Watanabe Takanori, Scott Clayton D, Kessler Daniel, Angstadt Michael, Sripada Chandra S
Dept. of EECS, University of Michigan, Ann Arbor, MI, 48109.
Dept. of Psychiatry, University of Michigan, Ann Arbor, MI, 48109.
Proc IEEE Int Conf Acoust Speech Signal Process. 2014 May;2014:5989-5993. doi: 10.1109/ICASSP.2014.6854753.
There is substantial interest in developing machine-based methods that reliably distinguish patients from healthy controls using high dimensional correlation maps known as (FC's) generated from resting state fMRI. To address the dimensionality of FC's, the current body of work relies on feature selection techniques that are blind to the spatial structure of the data. In this paper, we propose to use the fused Lasso regularized support vector machine to explicitly account for the 6-D structure of the FC (defined by pairs of points in 3-D brain space). In order to solve the resulting nonsmooth and large-scale optimization problem, we introduce a novel and scalable algorithm based on the alternating direction method. Experiments on real resting state scans show that our approach can recover results that are more neuroscientifically informative than previous methods.
利用静息态功能磁共振成像（fMRI）生成的称为功能连接（FC）的高维相关图来开发能够可靠地区分患者与健康对照的基于机器的方法，这引起了人们极大的兴趣。为了解决功能连接的维度问题，当前的研究工作依赖于对数据空间结构视而不见的特征选择技术。在本文中，我们建议使用融合套索正则化支持向量机来明确考虑功能连接的六维结构（由三维脑空间中的点对定义）。为了解决由此产生的非光滑和大规模优化问题，我们引入了一种基于交替方向法的新颖且可扩展的算法。对真实静息态扫描数据的实验表明，我们的方法能够获得比以前的方法更具神经科学信息的结果。
