Activity classification based on inertial and barometric pressure sensors at different anatomical locations

Abstract

Miniature, wearable sensor modules are a promising technology to monitor activities of daily living (ADL) over extended periods of time. To assure both user compliance and meaningful results, the selection and placement site of sensors requires careful consideration. We investigated these aspects for the classification of 16 ADL in 6 healthy subjects under laboratory conditions using ReSense, our custom-made inertial measurement unit enhanced with a barometric pressure sensor used to capture activity-related altitude changes. Subjects wore a module on each wrist and ankle, and one on the trunk. Activities comprised whole body movements as well as gross and dextrous upper-limb activities. Wrist-module data outperformed the other locations for the three activity groups. Specifically, overall classification accuracy rates of almost 93% and more than 95% were achieved for the repeated holdout and user-specific validation methods, respectively, for all 16 activities. Including the altitude profile resulted in a considerable improvement of up to 20% in the classification accuracy for stair ascent and descent. The gyroscopes provided no useful information for activity classification under this scheme. The proposed sensor setting could allow for robust long-term activity monitoring with high compliance in different patient populations.

Monitoring Upper Limb Recovery after Cervical Spinal Cord Injury: Insights beyond Assessment Scores

“…Additionally, ambulant SCI subjects use many different assistive devices (e.g., crutches and rollers). For these reasons, algorithms developed for walking detection in other neurological diseases (39–41) have not yet been validated in SCI.…”

Parameterization and reliability of single-leg balance test assessed with inertial sensors in stroke survivors: a cross-sectional study

“…They are validated tools [27] and easily accessible because they are commonly found in quotidian devices such as smartphones for example [12, 14, 15]. These features increase the potential for their use in clinical research, such as the study presented here [12].…”

Sitting is the new smoking: online complex human activity recognition with smartphones and wearables

“…Its independence of orientation and position makes it very useful. The usefulness of fusing a barometer with the accelerometer for better activity recognition has been shown in a few studies [85,86]. For example, in [86], the authors show that the use of a barometer increases the recognition performance for walking upstairs and downstairs by up to 20%.…”

“…The usefulness of fusing a barometer with the accelerometer for better activity recognition has been shown in a few studies [85,86]. For example, in [86], the authors show that the use of a barometer increases the recognition performance for walking upstairs and downstairs by up to 20%. Higher accuracies have also been reported for climbing up and down activities in [85] using a combination of the barometer and accelerometer.…”

“…We also place using stairs in this category because it is not easy to recognize such activities with a single accelerometer. Usually, additional sensors, such as a gyroscope [20] and a barometer [86], are combined with the accelerometer to reliably recognize using stairs: walking upstairs and downstairs.…”