A multi-layer Gaussian process for motor symptom estimation in people with Parkinson's Disease

Muriel Lang, Franz M. J. Pfister, Jakob Frohner, Kian Abedinpour, Daniel Pichler, Urban Fietzek, Terry Taewoong Um, Dana Kulic, Satoshi Endo, Sandra Hirche

Research output: Contribution to journalArticleResearchpeer-review

Abstract

The assessment of Parkinson's disease (PD) poses a significant challenge, as it is influenced by various factors that lead to a complex and fluctuating symptom manifestation. Thus, a frequent and objective PD assessment is highly valuable for effective health management of people with Parkinson's disease (PwP). Here, we propose a method for monitoring PwP by stochastically modeling the relationships between wrist movements during unscripted daily activities and corresponding annotations about clinical displays of movement abnormalities. We approach the estimation of PD motor signs by independently modeling and hierarchically stacking Gaussian process models for three classes of commonly observed movement abnormalities in PwP including tremor, (non-tremulous) bradykinesia, and (non-tremulous) dyskinesia. We use clinically adopted severity measures as annotations for training the models, thus allowing our multi-layer Gaussian process prediction models to estimate not only their presence but also their severities. The experimental validation of our approach demonstrates strong agreement of the model predictions with these PD annotations. Our results show that the proposed method produces promising results in objective monitoring of movement abnormalities of PD in the presence of arbitrary and unknown voluntary motions, and makes an important step toward continuous monitoring of PD in the home environment.

Original languageEnglish
Pages (from-to)3038-3049
Number of pages12
JournalIEEE Transactions on Biomedical Engineering
Volume66
Issue number11
DOIs
Publication statusPublished - 1 Nov 2019
Externally publishedYes

Keywords

  • Ambient intelligence
  • Gaussian processes
  • medical information system
  • wearable sensors

Cite this

Lang, M., Pfister, F. M. J., Frohner, J., Abedinpour, K., Pichler, D., Fietzek, U., ... Hirche, S. (2019). A multi-layer Gaussian process for motor symptom estimation in people with Parkinson's Disease. IEEE Transactions on Biomedical Engineering, 66(11), 3038-3049. https://doi.org/10.1109/TBME.2019.2900002
Lang, Muriel ; Pfister, Franz M. J. ; Frohner, Jakob ; Abedinpour, Kian ; Pichler, Daniel ; Fietzek, Urban ; Um, Terry Taewoong ; Kulic, Dana ; Endo, Satoshi ; Hirche, Sandra. / A multi-layer Gaussian process for motor symptom estimation in people with Parkinson's Disease. In: IEEE Transactions on Biomedical Engineering. 2019 ; Vol. 66, No. 11. pp. 3038-3049.
@article{ef007329ad9740c2ac696bad37da2032,
title = "A multi-layer Gaussian process for motor symptom estimation in people with Parkinson's Disease",
abstract = "The assessment of Parkinson's disease (PD) poses a significant challenge, as it is influenced by various factors that lead to a complex and fluctuating symptom manifestation. Thus, a frequent and objective PD assessment is highly valuable for effective health management of people with Parkinson's disease (PwP). Here, we propose a method for monitoring PwP by stochastically modeling the relationships between wrist movements during unscripted daily activities and corresponding annotations about clinical displays of movement abnormalities. We approach the estimation of PD motor signs by independently modeling and hierarchically stacking Gaussian process models for three classes of commonly observed movement abnormalities in PwP including tremor, (non-tremulous) bradykinesia, and (non-tremulous) dyskinesia. We use clinically adopted severity measures as annotations for training the models, thus allowing our multi-layer Gaussian process prediction models to estimate not only their presence but also their severities. The experimental validation of our approach demonstrates strong agreement of the model predictions with these PD annotations. Our results show that the proposed method produces promising results in objective monitoring of movement abnormalities of PD in the presence of arbitrary and unknown voluntary motions, and makes an important step toward continuous monitoring of PD in the home environment.",
keywords = "Ambient intelligence, Gaussian processes, medical information system, wearable sensors",
author = "Muriel Lang and Pfister, {Franz M. J.} and Jakob Frohner and Kian Abedinpour and Daniel Pichler and Urban Fietzek and Um, {Terry Taewoong} and Dana Kulic and Satoshi Endo and Sandra Hirche",
year = "2019",
month = "11",
day = "1",
doi = "10.1109/TBME.2019.2900002",
language = "English",
volume = "66",
pages = "3038--3049",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
number = "11",

