• Rouzbeh Maani, Yee-Hong Yang, Derek Emery, and Sanjay Kalra.
• Department of Computing Science, University of Alberta Edmonton, AB, Canada.
• Front Neurosci. 2016 Jan 1; 10: 120.

IntroductionRoutine MR images do not consistently reveal pathological changes in the brain in ALS. Texture analysis, a method to quantitate voxel intensities and their patterns and interrelationships, can detect changes in images not apparent to the naked eye. Our objective was to evaluate cerebral degeneration in ALS using 3-dimensional texture analysis of MR images of the brain.MethodsIn a case-control design, voxel-based texture analysis was performed on T1-weighted MR images of 20 healthy subjects and 19 patients with ALS. Four texture features, namely, autocorrelation, sum of squares variance, sum average, and sum variance were computed. Texture features were compared between the groups by statistical parametric mapping and correlated with clinical measures of disability and upper motor neuron dysfunction.ResultsTexture features were different in ALS in motor regions including the precentral gyrus and corticospinal tracts. To a lesser extent, changes were also found in the thalamus, cingulate gyrus, and temporal lobe. Texture features in the precentral gyrus correlated with disease duration, and in the corticospinal tract they correlated with finger tapping speed.ConclusionsChanges in MR image textures are present in motor and non-motor regions in ALS and correlate with clinical features. Whole brain texture analysis has potential in providing biomarkers of cerebral degeneration in ALS.

4 keywords...

hide…