Published on Wed Sep 22 2021

Identification of REM Sleep Behavior Disorder by Structural Magnetic Resonance Imaging and Machine Learning

Mei, J., Rahayel, S., Desrosiers, C., Postuma, R. B., Montplaisir, J., Carrier, J., Monchi, O., Frasnelli, J., Gagnon, J.-F.

Iiopathic rapid eye movement sleep behavior disorder (iRBD) is a major risk factor for synucleinopathies. The brain regions that are more vulnerable to neurodegeneration remain to be determined. Morphology-based machine learning approaches may allow for automated detection and subtyping of iRBD.

1
0
2
Abstract

Background Idiopathic rapid eye movement sleep behavior disorder (iRBD) is a major risk factor for synucleinopathies, and patients often present with clinical signs and morphological brain changes. However, there is an important heterogeneity in the presentation and progression of these alterations, and the brain regions that are more vulnerable to neurodegeneration remain to be determined. Objectives To assess the feasibility of morphology-based machine learning approaches in the identification and subtyping of iRBD. Methods For the three classification tasks [iRBD (n=48) vs controls (n=41); iRBD vs Parkinson's disease (n=29); iRBD with mild cognitive impairment (n=16) vs without mild cognitive impairment (n=32)], machine learning models were trained with morphometric measurements (thickness, surface area, volume, and deformation) extracted from T1- weighted structural magnetic resonance imaging. Model performance and the most discriminative brain regions were analyzed and identified. Results A high accuracy was reported for iRBD vs controls (79.6%, deformation of the caudal middle frontal gyrus and putamen, thinning of the superior frontal gyrus, and reduced volume of the inferior parietal cortex and insula), iRBD vs Parkinson's disease (82%, larger volume and surface area of the insula, thinning of the entorhinal cortex and lingual gyrus, and reduced volume of the fusiform gyrus), and iRBD with vs without mild cognitive impairment (84.8%, thinning of the pars triangularis, superior temporal gyrus, transverse temporal cortex, larger surface area of the superior temporal gyrus, and deformation of isthmus of the cingulate gyrus). Conclusions Morphology-based machine learning approaches may allow for automated detection and subtyping of iRBD, potentially enabling efficient preclinical identification of synucleinopathies.