Attention deficit hyperactivity disorder (ADHD) is a common neurodevelopmental problem in children. Resting state functional magnetic resonance imaging (rs-fMRI) provides an important tool in understanding the aberrant functional mechanisms in ADHD patients and assist in clinical diagnosis. Recently, spatio-temporal decomposition via spatial filtering (Fukunaga-Koontz transform, ICA) have gained attention in the analysis of fMRI time-series data. Their ability to decompose the blood oxygen level dependent (BOLD) rs-fMRI time series data into discriminative spatial and temporal components have resulted in better classification accuracy and the ability to isolate the important brain circuits responsible for the observed differences in brain activity. However, they are prone to errors in the estimation of covariance matrices due to the significant presence of atypical samples in the ADHD dataset. In this paper, we present a regularization framework to obtain a robust estimation of the covariance matrices such that the effect of atypical samples is reduced. The resulting approach called as regularized spatial filtering method (R-SFM) further uses Mahalanobis whitening to lower the effect of two-way correlations while preserving the spatial arrangement of the data in the feature extraction process. R-SFM was evaluated on the benchmark ADHD200 dataset and not only obtained a 6% improvement in classification accuracy, but also a 66.66% decrease in standard deviation over the previously developed SFM approach. Also R-SFM produces higher specificity which results in lower misclassification of ADHD, thereby reducing the risk of misdiagnosis. These results clearly show that R-SFM provides an accurate and reliable tool for detection of ADHD from BOLD rs-fMRI time series data.

Analysis of resting state - functional Magnetic Resonance Imaging (rs-fMRI) data has been a challenging problem due to a high homogeneity, large intra-class variability, limited samples and difference in acquisition technologies/techniques. These issues are predominant in the case of Attention Deficit Hyperactivity Disorder (ADHD). In this paper, we propose a new Deep Transformation Method (DTM) that extracts the discriminant latent feature space from rsfMRI and projects it in the subsequent layer for classification of rs-fMRI data. The hidden transformation layer in DTM projects the original rs-fMRI data into a new space using the learning policy and extracts the spatio-temporal correlations of the functional activities as a latent feature space. The subsequent convolution and decision layers transform the latent feature space into high-level features and provide accurate classification. The performance of DTM has been evaluated using the ADHD200 rs-fMRI benchmark data with crossvalidation. The results show that the proposed DTM achieves a mean classification accuracy of 70.36% and an improvement of 8.25% on the state of the art methodologies was observed. The improvement is due to concurrent analysis of the spatio-temporal correlations between the different regions of the brain and can be easily extended to study other cognitive disorders using rs-fMRI. Further, brain network analysis has been studied to identify the difference in functional activities and the corresponding regions behind cognitive symptoms in ADHD.

No abstract

No abstract

No abstract