Respiratory rate is a vital sign that signifies the movement of air into and out of the lungs through ventilation. In modern medical practices, this vital sign is often inaccurately measured by hospital staff [1]. This phenomenon is problematic considering a fluctuation in respiratory rate is often observed as an early indicator of illness. The recent COVID-19 pandemic resulted in an influx of patients visiting hospitals and a greater need for hospital staff to monitor the vitals of patients in case of respiratory distress. To aid in the important measuring of patient vitals, a device was created to non-invasively measure respiratory rate. This portable, cost-effective device utilizes thermal analysis and facial recognition through two cameras to measure and calculate a patient's respiratory rate from a distance of 30-60 cm away. The respiratory rate is then displayed via a smartphone application. A group of subjects participated in a testing procedure to determine the accuracy of the device. Subject respiration was recorded by the device over a period of 30 seconds. The results were compared to a manual count, verified by a real-time heart rate monitor. The device’s algorithm was found to have an increase in accuracy compared to a previous study of respiratory rates manually counted by trained medical staff [2]. This device, the LeTourneau Engineering Vital Imaging System (LEVIS), provides a more accurate measurement of respiratory rate, which can enable medical staff to attend to other tasks during the measurement period.

© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.

Savannah M. Orton, Erin R. O'Brien, Ashley D. DeCock, Julian J. Lee, Samson A. Segovia, Ian C. Crochet, Oladimeji A. Apaokagi, Evan J. Kowal, Morgan B. Nix, Francis I. Ko, Esther H. Kim, and Joon W. Kim, "Non-invasive respiratory rate detection using thermal imaging and facial recognition," Proc. SPIE 12488, Health Monitoring of Structural and Biological Systems XVII, 124881G (Presented at SPIE Smart Structures + Nondestructive Evaluation: March 15, 2023; Published: 25 April 2023); https://doi.org/10.1117/12.2657803.