Hemant Tagare, PhD
Professor of Radiology and Biomedical Imaging and of Biomedical Engineering
Research & Publications
Biography
News
Research Summary
Dr. Hemant D. Tagare's research interests include new and innovative approaches to bio-medical image analysis, especially in Cryogenic electron microscopy of proteins, MRI and ultrasound image segmentation, non-rigid registration, shape spaces, and image indexing, and each of the former's foundational issues.
Specialized Terms: Heart motion analysis
Extensive Research Description
Dr. Hemant D. Tagare's research interests include new and innovative approaches to bio-medical image analysis, especially in Cryogenic electron microscopy of proteins, MRI and ultrasound image segmentation, non-rigid registration, shape spaces, and image indexing, and each of the former's foundational issues.
Dr. Tagare works on the theoretical and practical problems in bio-medical image analysis. His research focuses on image segmentation, non-rigid registration, 3D reconstruction, and shape theory. Within segmentation, Dr. Tagare mainly studies cardiac ultrasound and brain MRI segmentation, and especially the development of new methodology that draws on machine-learning and advanced numerical optimization techniques. Regarding non-rigid registration, he is developing an axiomatic framework for registration that is applied to brain deformation and atlas building. In 3D reconstruction, Dr. Tagare focuses his research on reconstructing protein structure from Cryogenic Electron Microscopy. In the field of shape theory, his work addresses questions of the topology and geometry of affine shape spaces and non-rigid shape comparison. Dr. Tagare receives research funding from the N.I.H.
Research Interests
Biomedical Engineering; Magnetic Resonance Imaging; Radiology; Cryoelectron Microscopy
Selected Publications
- Symmetric Non-rigid Registration: A Geometric Theory and Some Numerical TechniquesTagare H.D., Groisser D., Skrinjar O., “Symmetric Non-rigid Registration: A Geometric Theory and Some Numerical Techniques”, Journal of Mathematical Imaging and Vision 2008.
- Comparison of EM-based and Level Set Partial Volume Segmentations of MR Brain ImagesTagare H. D., Chen Y., Fulbright R. K., “Comparison of EM-based and Level Set Partial Volume Segmentations of MR Brain Images”, (J. M. Reinhardt, J. P. W. Pluim, eds.), Medical Imaging 2208, Proc. SPIE vol6914, pp.69140N-1 - 69140N-7, 2008.
- Fast, Adaptive Expectation-Maximization Alignment for Cryo-EMTagare HD, Sigworth F, Barthel A. Fast, Adaptive Expectation-Maximization Alignment for Cryo-EM. 2008, 11: 855-862. PMID: 18982685, DOI: 10.1007/978-3-540-85990-1_103.
- Evaluation of a Cardiac Ultrasound Segmentation Algorithm Using a PhantomYue Y, Tagare HD, Madsen EL, Frank GR, Hobson MA. Evaluation of a Cardiac Ultrasound Segmentation Algorithm Using a Phantom. 2008, 11: 101-109. PMID: 18979737, PMCID: PMC2840385, DOI: 10.1007/978-3-540-85988-8_13.
- A Geometric Theory of Symmetric RegistrationTagare H. D., Groisser D., Skrinjar O. "A Geometric Theory of Symmetric Registration." Mathematical Methods in Biomedical Image Analysis, Intl. Conference on Computer Vision and Pattern Recognition, New York, NY, 2006.