University of Twente Student Theses


Validation of a wrist-worn photoplethysmographic sensor (E4): comparison of finger and wrist

Enewoldsen, N.M. (2021) Validation of a wrist-worn photoplethysmographic sensor (E4): comparison of finger and wrist.

[img] PDF
Abstract:The fast rise of wearable technology has led to the development of photoplethysmography (PPG) sensors which enable the monitoring of physical and emotional states (e.g., stress) of users in real-time and in daily life. Thereby, detection and management of stress is a key component of the research, due to the detrimental negative effects of stress on health. However, studies have shown the sensors’ vulnerability towards artifacts, especially motion artifacts, which emphasize the need for accurate validity assessments. The present study implements the validity assessment protocol of van Lier et al. (2019), whereby the wrist-worn PPG sensor of the Empatica E4 is validated by comparing it to a PPG sensor at the fingertip which reflects the reference device (RD). A modified version of the Sing-A-Song-Stress task (SSST) and the Stroop task were used as social and cognitive stressor, in order to induce different intensities of stress. Beside the recording of the physiological arousal, the State Trait Anxiety Inventory (STAI) was used to measure the subjective stress experience. The analysis was performed at the signal, parameter and event level, enabling a comprehensive and standardized validation of the E4. The successful induction of subjective stress for both tasks could not be measured by the physiological measurements. On the one hand, the strict quality inclusion criteria for the analysis resulted in the exclusion of around 40% of the data, while on the other hand, the stressors were found to be not strong enough to induce detectable physiological arousal. Furthermore, at the signal level of the PPG, no relationship was found between the measurements of the two sensors, which is supported by the differences in signal, demonstrated by the analysis at the parameter level. This study questions the validity of the E4 by showing that the validity is lower than described in the literature. Nevertheless, previous research has shown that the E4 can be used for strong and long-lasting stressors, as well as for averages of HR over a longer period of time. Moreover, the validity assessment protocol showed its potential, by enabling the less time- consuming validation with a PPG sensor as RD instead of an electrocardiogram, consequently, making the protocol more available and less restricting with regard to participants and setting.
Item Type:Essay (Master)
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:77 psychology
Programme:Psychology MSc (66604)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page