In this study, we investigate if differences in interaction between different brain regions for subjects with autismspectrum disorder (ASD) and healthy controls can be captured using resting-state fMRI. To this end, we investigate theuse of mutual connectivity analysis with Local Models (MCA-LM), which estimates nonlinear measures of interactionbetween pairs of time-series in terms of cross-predictability. These pairwise measures provide a high-dimensionalrepresentation of connectivity profiles for subjects and are used as features for classification. Subsequently, we performfeature selection, reducing the dimension of the input space with the Kendall's τ coefficient method. The RandomForests (RF) and AdaBoost classifiers are used. Performing machine learning on functional connectivity measures iscommonly known as multi-voxel pattern analysis (MVPA). Traditionally, measures of functional connectivity areobtained with cross-correlation. Hence, as a metric to evaluate MCA-LM against, we also investigate classificationperformance with cross-correlation. The high area under receiver operating curve (AUC) and accuracy values for 100different train/test separations across both classifiers using MCA-LM (mean AUC ranges between 0.78 - 0.85 and meanaccuracy between 0.7 - 0.81) compared with standard MVPA analysis using cross-correlation between fMRI time-series(mean AUC ranges between 0.54 - 0.6 and mean accuracy between 0.50 - 0.57), across all the number of featuresselected demonstrates that such a nonlinear measure may be better suited at extracting information from the time-seriesdata and has potential for the development of novel neuro-imaging biomarkers for ASD.
展开▼