Benchtop validation of a low-cost wearable triaxial accelerometer

Area of Science:

• Biomedical Engineering
• Sensor Technology
• Gerontology

Background:

• Wearable inertial sensors are crucial for detecting fall-related events in older adults.
• Rigorous validation of low-cost devices is essential to ensure measurement fidelity and stability.

Purpose of the Study:

• To evaluate the concurrent validity of a new, low-cost triaxial accelerometer.
• To compare its linear acceleration measurements against a commercial reference sensor (RS).

Main Methods:

• A methodological validation study involving 49 sensor under test (SUT) units and a Physilog®5 RS.
• Simultaneous collection of triaxial linear acceleration data across three axes with repeated trials.
• Analysis using bias, root mean squared error (RMSE), mean absolute error (MAE), waveform comparison, and Bland-Altman analysis.

Main Results:

• The SUT showed minimal bias on X and Y axes, with slightly higher errors on the Z-axis.
• RMSE and MAE were similar for X and Y axes, with moderate increases on the Z-axis.
• Strong waveform overlap and near-unity correlations indicated high signal correspondence; Bland-Altman showed minimal bias and narrow limits of agreement for X and Y axes.

Conclusions:

• The low-cost accelerometer (~$60 USD) demonstrates strong agreement with a high-cost reference device (> $1,000 USD).
• The device is validated for three-dimensional linear acceleration measurement under controlled conditions.
• This accelerometer offers a promising, cost-effective solution for scalable motion monitoring and fall detection in older adults.