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

Results of GLM, GIFT, and SSICA in extracting active clusters during finger-tapping task. (A)Z-score maps showing active clusters for different contrasts during finger-tapping tasks using GLM analysis. Only the slices with significant activity are shown. The left (Visual), middle (Auditory), and right (Motor) panels show the significantly active clusters for (VFT − RS) − (AFT − RS), (AFT − RS) − (VFT − RS), and (AFT − RS) + (VFT − RS) contrasts, respectively (corrected using Gaussian random field theory, cluster level p < 0.05). (B) Left, middle, and right panels show the results of GIFT software for the contrasts of (VFT − AFT) on visual, (AFT − VFT) on auditory, and (AFT + VFT) on motor network, respectively. (C) Left, middle and right panels show t-statistics applied on back-reconstructed SSICA maps corresponding to specific components of VFT, AFT, and a shared component comprised of motor network, respectively. The statistical maps reported in (A) are generated based on fMRI data in AFT, VFT, and RS conditions, while the statistical maps in (B) and (C) are calculated based on the data in AFT and VFT conditions only. All ICA maps are back-reconstructed and resulting group level t-maps were thresholded at t = 3.17, p < 0.005, uncorrected.
© Copyright Policy
Related In: Results  -  Collection

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

Figure 8: Results of GLM, GIFT, and SSICA in extracting active clusters during finger-tapping task. (A)Z-score maps showing active clusters for different contrasts during finger-tapping tasks using GLM analysis. Only the slices with significant activity are shown. The left (Visual), middle (Auditory), and right (Motor) panels show the significantly active clusters for (VFT − RS) − (AFT − RS), (AFT − RS) − (VFT − RS), and (AFT − RS) + (VFT − RS) contrasts, respectively (corrected using Gaussian random field theory, cluster level p < 0.05). (B) Left, middle, and right panels show the results of GIFT software for the contrasts of (VFT − AFT) on visual, (AFT − VFT) on auditory, and (AFT + VFT) on motor network, respectively. (C) Left, middle and right panels show t-statistics applied on back-reconstructed SSICA maps corresponding to specific components of VFT, AFT, and a shared component comprised of motor network, respectively. The statistical maps reported in (A) are generated based on fMRI data in AFT, VFT, and RS conditions, while the statistical maps in (B) and (C) are calculated based on the data in AFT and VFT conditions only. All ICA maps are back-reconstructed and resulting group level t-maps were thresholded at t = 3.17, p < 0.005, uncorrected.

Mentions: Figure 8A shows the results of standard GLM analysis as described in Section Analysis of Finger-Tapping fMRI Dataset on the fMRI data acquired during finger-tapping conditions cued with visual and auditory stimuli. As expected, the between-condition contrasts of (VFT − RS) − (AFT − RS) and (AFT − RS) − (VFT − RS) included mostly clusters of activity in the visual and auditory areas of the brain, respectively. The auditory network was found unilateral within the right hemisphere. The contrast representing the average activity during both conditions compared to the baseline, i.e., (AFT − RS) + (VFT − RS), included clusters in the motor network comprising M1, SI, premotor cortex, and supplementary motor area.


Validation of Shared and Specific Independent Component Analysis (SSICA) for Between-Group Comparisons in fMRI
Results of GLM, GIFT, and SSICA in extracting active clusters during finger-tapping task. (A)Z-score maps showing active clusters for different contrasts during finger-tapping tasks using GLM analysis. Only the slices with significant activity are shown. The left (Visual), middle (Auditory), and right (Motor) panels show the significantly active clusters for (VFT − RS) − (AFT − RS), (AFT − RS) − (VFT − RS), and (AFT − RS) + (VFT − RS) contrasts, respectively (corrected using Gaussian random field theory, cluster level p < 0.05). (B) Left, middle, and right panels show the results of GIFT software for the contrasts of (VFT − AFT) on visual, (AFT − VFT) on auditory, and (AFT + VFT) on motor network, respectively. (C) Left, middle and right panels show t-statistics applied on back-reconstructed SSICA maps corresponding to specific components of VFT, AFT, and a shared component comprised of motor network, respectively. The statistical maps reported in (A) are generated based on fMRI data in AFT, VFT, and RS conditions, while the statistical maps in (B) and (C) are calculated based on the data in AFT and VFT conditions only. All ICA maps are back-reconstructed and resulting group level t-maps were thresholded at t = 3.17, p < 0.005, uncorrected.
© Copyright Policy
Related In: Results  -  Collection

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

Figure 8: Results of GLM, GIFT, and SSICA in extracting active clusters during finger-tapping task. (A)Z-score maps showing active clusters for different contrasts during finger-tapping tasks using GLM analysis. Only the slices with significant activity are shown. The left (Visual), middle (Auditory), and right (Motor) panels show the significantly active clusters for (VFT − RS) − (AFT − RS), (AFT − RS) − (VFT − RS), and (AFT − RS) + (VFT − RS) contrasts, respectively (corrected using Gaussian random field theory, cluster level p < 0.05). (B) Left, middle, and right panels show the results of GIFT software for the contrasts of (VFT − AFT) on visual, (AFT − VFT) on auditory, and (AFT + VFT) on motor network, respectively. (C) Left, middle and right panels show t-statistics applied on back-reconstructed SSICA maps corresponding to specific components of VFT, AFT, and a shared component comprised of motor network, respectively. The statistical maps reported in (A) are generated based on fMRI data in AFT, VFT, and RS conditions, while the statistical maps in (B) and (C) are calculated based on the data in AFT and VFT conditions only. All ICA maps are back-reconstructed and resulting group level t-maps were thresholded at t = 3.17, p < 0.005, uncorrected.
Mentions: Figure 8A shows the results of standard GLM analysis as described in Section Analysis of Finger-Tapping fMRI Dataset on the fMRI data acquired during finger-tapping conditions cued with visual and auditory stimuli. As expected, the between-condition contrasts of (VFT − RS) − (AFT − RS) and (AFT − RS) − (VFT − RS) included mostly clusters of activity in the visual and auditory areas of the brain, respectively. The auditory network was found unilateral within the right hemisphere. The contrast representing the average activity during both conditions compared to the baseline, i.e., (AFT − RS) + (VFT − RS), included clusters in the motor network comprising M1, SI, premotor cortex, and supplementary motor area.

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 &ldquo;shared and specific independent component analysis&rdquo; (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 &ldquo;specific&rdquo; 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