What question did this study set out to answer?

This study aims to evaluate the accuracy and comfort of a novel PPG-based earsensor for vital sign monitoring in surgical patients.

February 14, 2026Open Access

Continuous Vital Signs Monitoring with a Wireless and Wearable Earsensor in Surgical Patients: A Clinical Validation Study

Key Points

This study aims to evaluate the accuracy and comfort of a novel PPG-based earsensor for vital sign monitoring in surgical patients.
Conducted an observational method comparison study
Simultaneously recorded SpO2, PR, and RR using the earsensor and standard monitoring
Analyzed bias, limits of agreement, and average root mean square
Evaluated data loss and patient comfort via a questionnaire
Bias for SpO2 was 1.7% with ARMS of 2.4%
Pulse rate had a bias of 1.2 bpm, ARMS of 3.9 bpm
Respiratory rate bias was 0.3 brpm, ARMS of 4.4 brpm
Data loss was significant across parameters, particularly 42% for SpO2
Patient-reported comfort level was high at 77%

Abstract

(1) Background: Evidence on the clinical accuracy of wireless photoplethysmography (PPG)-based vital sign monitoring is limited. This study evaluated the accuracy, technical performance, and patient comfort of a novel PPG-based earsensor for measuring oxygen saturation (SpO2), pulse rate (PR), and respiratory rate (RR) in postoperative patients. (2) Methods: In this observational method comparison study, SpO2, PR, and RR were simultaneously recorded using the earsensor and compared with continuous monitoring in patients admitted overnight to the post-anesthesia care unit. Outcome measures were bias, 95% limits of agreement (LoA), and average root mean square (ARMS). Technical performance was evaluated by data loss and data gap duration. Patient comfort was assessed using a questionnaire. (3) Results: Twenty-one patients contributed to 264 h of data. Bias was 1.7% for SpO2 (ARMS 2.4%; LoA −1.8% to 5.1%), 1.2 bpm for PR (ARMS 3.9 bpm; LoA –6.1 to 8.4 bpm), and 0.3 brpm for RR (ARMS 4.4 brpm; LoA –8.4 to 8.9 brpm). Overall, data loss was 42% for SpO2, 33% for RR, and 29% for PR; most data gaps were under 30 min. Patient-reported comfort was high (77%). (4) Conclusions: The earsensor accurately measured SpO2 and PR. RR accuracy was outside the predefined criteria. Despite substantial data loss, patient comfort was high, supporting the potential of PPG-based sensors for unobtrusive vital sign trend monitoring in low-acuity settings.

Bookmark

View Full Paper