AI in Medical Image Computing
Increasing availability of medical imaging data with respect to quantity, resolution and modalities, pose challenges to traditional processing and analysis methods. Large cohort studies such as the ADNI, Human Connectome Project, UK Biobank, Rhineland study etc. collect rich data on 10-thousands of individuals. In order to investigate etiology, progression, treatment, risk and preserving factors of neurodegenerative diseases we need descriptive features, obtained by scalable automatic processing methods. Traditional approaches that depend on non-linear atlas registration and segmentation, however, require long processing times (many hours up to a day for a single image) and are thus not efficient enough to handle big data or provide rapid results for personalised medicine. We develop fast deep learning approaches,for example, for full brain segmentation and cortical parcellation obtained in under 1 minute (FastSurfer). Furthermore, deep learning approaches can help accelerate demanding diffusion reconstruction and processing pipelines. Finally, we introduce advanced geometry-based features (such as shape and lateral asymmetry) as well as hippocampal thickness estimates for the localized and sensitive analysis of neuroanatomical changes.
Short Bio:Prof. Reuter is Director of Image Analysis at the German Center for Neurodegenerative Diseases (DZNE) in Bonn. He is Assistant Professor of Radiology and Assistant Professor of Neurology at the Harvard Medical School in Boston, USA. He, furthermore, directs the Laboratory for Computational Longitudinal Neuroimaging at the Martinos Center for Biomedical Imaging, Massachusetts General Hospital, and is Research Affiliate at the Computer Science and Artificial Intelligence Lab (CSAIL) of the Massachusetts Institute of Technology. Prof. Reuter obtained a degree in Mathematics and Computer Science in 2001 and a Ph.D. in Computational Geometry in 2005 from the Leibniz University Hanover, Germany. He pursued his postdoctoral studies with a Feodor Lynen Fellowship of the Alexander von Humboldt Foundation at the Massachusetts Institute of Technology until 2008 and subsequently moved to the Martinos Center for Biomedical Imaging at the Massachusetts General Hospital. He has received prizes and awards from international journals, workshops, competitions, industry and funding bodies, including a competitive career award of the National Institute of Health (NIH). Prof. Reuter’s research contributions include advanced non-rigid shape analysis and highly sensitive computational and AI methods for neuroimaging and computer-aided diagnosis. His methods are widely employed as part of the FreeSurfer software suite, for example, to uncover risk and preserving factors of neurodegenerative diseases or to assess disease modifying therapies. Furthermore, various large-cohort neuroimaging studies, such as the Alzheimer’s Disease Neuroimaging Initiative (ADNI), rely on his contributions for the automated quantification of relevant MR-markers.
Posted by: Hong Xu