期刊
MACHINE LEARNING IN MEDICAL IMAGING (MLMI 2019)
卷 11861, 期 -, 页码 265-273出版社
SPRINGER INTERNATIONAL PUBLISHING AG
DOI: 10.1007/978-3-030-32692-0_31
关键词
-
类别
资金
- UK Engineering and Physical Sciences Research Council [EP/R014507/1]
- Medical Research Council 515 [MR/J004146/1]
- European Research Council [GAP: 670428 - BRAIN2MIND NEUROCOMP]
In neural decoding, there has been a growing interest in machine learning on functional magnetic resonance imaging (fMRI). However, the size discrepancy between the whole-brain feature space and the training set poses serious challenges. Simply increasing the number of training examples is infeasible and costly. In this paper, we propose a domain adaptation framework for whole-brain fMRI (DawfMRI) to improve whole-brain neural decoding on target data leveraging source data. DawfMRI consists of two steps: (1) source and target feature adaptation, and (2) source and target classifier adaptation. We evaluate its four possible variations, using a collection of fMRI datasets from OpenfMRI. The results demonstrated that appropriate choices of source domain can help improve neural decoding accuracy for challenging classification tasks. The best-case improvement is 10.47% (from 77.26% to 87.73%). Moreover, visualising and interpreting voxel weights revealed that the adaptation can provide additional insights into neural decoding.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据