学术报告：Multimodal Brain Imaging for the Computational Mapping of Neuroanatomy

报告时间：2015.3.27（星期五）上午10：00

报告地点：博习楼327

报告人：Dr. Yonggang Shi, Tenure-Track Assistant Professor of University of Southern California

邀请人：陈新建 特聘教授

报告摘要:In this talk, I will present our recent work on the automatedanalysis of multimodal MR images for large scale brain mapping. I will firstpresent a suite of novel algorithms for mapping brain structures usingintrinsic geometry. The key idea in our method is the use ofLaplace-Beltrami (LB) eigenfunctions for modeling brain shapes, such ashippocampus and cortex. These tools have the advantage of being invariant topose and scale variances, and robust todeformations from development andpathology. Using the LB eigenfunctions and topology-preserving evolution, wehave developed a robust approach for surface reconstruction from segmentedmasks. This method can remove outliers while accurately retaining volumeinformation. For the challenging problem of cortical surface reconstruction,we have developed a unified approach for the joint correction of geometricand topological outliers with the Reeb graph of LB eigenfunctions. By usingthe LB embeddings of surfaces, we have developed a novel and generalapproach for surface mapping via the optimization of their conformalmetrics. Based on these cutting-edge algorithms for image and shapeanalysis, completely automated workflows have been created for the largescale analysis of brain morphometry. In our current research, these intrinsic modeling techniques are being extended to multimodal imageanalysis for the moreaccurateand robust mapping of brain structure andfunction. Using the reconstructed cortical surfaces, we have developed moreaccurate ways of normalizing cerebral blood perfusion (CBF) images withcortical thickness and area, and successfully applied them to map sexdifferences in brain development. For the analysis of brain connectivity, wedeveloped a novel algorithm for fiber orientation distribution (FOD)reconstruction that can be applied to diffusion imaging data collected froma wide range of acquisition schemes. With the help of FODs and intrinsicanalysis, we are able to automatically extract fiber bundles withsignificantly improved details androbustnessusing the state-of-the-artdata from the Human Connectome Project.

报告人简介: Dr.Yonggang Shi received his Bachelor and Master degree inElectricalEngineering from the Southeast University of China in 1996 and 1999 respectively. He received his Ph.D. in Electrical Engineering from Boston University in 2005. From 2005 to 2009, he was a Post-Doctoral fellow at the Laboratory of Neuro Imaging (LONI) at UCLA. He was promoted to Assistant Professor at LONI in 2009. In July 2013, Dr. Shi was recruited to USC a tenure-track Assistant Professor of Neurology and Electrical Engineering. He joins USC along with other faculty members that previously had formed the Laboratory of Neuro Imaging (LONI) at UCLA to found the newly established editor Institute for Neuroimaging and Informatics (INI). Dr. Shi is an Associate Editor of IEEE Transactions on Image Processing. Dr. Shi was a winner of student paper competition at the 2005 ICASSP for his work on a fast level set algorithm. He also won the Best Paper Award at the 2008 MMBIA for his work on using Reeb graphs of LB eigenfunctions to construct shape skeletons.