Automatic Recognition of fMRI-Derived Functional Networks Using 3-D Convolutional Neural Networks

Abstract

Current functional magnetic resonance imaging (fMRI) data modeling techniques, such as independent component analysis and sparse coding methods, can effectively reconstruct dozens or hundreds of concurrent interacting functional brain networks simultaneously from the whole brain fMRI signals. However, such reconstructed networks have no correspondences across different subjects. Thus, automatic, effective, and accurate classification and recognition of these large numbers of fMRI-derived functional brain networks are very important for subsequent steps of functional brain analysis in cognitive and clinical neuroscience applications. However, this task is still a challenging and open problem due to the tremendous variability of various types of functional brain networks and the presence of various sources of noises. In recognition of the fact that convolutional neural networks (CNN) has superior capability of representing spatial patterns with huge variability and dealing with large noises, in this paper, we design, apply, and evaluate a deep 3-D CNN framework for automatic, effective, and accurate classification and recognition of large number of functional brain networks reconstructed by sparse representation of whole-brain fMRI signals. Our extensive experimental results based on the Human Connectome Project fMRI data showed that the proposed deep 3-D CNN can effectively and robustly perform functional networks classification and recognition tasks, while maintaining a high tolerance for mistakenly labeled training instances. This study provides a new deep learning approach for modeling functional connectomes based on fMRI data.

A Novel Deep Learning Framework on Brain Functional Networks for Early MCI Diagnosis

“…To our best knowledge, this is the first BFN-based deep learning for the disease diagnosis. Zhao et al, proposed a 3D CNN model for automatic ICA component labeling for each BFN [18]; however, that study used only healthy subjects for a much easier task (component labeling) than the current eMCI diagnosis study. In the proposed method, we first extracted subject-specific spatial maps of high-level cognitive function-related BFNs by using GIG-ICA, and then constructed a 3D CNN for each BFN, respectively, to learn deeply embedded spatial patterns of each BFN.…”

Deep Learning Methods to Process fMRI Data and Their Application in the Diagnosis of Cognitive Impairment: A Brief Overview and Our Opinion

“…The 3D-CNN can also be used to process fMRI data. The data from Human Connectome project (HCP) was utilized to reconstruct the brain network with a higher accuracy (Zhao et al, 2017b ). The methods of combining fMRI with structural magnetic resonance imaging had also been used for the classification of Attention deficit hyperactivity disorder (ADHD) (Zou et al, 2017 ).…”

A Multi-Domain Connectome Convolutional Neural Network for Identifying Schizophrenia From EEG Connectivity Patterns

“…Applications of CNNs to brain connectome data in classifying spatial maps of functional networks are introduced only recently and in its very early stage. A few related works include classification of fMRIderived functional connectivity in mild cognitive impairment [35] and in resting-state networks [36], as well as DTI-based structural connectivity for predicting neurodevelopment in infants [37]. However, the convolutional architectures proposed in these studies focused on one type of connectivity measures from a single domain such as the Pearson correlation for functional connectivity, which did not take into account the connection directionality and topological organization of the brain networks.…”