Comparison of algorithms and classifiers for stride detection using wearables

Abstract

Sensor-based systems for diagnosis or therapy support of motor dysfunctions need methodologies of automatically stride detection from movement sequences. In this proposal, we developed a stride detection system for daily life use. We compared mostly used algorithms min-max patterns, dynamic time warping, convolutional neural networks (CNN), and automatic framing using two data sets of 32 healthy and 28 Parkinson's disease (PD) persons. We developed an insole with force and IMU sensors to record the gait data. The PD patients carried out the standardized time up and go test, and the healthy persons a daily life activities test (walking, sitting, standing, ascending and descending stairs). As an automatically stride detection process for daily life use, we propose a first stride detection using automatic framing, and after normalization and resampling data a CNN is used. A F1-score of 0.938 (recall 0.968, precision 0.910) for time up and go test and of 0.944 (recall 0.992, precision 0.901) for daily life activities test were obtained for CNN. Compared to the other detection methods, up to 6% F-measure improvement was shown. Keywords Stride detection Á Gait analysis Á Inertial sensors Á Parkinson's disease Á Validation Á Dynamic time warping Á Time up and go test Á Convolutional neural networks & Tobias Steinmetzer

Analyzing gait symmetry with automatically synchronized wearable sensors in daily life

“…The use of CNN's for stride detection has proven to be very useful for us. Other work has already been able to benefit from the technology [35] . The symmetry of the legs is analyzed with DTW.…”

“…For the results we have specified precision, recall, F1-Score and Accuracy. For each column we have given the average value and standard deviation [35] .…”

Analysis and Classification of Motor Dysfunctions in Arm Swing in Parkinson’s Disease

“…Precision was the proportion of correctly predicted MD to all MD; see Equation (9). Accuracy was the ratio of all correctly recognized MD and no MD to all test data; see Equation (10). The F1-score (F1) was the harmonious average between precision and recall; see Equation (11).…”

“…Most of the research on gait analysis deals with the analysis of leg motion [4][5][6][7][8][9][10]. However, the analysis of the arm movement is also important for the assessment of a gait disorder.…”

Special issue on developing nature-inspired intelligence by neural systems

“…The accuracy was 97.78% with an equal error rate of 2.22% for abnormal and normal heart sound classification. Finally, there is a work, which presents a stride detection for patient with Parkinson's disease and control persons [3]. It is implemented by convolutional neural networks, dynamic time warping and algorithms min-max patterns.…”