Neurological disorder classification using log-regularized laplacian sparse functional connectivity and ensemble graph-based deep learning

A computer program product for classifying a neurological disorder, the computer program product comprising one or more computer readable storage mediums having program code embodied therewith, the program code comprising programming instructions for:

• receiving resting-state functional magnetic resonance imaging (rs-fMRI) data and phenotypic data associated with a subject;
• estimating a sparse functional connectivity (FC) matrix from said rs-fMRI data by minimizing an objective function comprising a Laplacian graph trace term and a log-regularized weight term, wherein said log-regularized weight term comprises a logarithmic barrier to support positive node degrees;
• generating discriminative brain connectivity features from said sparse FC matrix using an Adaptive-Hierarchical Graph Convolutional Network (AH-GCN);
• processing said phenotypic data through a plurality of machine learning classifiers to generate phenotypic probability scores;
• and
• synthesizing said discriminative brain connectivity features and said phenotypic probability scores using an ensemble stacking model to generate a classification for said neurological disorder.

A system, comprising:

• a memory for storing a computer program for classifying a neurological disorder;
• and
• a processor connected to the memory, wherein the processor is configured to execute program instructions of the computer program comprising:
• receiving resting-state functional magnetic resonance imaging (rs-fMRI) data and phenotypic data associated with a subject;
• estimating a sparse functional connectivity (FC) matrix from said rs-fMRI data by minimizing an objective function comprising a Laplacian graph trace term and a log-regularized weight term, wherein said log-regularized weight term comprises a logarithmic barrier to support positive node degrees;
• generating discriminative brain connectivity features from said sparse FC matrix using an Adaptive-Hierarchical Graph Convolutional Network (AH-GCN);
• processing said phenotypic data through a plurality of machine learning classifiers to generate phenotypic probability scores;
• and
• synthesizing said discriminative brain connectivity features and said phenotypic probability scores using an ensemble stacking model to generate a classification for said neurological disorder.