国家数学与交叉科学中心合肥分中心报告【Ting Chen】

报告题目：Group-wise shape atlas construction

报告人：Ting Chen 

      Roche Tissue Diagnostics

时  间： 2015年5月13号（星期三）下午2点

地  点：东区管理科研楼  365英国上市官网1208室

内容提要：

The first part of the talk will present a novel and robust technique for group-wise point set atlas construction, simultaneous registration of point sets with unknown correspondence. A new information theoretic measure based on Havrda-Charvát cumulative residual entropy is proposed to quantify the dis-similarity between CDFs estimated from each point-set in the given population of point sets.  Closed-form solutions for both the cost function and the analytic gradient with respect to the non-rigid registration parameters have been derived.

The second part of the talk will discuss a technique for constructing a shape complex atlas using an information geometry framework. We represent the boundary of the entire shape complex using the zero level set of a distance function. We then leverage the well known relationship between the stationary state wave function of the Schrodinger equation and the eikonal equation for distance transform. A shape complex atlas is constructed by first computing the Karcher mean of the wave functions, followed by an inverse mapping of the estimated mean back to the space of distance transforms in order to realize the atlas.

An interesting extension of this work to point set registration can be found here:

http://www.cv-foundation.org/openaccess/content_cvpr_2014/papers/Deng_A_Riemannian_Framework_2014_CVPR_paper.pdf

Speaker: Dr. Ting Chen is currently working as a Senior Imaging Scientist at Roche Tissue Diagnostics. She was a postdoctoral researcher at IBM Research Center before joining Roche in 2012. She received Ph.D. in Computer Engineering from University of Florida in 2011 and B.E. degree in Computer Science and Technology from University of Science and Technology of China in 2006. Her research interests are Medical Image Analysis, Machine learning and Computer Vision. She serves as a reviewer for top journals and conferences such as IJCV, MedIA, CVPR, ICCV, MICCAI, ISBI etc. since 2007 and is a committee member of MICCAI workshop on Machine Learning in Medical Imaging since 2012. She is the winner of 2011 MICCAI Young Scientist Award and holds more than 12 patents and invention disclosures.

欢迎感兴趣的老师参加！