TY - JOUR
T1 - Detecting freezing of gait with a tri-axial accelerometer in Parkinson’s disease patients
AU - Ahlrichs, Claas
AU - Samà, Albert
AU - Lawo, Michael
AU - Cabestany, Joan
AU - Rodríguez-Martín, Daniel
AU - Pérez-López, Carlos
AU - Sweeney, Dean
AU - Quinlan, Leo R.
AU - Laighin, Gearòid
AU - Counihan, Timothy
AU - Browne, Patrick
AU - Hadas, Lewy
AU - Vainstein, Gabriel
AU - Costa, Alberto
AU - Annicchiarico, Roberta
AU - Alcaine, Sheila
AU - Mestre, Berta
AU - Quispe, Paola
AU - Bayes, Àngels
AU - Rodríguez-Molinero, Alejandro
PY - 2015/10/1
Y1 - 2015/10/1
N2 - 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.
AB - 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.
KW - Freezing of Gait
KW - Machine learning
KW - Parkinson’s disease
KW - Support vector machines
UR - http://www.scopus.com/inward/record.url?scp=84944707200&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84944707200&partnerID=8YFLogxK
U2 - 10.1007/s11517-015-1395-3
DO - 10.1007/s11517-015-1395-3
M3 - Article
AN - SCOPUS:84944707200
JO - Medical and Biological Engineering and Computing
JF - Medical and Biological Engineering and Computing
SN - 0140-0118
ER -