Chao Chen (Stony Brook University)
Topology-Driven Learning for Biomedical Imaging Informatics
Abstract
Thanks to decades of technology development, we are now able to visualize in high quality complex biomedical structures such as neurons, vessels, trabeculae and breast tissues. We need innovative methods to fully exploit these structures, which encode important information about underlying biological mechanisms. In this talk, we explain how topology, i.e., connected components, handles, loops, and branches, can be seamlessly incorporated into different parts of a learning pipeline. Under the hood is a formulation of the topological computation as a differentiable operator, based on the theory of topological data analysis. This leads to a series of novel methods for segmentation, generation, and analysis of these topology-rich biomedical structures.
Biography
Dr. Chao Chen is an assistant professor at Stony Brook University. His research interest spans biomedical image analysis, topological data analysis (TDA) and machine learning. He develops principled learning methods inspired by the theory from TDA, such as persistent homology and discrete Morse theory. These methods address problems in biomedical image analysis and machine learning from a unique topological view. His research results have been published in major venues in machine learning, computer vision, and medical image analysis. He served as an area chair/senior PC in these venues including MICCAI, AAAI, CVPR and NeurIPS. He was awarded an NSF CAREER award to develop topology-driven learning for biomedical images.
Cardiac Motion Analysis using Transport Distance
Abstract
In this talk, we show that transport detected by Wasserstein distance along a temporal sequence of volumetric cardiac images separates heart wall motion from images. Sequential application of registration provides tracking of temporal deformation of organs. Wasserstein distance allows us to detect deformation between a pair of images without computing geometric transforms. Sequential application of registration provides tracking of temporal deformation of organs from appearances. Numerical examples show the method aligns sequences of temporal volumetric MRI heart images and tracks the motion of heart valves and heart wall in MRI temporal image sequences.
Biography
Atsushi Imiya is a professor emeritus of Chiba University. He had completed his doctor of engineering degree (Dr Engg.) on computer science from Tokyo Institute of Technology in 1985. Since then, he had hold teaching positions in Kanazawa University, NII Tokyo and Chiba University. He is a fellow of IAPR and a fellow of IEICE, Japan. His main interests are discrete geometry, mathematical images analysis and their applications to medicine and biology. He had published more than 200 refereed papers in this field. Since 1998, he is the officer of a gifted high-school-student education programme in STEM at Chiba University.
Loading...