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

Claas Ahlrichs, Albert Samà, Michael Lawo, Joan Cabestany, Daniel Rodríguez-Martín, Carlos Pérez-López, Dean Sweeney, Leo R. Quinlan, Gearòid Laighin, Timothy Counihan, Patrick Browne, Lewy Hadas, Gabriel Vainstein, Alberto Costa, Roberta Annicchiarico, Sheila Alcaine, Berta Mestre, Paola Quispe, Àngels Bayes, Alejandro Rodríguez-Molinero

Research output: Contribution to journalArticlepeer-review


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.

Original languageEnglish
JournalMedical and Biological Engineering and Computing
Publication statusAccepted/In press - Oct 1 2015


  • Freezing of Gait
  • Machine learning
  • Parkinson’s disease
  • Support vector machines

ASJC Scopus subject areas

  • Biomedical Engineering
  • Computer Science Applications


Dive into the research topics of 'Detecting freezing of gait with a tri-axial accelerometer in Parkinson’s disease patients'. Together they form a unique fingerprint.

Cite this