Classification of patients with migraine and healthy subjects using gross brain volume changes

Abstract

AbstractBackground and PurposeDespite reports of changes in gray matter and white matter in individuals with migraines, research on detecting whole‐brain gross volume changes in migraines is very rare. Therefore, we aimed to investigate gross volume changes associated with migraine and test its potential in developing a neuroimaging biomarker for objective diagnosis of migraine.MethodologyIn this study, 45 patients with migraine and 46 healthy controls were scanned using a 3T MRI scanner, and 3D, T1‐weighted brain MR images were obtained. First, Tensor‐based morphometry and Voxel‐based morphometry were performed using Computational Anatomy Toolbox (CAT 12) to compute gross volumes, gray matter volumes, white matter volumes, and cerebrospinal fluid volumes. Univariate analyses were performed using two‐sample t‐test and the results were corrected for multiple comparisons. Second, multivariate pattern analyses (MVPA) were conducted to classify the patients with migraine and healthy controls.ResultsReduced gross volumes in patients were detected in middle frontal gyrus, superior frontal gyrus, inferior temporal gyrus and middle temporal gyrus of right cerebrum, middle temporal gyrus and angular gyrus of the left cerebrum, cuneus, and left cerebellum. Reduced gray matter volumes were detected in superior frontal, right middle frontal, right inferior temporal, and right middle temporal. The results of MVPA indicated that gross volume changes can improve the classification accuracy (73.63%) than other structural metrics such as gray matter (70%), white matter (66.3%), and cerebrospinal fluid volumes (58.69%).ConclusionWe conclude that the gross volumetric alterations can serve as a better neuroimaging marker for the objective diagnosis of migraine.