Disjoint subspaces for common and distinct component analysis: Application to the fusion of multi-task FMRI data.
Disjoint subspaces for common and distinct component analysis: Application to the fusion of multi-task FMRI data.
复制标题
DOI：
10.1016/j.jneumeth.2021.109214
复制
发表时间：
2021-07-01
影响因子：
3
通讯作者：
Adali T
中科院分区：
文献类型：
--
作者：
Akhonda MABS;Gabrielson B;Bhinge S;Calhoun VD;Adali T
独立成分分析（ICA）等数据驱动方法对数据和多个数据集之间的关系做出很少的假设，因此对于医学成像数据的融合很有吸引力。用于多集融合的 ICA 的两个重要扩展是联合 ICA (jICA) 和多集典型相关分析和联合 ICA (MCCA-jICA) 技术。两种方法都假设相同的混合矩阵，强调多个数据集中常见的成分。然而，一般来说，人们希望拥有在数据集中通用且每个数据集都不同的组件。我们提出了一个通用框架，即使用 ICA 的不相交子空间分析 (DS-ICA)，它不仅可以识别和提取多个数据集中的共同成分，还可以提取不同的成分。该方法的关键组成部分是在后续分析之前识别这些子空间及其分离，这有助于建立更好的模型匹配并提供算法和顺序选择的灵活性。我们通过模拟和应用从健康对照者以及精神分裂症患者收集的多组功能磁共振成像 (fMRI) 任务数据，将 DS-ICA 与 jICA 和 MCCA-jICA 进行比较。结果表明，与 jICA 和 MCCA-jICA 相比，DS-ICA 估计了健康对照和患者之间更多的区分成分，并且具有更高的区分能力，显示有意义区域的激活差异。当应用于分类框架时，DS-ICA 估计的组件对于不同的数据集组合比其他两种方法具有更高的分类性能。这些结果表明 DS-ICA 是一种有效的多数据集融合方法。
Data-driven methods such as independent component analysis (ICA) makes very few assumptions on the data and the relationships of multiple datasets, and hence, are attractive for the fusion of medical imaging data. Two important extensions of ICA for multiset fusion are the joint ICA (jICA) and the multiset canonical correlation analysis and joint ICA (MCCA-jICA) techniques. Both approaches assume identical mixing matrices, emphasizing components that are common across the multiple datasets. However, in general, one would expect to have components that are both common across the datasets and distinct to each dataset. We propose a general framework, disjoint subspace analysis using ICA (DS-ICA), which identifies and extracts not only the common but also the distinct components across multiple datasets. A key component of the method is the identification of these subspaces and their separation before subsequent analyses, which helps establish better model match and provides flexibility in algorithm and order choice. We compare DS-ICA with jICA and MCCA-jICA through both simulations and application to multiset functional magnetic resonance imaging (fMRI) task data collected from healthy controls as well as patients with schizophrenia. The results show DS-ICA estimates more components discriminative between healthy controls and patients than jICA and MCCA-jICA, and with higher discriminatory power showing activation differences in meaningful regions. When applied to a classification framework, components estimated by DS-ICA results in higher classification performance for different dataset combinations than the other two methods. These results demonstrate that DS-ICA is an effective method for fusion of multiple datasets.
登录
查看更多内容
DOI：
10.1109/jproc.2015.2461624
发表时间：
2015-09-01
期刊：
Proceedings of the IEEE. Institute of Electrical and Electronics Engineers
影响因子：
--
作者：
Adali T;Levin-Schwartz Y;Calhoun VD
通讯作者：
Calhoun VD
DOI：
10.1109/titb.2008.923773
发表时间：
2009-09-01
影响因子：
--
作者：
Calhoun, Vince D.;Adali, Tuelay
通讯作者：
Adali, Tuelay
影响因子：
5.4
作者：
Anderson, Matthew;Fu, Geng-Shen;Adali, Tulay
通讯作者：
Adali, Tulay
影响因子：
18.6
作者：
James, Alex Pappachen;Dasarathy, Belur V.
通讯作者：
Dasarathy, Belur V.
影响因子：
17.6
作者：
Calhoun VD;Adalı T
通讯作者：
Adalı T