A standardized validity assessment protocol for physiological signals from wearable technology: Methodological...

Area of Science:

• Biomedical Engineering
• Physiological Measurement
• Wearable Technology

Background:

• Increasing use of wearable devices in ambulatory research necessitates standardized validation methods.
• Current validation approaches for novel sensing technology lack consensus, leading to inconsistent results.
• Existing methods are often unsuitable for assessing physiological signals from wearables.

Purpose of the Study:

• Introduce a comprehensive validity assessment protocol for physiological signals from wearable devices.
• Evaluate the validity of the E4 wearable device using the proposed three-level protocol.
• Establish a standardized framework for assessing the quality of wearable sensor data.

Main Methods:

• Developed a three-level validity assessment protocol: signal, parameter, and event levels.
• Utilized cross-correlation for signal-level validation.
• Employed Bland-Altman plots for parameter-level validation and event difference plots for event-level validation.
• Tested the E4 wearable device measuring electrodermal and cardiovascular activity.

Main Results:

• The E4 wearable demonstrated validity for heart rate, RMSSD, and SD at parameter and event levels.
• Total amplitude of skin conductance responses was valid at the event level during sustained stressors.
• Findings align with existing literature, supporting the protocol's applicability.

Conclusions:

• The proposed protocol offers a comprehensive, standardized, and feasible method for validating wearable sensor data.
• The E4 wearable is a valid tool for ambulatory research, particularly for heart rate and skin conductance measurements under stress.
• This work facilitates rigorous quality assessment of physiological data from emerging wearable technologies.