Use of machine learning classifiers and sensor data to detect nurological deficit in stroke patients

Eunjeong Park, Hyuk Jae Chang, Hyo Suk Nam

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Background: The pronator drift test (PDT), a neurological examination, is widely used in clinics to measure motor weakness of stroke patients. Objective: The aim of this study was to develop a PDT tool with machine learning classifiers to detect stroke symptoms based on quantification of proximal arm weakness using inertial sensors and signal processing. Methods: We extracted features of drift and pronation from accelerometer signals of wearable devices on the inner wrists of 16 stroke patients and 10 healthy controls. Signal processing and feature selection approach were applied to discriminate PDT features used to classify stroke patients. A series of machine learning techniques, namely support vector machine (SVM), radial basis function network (RBFN), and random forest (RF), were implemented to discriminate stroke patients from controls with leave-one-out cross-validation. Results: Signal processing by the PDT tool extracted a total of 12 PDT features from sensors. Feature selection abstracted the major attributes from the 12 PDT features to elucidate the dominant characteristics of proximal weakness of stroke patients using machine learning classification. Our proposed PDT classifiers had an area under the receiver operating characteristic curve (AUC) of .806 (SVM), .769 (RBFN), and .900 (RF) without feature selection, and feature selection improves the AUCs to .913 (SVM), .956 (RBFN), and .975 (RF), representing an average performance enhancement of 15.3%. Conclusions: Sensors and machine learning methods can reliably detect stroke signs and quantify proximal arm weakness. Our proposed solution will facilitate pervasive monitoring of stroke patients.

Original languageEnglish
Article numbere120
JournalJournal of medical Internet research
Volume19
Issue number4
DOIs
Publication statusPublished - 2017 Apr

Fingerprint

Stroke
Area Under Curve
Arm
Pronation
Machine Learning
Neurologic Examination
Physiologic Monitoring
Wrist
ROC Curve
Equipment and Supplies
Support Vector Machine

All Science Journal Classification (ASJC) codes

  • Health Informatics

Cite this

@article{29253cb44fa14e9aa5fb4faa746354db,
title = "Use of machine learning classifiers and sensor data to detect nurological deficit in stroke patients",
abstract = "Background: The pronator drift test (PDT), a neurological examination, is widely used in clinics to measure motor weakness of stroke patients. Objective: The aim of this study was to develop a PDT tool with machine learning classifiers to detect stroke symptoms based on quantification of proximal arm weakness using inertial sensors and signal processing. Methods: We extracted features of drift and pronation from accelerometer signals of wearable devices on the inner wrists of 16 stroke patients and 10 healthy controls. Signal processing and feature selection approach were applied to discriminate PDT features used to classify stroke patients. A series of machine learning techniques, namely support vector machine (SVM), radial basis function network (RBFN), and random forest (RF), were implemented to discriminate stroke patients from controls with leave-one-out cross-validation. Results: Signal processing by the PDT tool extracted a total of 12 PDT features from sensors. Feature selection abstracted the major attributes from the 12 PDT features to elucidate the dominant characteristics of proximal weakness of stroke patients using machine learning classification. Our proposed PDT classifiers had an area under the receiver operating characteristic curve (AUC) of .806 (SVM), .769 (RBFN), and .900 (RF) without feature selection, and feature selection improves the AUCs to .913 (SVM), .956 (RBFN), and .975 (RF), representing an average performance enhancement of 15.3{\%}. Conclusions: Sensors and machine learning methods can reliably detect stroke signs and quantify proximal arm weakness. Our proposed solution will facilitate pervasive monitoring of stroke patients.",
author = "Eunjeong Park and Chang, {Hyuk Jae} and Nam, {Hyo Suk}",
year = "2017",
month = "4",
doi = "10.2196/jmir.7092",
language = "English",
volume = "19",
journal = "Journal of Medical Internet Research",
issn = "1439-4456",
publisher = "Journal of medical Internet Research",
number = "4",

}

Use of machine learning classifiers and sensor data to detect nurological deficit in stroke patients. / Park, Eunjeong; Chang, Hyuk Jae; Nam, Hyo Suk.

In: Journal of medical Internet research, Vol. 19, No. 4, e120, 04.2017.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Use of machine learning classifiers and sensor data to detect nurological deficit in stroke patients

AU - Park, Eunjeong

AU - Chang, Hyuk Jae

AU - Nam, Hyo Suk

PY - 2017/4

Y1 - 2017/4

N2 - Background: The pronator drift test (PDT), a neurological examination, is widely used in clinics to measure motor weakness of stroke patients. Objective: The aim of this study was to develop a PDT tool with machine learning classifiers to detect stroke symptoms based on quantification of proximal arm weakness using inertial sensors and signal processing. Methods: We extracted features of drift and pronation from accelerometer signals of wearable devices on the inner wrists of 16 stroke patients and 10 healthy controls. Signal processing and feature selection approach were applied to discriminate PDT features used to classify stroke patients. A series of machine learning techniques, namely support vector machine (SVM), radial basis function network (RBFN), and random forest (RF), were implemented to discriminate stroke patients from controls with leave-one-out cross-validation. Results: Signal processing by the PDT tool extracted a total of 12 PDT features from sensors. Feature selection abstracted the major attributes from the 12 PDT features to elucidate the dominant characteristics of proximal weakness of stroke patients using machine learning classification. Our proposed PDT classifiers had an area under the receiver operating characteristic curve (AUC) of .806 (SVM), .769 (RBFN), and .900 (RF) without feature selection, and feature selection improves the AUCs to .913 (SVM), .956 (RBFN), and .975 (RF), representing an average performance enhancement of 15.3%. Conclusions: Sensors and machine learning methods can reliably detect stroke signs and quantify proximal arm weakness. Our proposed solution will facilitate pervasive monitoring of stroke patients.

AB - Background: The pronator drift test (PDT), a neurological examination, is widely used in clinics to measure motor weakness of stroke patients. Objective: The aim of this study was to develop a PDT tool with machine learning classifiers to detect stroke symptoms based on quantification of proximal arm weakness using inertial sensors and signal processing. Methods: We extracted features of drift and pronation from accelerometer signals of wearable devices on the inner wrists of 16 stroke patients and 10 healthy controls. Signal processing and feature selection approach were applied to discriminate PDT features used to classify stroke patients. A series of machine learning techniques, namely support vector machine (SVM), radial basis function network (RBFN), and random forest (RF), were implemented to discriminate stroke patients from controls with leave-one-out cross-validation. Results: Signal processing by the PDT tool extracted a total of 12 PDT features from sensors. Feature selection abstracted the major attributes from the 12 PDT features to elucidate the dominant characteristics of proximal weakness of stroke patients using machine learning classification. Our proposed PDT classifiers had an area under the receiver operating characteristic curve (AUC) of .806 (SVM), .769 (RBFN), and .900 (RF) without feature selection, and feature selection improves the AUCs to .913 (SVM), .956 (RBFN), and .975 (RF), representing an average performance enhancement of 15.3%. Conclusions: Sensors and machine learning methods can reliably detect stroke signs and quantify proximal arm weakness. Our proposed solution will facilitate pervasive monitoring of stroke patients.

UR - http://www.scopus.com/inward/record.url?scp=85018752347&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85018752347&partnerID=8YFLogxK

U2 - 10.2196/jmir.7092

DO - 10.2196/jmir.7092

M3 - Article

C2 - 28420599

AN - SCOPUS:85018752347

VL - 19

JO - Journal of Medical Internet Research

JF - Journal of Medical Internet Research

SN - 1439-4456

IS - 4

M1 - e120

ER -