Machine learning analysis of motor evoked potential time series to predict disability progression in multiple sclerosis.

Publication date: Mar 21, 2020

Evoked potentials (EPs) are a measure of the conductivity of the central nervous system. They are used to monitor disease progression of multiple sclerosis patients. Previous studies only extracted a few variables from the EPs, which are often further condensed into a single variable: the EP score. We perform a machine learning analysis of motor EP that uses the whole time series, instead of a few variables, to predict disability progression after two years. Obtaining realistic performance estimates of this task has been difficult because of small data set sizes. We recently extracted a dataset of EPs from the Rehabiliation & MS Center in Overpelt, Belgium. Our data set is large enough to obtain, for the first time, a performance estimate on an independent test set containing different patients.

We extracted a large number of time series features from the motor EPs with the highly comparative time series analysis software package. Mutual information with the target and the Boruta method are used to find features which contain information not included in the features studied in the literature. We use random forests (RF) and logistic regression (LR) classifiers to predict disability progression after two years. Statistical significance of the performance increase when adding extra features is checked.

Including extra time series features in motor EPs leads to a statistically significant improvement compared to using only the known features, although the effect is limited in magnitude (?AUC = 0.02 for RF and ?AUC = 0.05 for LR). RF with extra time series features obtains the best performance (AUC = 0.75+/-0.07 (mean and standard deviation)), which is good considering the limited number of biomarkers in the model. RF (a nonlinear classifier) outperforms LR (a linear classifier).

Using machine learning methods on EPs shows promising predictive performance. Using additional EP time series features beyond those already in use leads to a modest increase in performance. Larger datasets, preferably multi-center, are needed for further research. Given a large enough dataset, these models may be used to support clinicians in their decision making process regarding future treatment.

Open Access PDF

Yperman, J., Becker, T., Valkenborg, D., , Popescu, Hellings, N., Wijmeersch, B.V., and Peeters, L.M. Machine learning analysis of motor evoked potential time series to predict disability progression in multiple sclerosis. 20513. 2020 BMC Neurol (20):1.

Concepts Keywords
AUC Electroencephalography
Belgium Evoked potential
Biomarkers Multiple sclerosis
BMC Clinical medicine
Central Nervous System Biology
Classifier
Conductivity
Disability
Evoked Potential
Logistic Regression
Magnitude
Multiple Sclerosis
Nonlinear
Overpelt
Random Forests
Standard Deviation
Statistical Significance

Semantics

Type Source Name
disease MESH multiple sclerosis
disease MESH disease progression
pathway REACTOME Reproduction
disease MESH sclerosis
disease MESH demyelination
disease MESH anxiety
disease MESH visual
disease MESH community
drug DRUGBANK Ilex paraguariensis leaf
drug DRUGBANK Trestolone
drug DRUGBANK Cysteamine
drug DRUGBANK Saquinavir
drug DRUGBANK Esomeprazole
disease MESH diagnosis
disease MESH development
drug DRUGBANK Coenzyme M
disease MESH syndrome
disease MESH separated
drug DRUGBANK Pentaerythritol tetranitrate
drug DRUGBANK Ranitidine
drug DRUGBANK Dimethyltryptamine
disease MESH uncertainty
disease MESH privacy
drug DRUGBANK Gold
disease MESH Mult
drug DRUGBANK Ribostamycin
disease MESH optic neuritis
disease MESH relapse
disease MESH cognitive impairment
drug DRUGBANK Tilmicosin
disease MESH stroke
drug DRUGBANK Kale
drug DRUGBANK Alemtuzumab
disease MESH relapsing remitting multiple sclerosis

Similar

Original Article

Leave a Comment

Your email address will not be published. Required fields are marked *