Publication

Detecting freezing of gait with a tri-axial accelerometer in parkinson’s disease patients

Ahlrichs, Claas
; 
Samà, Albert
; 
Lawo, Michael
; 
Cabestany, Joan
; 
Rodríguez-Martín, Daniel
; 
Pérez-López, Carlos
; 
Sweeney, Dean
; 
Quinlan, Leo R.
; 
Laighin, Gearòid Ò
; 
Counihan, Timothy
... show 10 more
Repository DOI
Publication Date
2015-10-01
Keywords
parkinson's disease, freezing of gait, machine learning, support vector machines, diagnosis, symptoms, models
Type
Article
Downloads
Show all metadata
Citation
Ahlrichs, Claas; Samà, Albert; Lawo, Michael; Cabestany, Joan; Rodríguez-Martín, Daniel; Pérez-López, Carlos; Sweeney, Dean; Quinlan, Leo R. Laighin, Gearòid Ò; Counihan, Timothy; Browne, Patrick; Hadas, Lewy; Vainstein, Gabriel; Costa, Alberto; Annicchiarico, Roberta; Alcaine, Sheila; Mestre, Berta; Quispe, Paola; Bayes, Àngels; Rodríguez-Molinero, Alejandro (2015). Detecting freezing of gait with a tri-axial accelerometer in parkinson’s disease patients. Medical & Biological Engineering & Computing 54 (1), 223-233
Abstract
Freezing of gait (FOG) is a common motor symptom of Parkinson's disease (PD), which presents itself as an inability to initiate or continue gait. This paper presents a method to monitor FOG episodes based only on acceleration measurements obtained from a waist-worn device. Three approximations of this method are tested. Initially, FOG is directly detected by a support vector machine (SVM). Then, classifier's outputs are aggregated over time to determine a confidence value, which is used for the final classification of freezing (i.e., second and third approach). All variations are trained with signals of 15 patients and evaluated with signals from another 5 patients. Using a linear SVM kernel, the third approach provides 98.7 % accuracy and a geometric mean of 96.1 %. Moreover, it is investigated whether frequency features are enough to reliably detect FOG. Results show that these features allow the method to detect FOG with accuracies above 90 % and that frequency features enable a reliable monitoring of FOG by using simply a waist sensor.
Funder
Publisher
Springer Nature
Publisher DOI
10.1007/s11517-015-1395-3
Rights
Attribution-NonCommercial-NoDerivs 3.0 Ireland
cbn
Collections
Externally hosted open access publications with University of Galway authors (2)