Automated Categorization of Parkinsonian Syndromes Using Magnetic Resonance Imaging in a Clinical Setting

Abstract

Background Machine learning algorithms using magnetic resonance imaging (MRI) data can accurately discriminate parkinsonian syndromes. Validation in patients recruited in routine clinical practice is missing. Objective The aim of this study was to assess the accuracy of a machine learning algorithm trained on a research cohort and tested on an independent clinical replication cohort for the categorization of parkinsonian syndromes. Methods Three hundred twenty‐two subjects, including 94 healthy control subjects, 119 patients with Parkinson's disease (PD), 51 patients with progressive supranuclear palsy (PSP) with Richardson's syndrome, 35 with multiple system atrophy (MSA) of the parkinsonian variant (MSA‐P), and 23 with MSA of the cerebellar variant (MSA‐C), were recruited. They were divided into a training cohort (n = 179) scanned in a research environment and a replication cohort (n = 143) examined in clinical practice on different MRI systems. Volumes and diffusion tensor imaging (DTI) metrics in 13 brain regions were used as input for a supervised machine learning algorithm. To harmonize data across scanners and reduce scanner‐dependent effects, we tested two types of normalizations using patient data or healthy control data. Results In the replication cohort, high accuracies were achieved using volumetry in the classification of PD–PSP, PD–MSA‐C, PSP–MSA‐C, and PD‐atypical parkinsonism (balanced accuracies: 0.840–0.983, area under the receiver operating characteristic curves: 0.907–0.995). Performances were lower for the classification of PD–MSA‐P, MSA‐C–MSA‐P (balanced accuracies: 0.765–0.784, area under the receiver operating characteristic curve: 0.839–0.871) and PD–PSP–MSA (balanced accuracies: 0.773). Performance using DTI was improved when normalizing by controls, but remained lower than that using volumetry alone or combined with DTI. Conclusions A machine learning approach based on volumetry enabled accurate classification of subjects with early‐stage parkinsonism, examined on different MRI systems, as part of their clinical assessment. © 2020 International Parkinson and Movement Disorder Society

Differentiation of Parkinson’s disease and Parkinsonism predominant multiple system atrophy in early disease stages by means of objective measurement in Susceptibility Weighted Imaging

“…Regional apparent diffusion coe cients of middle cerebellar peduncles completely differentiated MSA-P patients from Parkinson's disease patients with mean disease duration of 4.9 years [11]. Also by machine learning approach based on volumetry enabled accurate classi cation of subjects with early-stage parkinsonism of about 5.0 years' mean disease duration [12]. Proton magnetic resonance spectroscopy in the basal ganglia of MSA-P with mean disease duration of 3.4 years was also different from healthy controls [13].…”

A bibliometric analysis of neuroimaging biomarkers in Parkinson disease based on Web of Science

“…[ 45 ] T1-weighted MRI with voxel-based morphometry has some advantages in differentiating PD from atypical Parkinson syndrome. [ 46 ] However, there are limitations to using different NMBM. Bougea [ 18 ] suggested that a combination of multiple biomarkers can improve the early diagnosis and more accurate prognosis of PD.…”

Differentiation of Parkinson’s disease and Parkinsonism predominant multiple system atrophy in early stage by morphometrics in susceptibility weighted imaging

“…Regional apparent diffusion coefficients of middle cerebellar peduncles completely differentiated patients with MSA-P from those with PD, with a mean disease duration of 4.9 years ( Nicoletti et al, 2006 ). In addition, the machine learning approach based on volumetry enabled accurate classification of subjects with early stage Parkinsonism with a mean disease duration of 5.0 years ( Chougar et al, 2020 ). Proton magnetic resonance spectroscopy findings in the basal ganglia of patients with MSA-P with a mean disease duration of 3.4 years also differed from those of healthy controls ( Stamelou et al, 2015 ).…”