Brain State Decoding Based on fMRI Using Semisupervised Sparse Representation Classifications.

Brain State Decoding Based on fMRI Using Semisupervised Sparse Representation Classifications.
复制标题

使用半监督稀疏表示分类的基于 fMRI 的脑状态解码

DOI:
10.1155/2018/3956536
复制
发表时间:
2018
影响因子:
--
通讯作者:
Long Z
Long Z
中科院分区:
工程技术3区
文献类型:
--
作者:
Zhang J;Zhang C;Yao L;Zhao X;Long Z

文献摘要

参考文献

相似文献

多变量分类技术已广泛应用于功能磁共振成像(fMRI)的脑状态解码。由于fMRI数据的可变性和人类fMRI数据收集的局限性,对fMRI数据训练一个高效鲁棒的监督学习分类器并不容易。在各种分类技术中,稀疏表示分类器(SRC)在图像分类中表现出最先进的分类性能。然而,SRC很少应用于基于fmri的解码。本研究旨在使用未标记的测试样本改进SRC,使其能够有效地应用于基于fmri的解码。我们提出了一种基于平均系数的半监督学习SRC (semirc - ave)方法,该方法使用每个类的平均系数代替重建误差进行分类,并使用新的高置信度的测试数据有选择性地更新训练数据集,以提高SRC的性能。通过模拟和真实fMRI实验验证了semi - rc - ave的可行性和鲁棒性。模拟和真实fMRI实验结果表明,半监督学习方法(semi - rc - ave)显著优于平均系数监督学习方法(SRC- ave),且表现优于其他三种半监督学习方法。
Multivariate classification techniques have been widely applied to decode brain states using functional magnetic resonance imaging (fMRI). Due to variabilities in fMRI data and the limitation of the collection of human fMRI data, it is not easy to train an efficient and robust supervised-learning classifier for fMRI data. Among various classification techniques, sparse representation classifier (SRC) exhibits a state-of-the-art classification performance in image classification. However, SRC has rarely been applied to fMRI-based decoding. This study aimed to improve SRC using unlabeled testing samples to allow it to be effectively applied to fMRI-based decoding. We proposed a semisupervised-learning SRC with an average coefficient (semiSRC-AVE) method that performed the classification using the average coefficient of each class instead of the reconstruction error and selectively updated the training dataset using new testing data with high confidence to improve the performance of SRC. Simulated and real fMRI experiments were performed to investigate the feasibility and robustness of semiSRC-AVE. The results of the simulated and real fMRI experiments showed that semiSRC-AVE significantly outperformed supervised learning SRC with an average coefficient (SRC-AVE) method and showed better performance than the other three semisupervised learning methods.
DOI: 10.1371/journal.pone.0017191
发表时间: 2011-02-16
期刊: PloS one
影响因子: 3.7
作者:
Song S;Zhan Z;Long Z;Zhang J;Yao L
通讯作者: Yao L
基于平滑 L0 范数的改进稀疏分解,使用拉普拉斯核从 fMRI 数据中选择特征
DOI: 10.1016/j.jneumeth.2014.12.021
发表时间: 2015-04-30
影响因子: 3
作者:
Zhang, Chuncheng;Song, Sutao;Long, Zhiying
通讯作者: Long, Zhiying
DOI: 10.1109/lgrs.2015.2481181
发表时间: 2015-12-01
影响因子: 4.8
作者:
Zou, Jinyi;Li, Wei;Du, Qian
通讯作者: Du, Qian
DOI: 10.1109/tpami.2008.79
发表时间: 2009-02-01
影响因子: 23.6
作者:
Wright, John;Yang, Allen Y.;Ma, Yi
通讯作者: Ma, Yi
DOI: 10.1016/j.compbiomed.2015.08.017
发表时间: 2015-11-01
影响因子: 7.7
作者:
Shin, Younghak;Lee, Seungchan;Lee, Heung-No
通讯作者: Lee, Heung-No