Limits...
Validation of Shared and Specific Independent Component Analysis (SSICA) for Between-Group Comparisons in fMRI

View Article: PubMed Central - PubMed

ABSTRACT

Independent component analysis (ICA) has been widely used to study functional magnetic resonance imaging (fMRI) connectivity. However, the application of ICA in multi-group designs is not straightforward. We have recently developed a new method named “shared and specific independent component analysis” (SSICA) to perform between-group comparisons in the ICA framework. SSICA is sensitive to extract those components which represent a significant difference in functional connectivity between groups or conditions, i.e., components that could be considered “specific” for a group or condition. Here, we investigated the performance of SSICA on realistic simulations, and task fMRI data and compared the results with one of the state-of-the-art group ICA approaches to infer between-group differences. We examined SSICA robustness with respect to the number of allowable extracted specific components and between-group orthogonality assumptions. Furthermore, we proposed a modified formulation of the back-reconstruction method to generate group-level t-statistics maps based on SSICA results. We also evaluated the consistency and specificity of the extracted specific components by SSICA. The results on realistic simulated and real fMRI data showed that SSICA outperforms the regular group ICA approach in terms of reconstruction and classification performance. We demonstrated that SSICA is a powerful data-driven approach to detect patterns of differences in functional connectivity across groups/conditions, particularly in model-free designs such as resting-state fMRI. Our findings in task fMRI show that SSICA confirms results of the general linear model (GLM) analysis and when combined with clustering analysis, it complements GLM findings by providing additional information regarding the reliability and specificity of networks.

No MeSH data available.


Related in: MedlinePlus

Hybrid fMRI data generation. Top row: the source patches used to generate the hybrid fMRI data set. From left to right, it illustrates the patches (or components), which are located in the cerebellum, sensorimotor area, anterior cingulate and lingual regions, paracingulate gyrus and precuneus, and bilateral opercular cortex and middle visual area. Middle and bottom rows show the t-statistics maps of those components, respectively, extracted by the SSICA and the regular time-concatenation gICA approach, which show the highest correlation with the ground truth patches. In this example, the 2 patches shown on columns 1 and 2 were specific to group-1, the 2 patches on columns 4 and 5 were specific to group-2 and the one on column 3 was shared between groups. The specific maps are generated using two-sample t-statistics, while the shared map is generated using one-sample t-statistics. The SSICA results are more similar to the ground truth specific patches compared to the regular gICA approach.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5037228&req=5

Figure 2: Hybrid fMRI data generation. Top row: the source patches used to generate the hybrid fMRI data set. From left to right, it illustrates the patches (or components), which are located in the cerebellum, sensorimotor area, anterior cingulate and lingual regions, paracingulate gyrus and precuneus, and bilateral opercular cortex and middle visual area. Middle and bottom rows show the t-statistics maps of those components, respectively, extracted by the SSICA and the regular time-concatenation gICA approach, which show the highest correlation with the ground truth patches. In this example, the 2 patches shown on columns 1 and 2 were specific to group-1, the 2 patches on columns 4 and 5 were specific to group-2 and the one on column 3 was shared between groups. The specific maps are generated using two-sample t-statistics, while the shared map is generated using one-sample t-statistics. The SSICA results are more similar to the ground truth specific patches compared to the regular gICA approach.

Mentions: We randomly divided the twelve subjects into two groups, resulting in 12 resting-state functional runs (two runs for each of six subjects) in each group. To generate hybrid fMRI data, we added focal patches of activations to the sub-sampled normalized functional data of each subject during the resting-state condition. As shown in Figure 2 top row, five different patches configurations were generated. Each patch was actually composed of 2 or 3 distinct volumes of interest, consisting in clusters of neighboring voxels, which will be further denoted as blobs in this study. So each patch was composed of 2 or 3 blobs.


Validation of Shared and Specific Independent Component Analysis (SSICA) for Between-Group Comparisons in fMRI
Hybrid fMRI data generation. Top row: the source patches used to generate the hybrid fMRI data set. From left to right, it illustrates the patches (or components), which are located in the cerebellum, sensorimotor area, anterior cingulate and lingual regions, paracingulate gyrus and precuneus, and bilateral opercular cortex and middle visual area. Middle and bottom rows show the t-statistics maps of those components, respectively, extracted by the SSICA and the regular time-concatenation gICA approach, which show the highest correlation with the ground truth patches. In this example, the 2 patches shown on columns 1 and 2 were specific to group-1, the 2 patches on columns 4 and 5 were specific to group-2 and the one on column 3 was shared between groups. The specific maps are generated using two-sample t-statistics, while the shared map is generated using one-sample t-statistics. The SSICA results are more similar to the ground truth specific patches compared to the regular gICA approach.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5037228&req=5

Figure 2: Hybrid fMRI data generation. Top row: the source patches used to generate the hybrid fMRI data set. From left to right, it illustrates the patches (or components), which are located in the cerebellum, sensorimotor area, anterior cingulate and lingual regions, paracingulate gyrus and precuneus, and bilateral opercular cortex and middle visual area. Middle and bottom rows show the t-statistics maps of those components, respectively, extracted by the SSICA and the regular time-concatenation gICA approach, which show the highest correlation with the ground truth patches. In this example, the 2 patches shown on columns 1 and 2 were specific to group-1, the 2 patches on columns 4 and 5 were specific to group-2 and the one on column 3 was shared between groups. The specific maps are generated using two-sample t-statistics, while the shared map is generated using one-sample t-statistics. The SSICA results are more similar to the ground truth specific patches compared to the regular gICA approach.
Mentions: We randomly divided the twelve subjects into two groups, resulting in 12 resting-state functional runs (two runs for each of six subjects) in each group. To generate hybrid fMRI data, we added focal patches of activations to the sub-sampled normalized functional data of each subject during the resting-state condition. As shown in Figure 2 top row, five different patches configurations were generated. Each patch was actually composed of 2 or 3 distinct volumes of interest, consisting in clusters of neighboring voxels, which will be further denoted as blobs in this study. So each patch was composed of 2 or 3 blobs.

View Article: PubMed Central - PubMed

ABSTRACT

Independent component analysis (ICA) has been widely used to study functional magnetic resonance imaging (fMRI) connectivity. However, the application of ICA in multi-group designs is not straightforward. We have recently developed a new method named “shared and specific independent component analysis” (SSICA) to perform between-group comparisons in the ICA framework. SSICA is sensitive to extract those components which represent a significant difference in functional connectivity between groups or conditions, i.e., components that could be considered “specific” for a group or condition. Here, we investigated the performance of SSICA on realistic simulations, and task fMRI data and compared the results with one of the state-of-the-art group ICA approaches to infer between-group differences. We examined SSICA robustness with respect to the number of allowable extracted specific components and between-group orthogonality assumptions. Furthermore, we proposed a modified formulation of the back-reconstruction method to generate group-level t-statistics maps based on SSICA results. We also evaluated the consistency and specificity of the extracted specific components by SSICA. The results on realistic simulated and real fMRI data showed that SSICA outperforms the regular group ICA approach in terms of reconstruction and classification performance. We demonstrated that SSICA is a powerful data-driven approach to detect patterns of differences in functional connectivity across groups/conditions, particularly in model-free designs such as resting-state fMRI. Our findings in task fMRI show that SSICA confirms results of the general linear model (GLM) analysis and when combined with clustering analysis, it complements GLM findings by providing additional information regarding the reliability and specificity of networks.

No MeSH data available.


Related in: MedlinePlus