DATA AUGMENTATION FOR FMRI-BASED FUNCTIONAL CONNECTIVITY AND ITS APPLICATION TO CROSS-SITE ADHD CLASSIFICATION

Functional magnetic resonance imaging (fMRI) is an emerging neuroimaging modality that is widely used to study brain function and disorders due to its advantages of non-invasiveness, no radiation damage, and high spatial resolution. Existing studies have focused on fMRI-based recognition models to help diagnose brain disorders. However, due to the high cost of fMRI data acquisition and labeling, the amount of fMRI data is usually small, which largely limits the performance of recognition models. In addition, cross-site classification is always a challenge in fMRI study, because the heterogeneity of data collection at different sites increases the complexity and diversity of the data distribution, making the cross-site classification less robust than site-specific classification. In this paper, we propose three data augmentation methods based on functional connectivity networks (FCNs) of fMRI data, aided by a deep feature fusion method, for automatic disease identification. Firstly, Gaussian noise method, Mixup method, and sliding window method are proposed to effectively augment FCN data, respectively, this can balance the variability of sample distribution. Secondly, convolution neural network and graph attention network are separately employed to extract local and global features from FCN. Finally, the two kinds of features are integrated to classify subjects. The experimental results on the ADHD-200 dataset indicate that: (1) the data augmentation methods can effectively improve identification performance, in particular, the sliding window method performs best; (2) the cross-site attention deficit and hyperactivity disorder (ADHD) classification is improved by combining the data augmentation method of sliding window and deep feature fusion method; (3) the rationality of data augmentation for FCNs is explained by visualizing the hidden fused features with t-stochastic neighborhood embedding algorithm.