报告题目: Fiber orientation distribution estimation using a Peaceman-Rachford splitting metho
报告人:戴彧虹研究员
时间:2015-05-23
地点:数学院二楼北阶梯教室
摘要: n diffusion-weighted magnetic resonance imaging, the estimation of the orientations of multiple nerve fibers in each voxel (the fiber orientation distribution (FOD)) is a critical issue for exploring the connection of cerebral tissue. In this paper, we establish a convex semidefinite programming (CSDP) model for the FOD estimation. One feature of this new model is that it can ensure the statistical meaning of FOD. As a probability density function, FOD must be nonnegative and have a unit mass. To construct such a statistically meaningful FOD, we approximate it by a sum of squares (SOS) polynomial and impose the unitmass by a linear constraint. Another feature of the new model is that it combines the sparsity of nerve fibers. Due to the sparsity of the orientations of nerve fibers in cerebral white matter, we raise a new heuristic regularization, which is inspired by the Zeigenvalue of a symmetric tensor that closely relates to the SOS polynomial. To solve the CSDP efficiently, we propose a new Peaceman-Rachford splitting method and prove its global convergence. Numerical experiments on synthetic data show that the new approach gives a more accurate estimation of fiber orientations, and in real-world human brain study, the contour profiles of fibers reconstructed by the novel approach coincide with the results in neuroanatomy. 。
报告时间:5月23日上午10点
报告人简介:
戴彧虹,1992年毕业于北京理工大学应用数学系,1997年在中国科学院计算数学研究所获得博士学位(导师:袁亚湘)。毕业后,他一直在中国科学院数学与系统科学研究院工作,2006年被聘为创新基地研究员,2012年被聘为二级研究员。他主要研究非线性优化理论、算法及其应用,已出版专著一本,发表论文七十多篇。曾获中国数学会第五届钟家庆数学奖(1998年),国家自然科学奖二等奖(2006,排名第二),德国洪堡奖学金(2004),第十届中国青年科技奖(2007),国家杰出青年科学基金(2011),国际通信大会最佳论文奖(2011)。曾访问英国剑桥大学、邓迪大学、德国拜罗伊特大学等著名院校。目前,他担任中国运筹学会数学规划分会副理事长,中科院数学院优化与应用研究中心副主任,中国科学院数学与系统科学研究院计算数学与科学工程计算研究所所长助理。他还担任《ITOR》、《Science in China: Mathematics》、《APJOR》、《JSSC》、《JORC》等多个杂志编委。