JMIR Publications

ABSTRACT

Background:

Physiotherapy is a critical element in successful conservative management of low back pain (LBP). A gold standard in quantitatively measuring physiotherapy participation is crucial to understanding physiotherapy adherence managing recovery from LBP.

Objective:

The aim of this study was to develop and evaluate a system with wearable inertial sensors to objectively detect sitting postures and performance of unsupervised exercises for LBP containing movement in multiple planes.

Methods:

A quantitative classification design was used within a machine learning framework to detect exercise performance and posture in a cohort of healthy subjects. A set of 8 inertial sensors were placed on subjects and data was acquired as they performed 7 McKenzie low-back exercises and 3 sitting posture positions. Engineered time series features were extracted from the data and used to train nine models using a 6-fold cross validation approach, from which the best two models were selected for further study. In addition, a convolutional neural network (CNN) was trained directly on the time series data. A feature importance analysis was performed to identify sensor locations and channels that contributed most to the models. Finally, a subset of sensor locations and channels was included in a hyperparameter grid search to identify the optimal sensor configuration and the best performing algorithm(s) for exercise and posture classification. Final models were evaluated using F1-score in a 10-fold cross validation approach.

Results:

Nineteen healthy adult subjects with no history of low back pain each completed at least one full session of exercises and postures. Random forest (RF) and XGBoost (XGB) models performed best out of the initial nine engineered feature model set. The optimal hardware configuration was identified as a 3-sensor setup using lower back, left thigh, and right ankle sensors with acceleration, gyroscope, and magnetometer channels. The XGB model achieved the highest exercise (F1=0.94±0.03) and posture (F1=0.90±0.11) classification scores. The CNN achieved similar results with the same sensor locations, using only the accelerometer and gyroscope channels for exercise classification (F1=0.94±0.02) and the accelerometer channel alone for posture classification (F1=0.88±0.07).

Conclusions:

This study demonstrates the potential of a 3-sensor lower body wearable solution (e.g. smart pants) that can identify exercises in multiple planes and proper sitting postures, suitable for the treatment of LBP. This technology has the potential to improve the effectiveness of LBP rehabilitation by facilitating quantitative feedback, early problem diagnosis, and possible remote monitoring.