Discriminating progressive supranuclear palsy from Parkinson's disease using wearable technology and machine learning.
Gait
Inertial sensor array
Machine learning
Parkinson’s disease
Progressive supranuclear pasly
Wearables
Journal
Gait & posture
ISSN: 1879-2219
Titre abrégé: Gait Posture
Pays: England
ID NLM: 9416830
Informations de publication
Date de publication:
03 2020
03 2020
Historique:
received:
27
09
2019
revised:
05
02
2020
accepted:
06
02
2020
pubmed:
23
2
2020
medline:
1
12
2020
entrez:
21
2
2020
Statut:
ppublish
Résumé
Progressive supranuclear palsy (PSP), a neurodegenerative conditions may be difficult to discriminate clinically from idiopathic Parkinson's disease (PD). It is critical that we are able to do this accurately and as early as possible in order that future disease modifying therapies for PSP may be deployed at a stage when they are likely to have maximal benefit. Analysis of gait and related tasks is one possible means of discrimination. Here we investigate a wearable sensor array coupled with machine learning approaches as a means of disease classification. 21 participants with PSP, 20 with PD, and 39 healthy control (HC) subjects performed a two minute walk, static sway test, and timed up-and-go task, while wearing an array of six inertial measurement units. The data were analysed to determine what features discriminated PSP from PD and PSP from HC. Two machine learning algorithms were applied, Logistic Regression (LR) and Random Forest (RF). 17 features were identified in the combined dataset that contained independent information. The RF classifier outperformed the LR classifier, and allowed discrimination of PSP from PD with 86 % sensitivity and 90 % specificity, and PSP from HC with 90 % sensitivity and 97 % specificity. Using data from the single lumbar sensor only resulted in only a modest reduction in classification accuracy, which could be restored using 3 sensors (lumbar, right arm and foot). However for maximum specificity the full six sensor array was needed. A wearable sensor array coupled with machine learning methods can accurately discriminate PSP from PD. Choice of array complexity depends on context; for diagnostic purposes a high specificity is needed suggesting the more complete array is advantageous, while for subsequent disease tracking a simpler system may suffice.
Sections du résumé
BACKGROUND
Progressive supranuclear palsy (PSP), a neurodegenerative conditions may be difficult to discriminate clinically from idiopathic Parkinson's disease (PD). It is critical that we are able to do this accurately and as early as possible in order that future disease modifying therapies for PSP may be deployed at a stage when they are likely to have maximal benefit. Analysis of gait and related tasks is one possible means of discrimination.
RESEARCH QUESTION
Here we investigate a wearable sensor array coupled with machine learning approaches as a means of disease classification.
METHODS
21 participants with PSP, 20 with PD, and 39 healthy control (HC) subjects performed a two minute walk, static sway test, and timed up-and-go task, while wearing an array of six inertial measurement units. The data were analysed to determine what features discriminated PSP from PD and PSP from HC. Two machine learning algorithms were applied, Logistic Regression (LR) and Random Forest (RF).
RESULTS
17 features were identified in the combined dataset that contained independent information. The RF classifier outperformed the LR classifier, and allowed discrimination of PSP from PD with 86 % sensitivity and 90 % specificity, and PSP from HC with 90 % sensitivity and 97 % specificity. Using data from the single lumbar sensor only resulted in only a modest reduction in classification accuracy, which could be restored using 3 sensors (lumbar, right arm and foot). However for maximum specificity the full six sensor array was needed.
SIGNIFICANCE
A wearable sensor array coupled with machine learning methods can accurately discriminate PSP from PD. Choice of array complexity depends on context; for diagnostic purposes a high specificity is needed suggesting the more complete array is advantageous, while for subsequent disease tracking a simpler system may suffice.
Identifiants
pubmed: 32078894
pii: S0966-6362(20)30068-0
doi: 10.1016/j.gaitpost.2020.02.007
pii:
doi:
Types de publication
Journal Article
Observational Study
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
257-263Subventions
Organisme : Department of Health
Pays : United Kingdom
Informations de copyright
Crown Copyright © 2020. Published by Elsevier B.V. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of Competing Interest MDV, JP, TB, and CAA have no conflicts of interest. JJF has performed consultancy unrelated to the subject of this study for companies including Abbott, Renishaw, and Herantis, and also has a research grant from Abbott, for a project unrelated to this study.