The effect of preprocessing pipelines in subject classification and detection of abnormal resting state functional network connectivity using group ICA

Abstract

Resting state functional network connectivity (rsFNC) derived from functional magnetic resonance (fMRI) imaging is emerging as a possible biomarker to identify several brain disorders. Recently it has been pointed out that methods used to preprocess head motion variance might not fully remove all unwanted effects in the data. Proposed processing pipelines locate the treatment of head motion effects either close to the beginning or as one of the final steps. In this work, we assess several preprocessing pipelines applied in group independent component analysis (gICA) methods to study the rsFNC of the brain. The evaluation method utilizes patient/control classification performance based on linear support vector machines and leave-one-out cross validation. In addition, we explored group tests and correlation with severity measures in the patient population. We also tested the effect of removing high frequencies via filtering. Two real data cohorts were used: one consisting of 48 mTBI and one composed of 21 smokers, both with their corresponding matched controls. A simulation procedure was designed to test the classification power of each pipeline. Results show that data preprocessing can change the classification performance. In real data, regressing motion variance before gICA produced clearer group differences and stronger correlation with nicotine dependence.

Detection of Mild Traumatic Brain Injury by Machine Learning Classification Using Resting State Functional Network Connectivity and Fractional Anisotropy

“…36 In addition, these regressors are not part of the standard pre-processing procedure followed in gICA. 26,27,31 A full width half maximum Gaussian kernel of 6 mm was then used for spatial smoothing.…”

“…26 We followed the suggestion that resting state functional network connectivity (rsFNC) bias could be minimized by processing head motion variance before using gICA 27 rather than after gICA as has been typically done previously in the ICA literature. 20,26,[28][29][30] After investigating the effects of several pre-processing pipelines, Vergara and associates 31 found that pre-processing head movement before gICA leads to more accurate detection of group differences and higher classification accuracy after testing using both simulated and real data. Simulations in the same study also suggest that preprocessing head movement after gICA results in higher residual correlation with head movement and higher spatial variance between subjects.…”

“…Simulations in the same study also suggest that preprocessing head movement after gICA results in higher residual correlation with head movement and higher spatial variance between subjects. By following the guidelines presented by Vergara and associates, 31 we expected to achieve a more accurate classification because of a smaller residual head movement confounds.…”

The effect of preprocessing in dynamic functional network connectivity used to classify mild traumatic brain injury

“…These two states also exhibit higher occupancy rates and occupancy rate differences between mTBI and HC. Together, these observations suggest that PA and PB are appropriate pipelines for the detection of increased static connectivity in mTBI samples (Sours et al., 2013; Vergara, Mayer, Damaraju, Hutchison, & Calhoun, 2017; Vergara et al., 2015) previously mentioned. However, this panorama is not fully clear for the results in Figure 6.…”

“…The relationship between dFNC and diagnosis was higher for PC and PD in State 2, but higher for PA in State 3. Although PC and PD exhibited increased relationship with diagnosis in State 2, the occupancy rate is smaller for this state which explains why this increased sensitivity with diagnosis is not similarly observed in static connectivity (Vergara, Mayer, Damaraju, Hutchison, et al., 2017). These results suggest that the difference between static and dynamic connectivity is rooted on the difference in occupancy rate instead of connectivity strength.…”