Related Experiment Video
Updated: Aug 12, 2025
07:24
Home-Based Monitor for Gait and Activity Analysis
Published on: August 8, 2019
6.8K
Prediction of early-wheelchair dependence in multiple system atrophy based on machine learning algorithm: A
Lingyu Zhang1, Yanbing Hou1, Xiaojing Gu1
1Department of Neurology, Laboratory of Neurodegenerative Disorders, Rare Diseases Center, National Clinical Research Center for Geriatrics, West China Hospital, Sichuan University, Chengdu, China.
Clinical Parkinsonism & Related Disorders
|January 30, 2023
Summary
Predictors for wheelchair dependence in multiple system atrophy (MSA) include disease duration, age, and plasma neurofilament light chain (NFL) levels. These factors aid in early identification and intervention strategies for MSA patients.
Area of Science:
- Neurology
- Biomarkers
- Clinical Research
Background:
- Predictive factors for wheelchair dependence in multiple system atrophy (MSA) remain unclear.
- Early identification of wheelchair dependence is crucial for timely intervention in MSA patients.
Purpose of the Study:
- To explore predictive factors for early wheelchair dependence in MSA patients.
- To identify clinical features and blood biomarkers that predict wheelchair dependence.
Main Methods:
- Prospective cohort study including 100 probable MSA patients.
- Random forest (RF) model used to establish a predictive model for early wheelchair dependence.
- Model performance evaluated using accuracy, sensitivity, specificity, and AUC.
Main Results:
- Higher baseline plasma neurofilament light chain (NFL) levels were observed in patients with wheelchair dependence.
- Key predictive factors identified: disease duration, age of onset, UMSARS-II, NFL, UMSARS-I, CRP, and NLR.
- RF model achieved 72.29% accuracy, 70.82% sensitivity, 74.55% specificity, and 0.72 AUC.
Conclusions:
- Baseline NFL, CRP, and NLR are potential predictive biomarkers for early wheelchair dependence in MSA.
- Clinical features and blood biomarkers offer insights for early intervention trials in MSA.
Keywords:
AUC, area under the receiver operating characteristic curveCRP, C-reactive proteinCohort studyMSA, multiple system atrophyMSA-C, multiple system atrophy with predominate cerebellar ataxiaMSA-P, multiple system atrophy with predominate parkinsonismMultiple system atrophyNFL, neurofilament light chainNLR, neutrophil-to-lymphocyte ratioOH, orthostatic hypotensionRF, random forestSCA, spinocerebellar ataxiaTNF, tumor necrosis factorUMSARS, unified multiple system atrophy rating scaleWheelchair dependence