}

Lang, M, Pfister, FMJ, Frohner, J, Abedinpour, K, Pichler, D, Fietzek, U, Um, TT, Kulic, D, Endo, S & Hirche, S 2019, 'A multi-layer Gaussian process for motor symptom estimation in people with Parkinson's Disease', IEEE Transactions on Biomedical Engineering, vol. 66, no. 11, pp. 3038-3049. https://doi.org/10.1109/TBME.2019.2900002

A multi-layer Gaussian process for motor symptom estimation in people with Parkinson's Disease. / Lang, Muriel; Pfister, Franz M. J.; Frohner, Jakob; Abedinpour, Kian; Pichler, Daniel; Fietzek, Urban; Um, Terry Taewoong; Kulic, Dana; Endo, Satoshi; Hirche, Sandra.

In: IEEE Transactions on Biomedical Engineering, Vol. 66, No. 11, 01.11.2019, p. 3038-3049.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - A multi-layer Gaussian process for motor symptom estimation in people with Parkinson's Disease

AU - Lang, Muriel

AU - Pfister, Franz M. J.

AU - Frohner, Jakob

AU - Abedinpour, Kian

AU - Pichler, Daniel

AU - Fietzek, Urban

AU - Um, Terry Taewoong

AU - Kulic, Dana

AU - Endo, Satoshi

AU - Hirche, Sandra

PY - 2019/11/1

Y1 - 2019/11/1

N2 - The assessment of Parkinson's disease (PD) poses a significant challenge, as it is influenced by various factors that lead to a complex and fluctuating symptom manifestation. Thus, a frequent and objective PD assessment is highly valuable for effective health management of people with Parkinson's disease (PwP). Here, we propose a method for monitoring PwP by stochastically modeling the relationships between wrist movements during unscripted daily activities and corresponding annotations about clinical displays of movement abnormalities. We approach the estimation of PD motor signs by independently modeling and hierarchically stacking Gaussian process models for three classes of commonly observed movement abnormalities in PwP including tremor, (non-tremulous) bradykinesia, and (non-tremulous) dyskinesia. We use clinically adopted severity measures as annotations for training the models, thus allowing our multi-layer Gaussian process prediction models to estimate not only their presence but also their severities. The experimental validation of our approach demonstrates strong agreement of the model predictions with these PD annotations. Our results show that the proposed method produces promising results in objective monitoring of movement abnormalities of PD in the presence of arbitrary and unknown voluntary motions, and makes an important step toward continuous monitoring of PD in the home environment.

AB - The assessment of Parkinson's disease (PD) poses a significant challenge, as it is influenced by various factors that lead to a complex and fluctuating symptom manifestation. Thus, a frequent and objective PD assessment is highly valuable for effective health management of people with Parkinson's disease (PwP). Here, we propose a method for monitoring PwP by stochastically modeling the relationships between wrist movements during unscripted daily activities and corresponding annotations about clinical displays of movement abnormalities. We approach the estimation of PD motor signs by independently modeling and hierarchically stacking Gaussian process models for three classes of commonly observed movement abnormalities in PwP including tremor, (non-tremulous) bradykinesia, and (non-tremulous) dyskinesia. We use clinically adopted severity measures as annotations for training the models, thus allowing our multi-layer Gaussian process prediction models to estimate not only their presence but also their severities. The experimental validation of our approach demonstrates strong agreement of the model predictions with these PD annotations. Our results show that the proposed method produces promising results in objective monitoring of movement abnormalities of PD in the presence of arbitrary and unknown voluntary motions, and makes an important step toward continuous monitoring of PD in the home environment.

KW - Ambient intelligence

KW - Gaussian processes

KW - medical information system

KW - wearable sensors

UR - http://www.scopus.com/inward/record.url?scp=85073655476&partnerID=8YFLogxK

U2 - 10.1109/TBME.2019.2900002

DO - 10.1109/TBME.2019.2900002

M3 - Article

C2 - 30794163

AN - SCOPUS:85073655476

VL - 66

SP - 3038

EP - 3049

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 11

ER -