A Method to Analyze Correlations between Multiple Brain Imaging Tasks to Characterize Schizophrenia

Functional magnetic resonance imaging (fMRI) is widely used to develop techniques to better understand brain disorders like schizophrenia. Due to the complexity of the disorder it is useful to fuse data collected while a subject performs different tasks. It has been hypothesized that schizophrenia is associated with disruptions in connectivity within brain regions. It has also been shown that schizophrenics activate more similarly to two different tasks than do controls. This paper introduces a novel technique to fuse data from multiple fMRI tasks taking into account all brain voxels. This approach evaluates the distributive nature of the disorder while fusing multitask data to identify if patients and controls activate differently. In a study involving three tasks it was found that patients with schizophrenia do activate more similarly to most of the combinations of tasks than do controls. Results also affirm that schizophrenia manifests with widespread functional changes over the whole brain.