Detection of freezing of gait in people withParkinson’s disease using smartphones

Freezing of Gait (FOG) is one of the most trouble-some motor symptoms associated with Parkinson’s disease (PD),characterised by brief episodes of inability to step. It involvesincreased risk of falls and reduced quality of life, and correlateswith motor fluctuations and progression of the disease. Hence, theknowledge of FOG event frequency, duration, daily distributionand response to drug therapy is fundamental for a reliablepatient’s assessment. In this study, we propose a FOG detectionalgorithm that takes as input inertial data from a single waist-mounted smartphone, and provides information about presenceand duration of FOG episodes. Data acquisition was carried on38 PD patients and 21 elderly subjects executing a standard6-minute walking test. More than 3.5 hours of accelerationdata have been collected. A combination of Support VectorMachine and k-Nearest Neighbour classifiers has been designed.Sensitivity of 95.4%, specificity of 98.8%, precision of 92.8%and accuracy of 98.3% in the 10-fold cross validation, and adetection rate of 84% in Leave-one-Subject-Out validation were obtained. These results, along with a good time resolution in theFOG duration identification and very efficient processing times,make the algorithm a promising tool for reliable FOG assessmentduring activities of daily living