Background: Schizophrenia is a long-term mental disorder involving altered relations between thought, emotion, behavior and cognition. There is increasing demand for objective diagnoses of schizophrenia. The diagnosis of schizophrenia has been the subject of many research debates. MRI studies postulate that schizophrenia is a brain disorder with altered brain structure, whereas fMRI studies suggest that schizophrenia is a reduced BOLD signal in various brain regions. We hypothesize that the brain vessels and vascularity may serve as important factors promoting differences in volume and functionality between schizophrenia patients and controls. We tested this hypothesis by comparing precontrast and postcontrast DCE-T1- MRI.
Methods: 42 first episode schizophrenia patients and 28 matched controls participated in this study. MR imaging was performed using a DCE-MRI with GD-DTPA. All subjects underwent the same imaging protocol. FreeSurfer software was used to obtain brain volumes of the subcortical and cortex for all patients. The neuroanatomical volumes were inspected for accuracy in all subjects. No obvious errors were observed for any subject, and so all analyses were 100% automated. The vascular dynamics was obtained by a standard subtraction – early post- minus pre- contrast scan.
Results: Based on the volume analysis we show the ability to classify a novel subject (as either schizophrenia or control) with up to 73% sensitivity and 65% specificity in various brain regions including the basal ganglia hippocampus and amygdala. Taken together, volumetric analyses with the dynamic vascularity improve the classification accuracy to 93% sensitivity and 65% specificity.
Conclusion: Our method shows promising results in the direction of objective findings in schizophrenia. The combination of vascular and volumetric findings may suggest a possible common pathway in the schizophrenic process as altered vascularity may lead to volume abnormalities.