature Methods 17, 442–449 Abstract
Image: Modified from A. Heddergott / Munich