Noncontact Measurement of Respiratory Rate by an Optical Time of Flight Sensor

Depth measurements are becoming increasingly popular in many fields, particularly biomedical applications. Existing time-of-flight cameras suffer from low spatial resolution due to additional processing at the sensor level. In this paper, we propose a low-cost, off-the-shelf, and non-contact measurement system based on an indirect time-of-flight sensor for respiratory rate measurements. To this end, the performance of the single depth sensor in short-range distance measurements has been examined using a ground truth laser interferometer. Tissue-simulating abdominal wall surfaces were used to validate the accuracy of the proposed device. The simulated surfaces had six different reflectance levels, corresponding to the standard Fitzpatrick scale of human skin color. Our results show excellent agreement between the expected and extracted values of displacement and vibration Frequency using a fast Fourier transform of the time-varying displacements. The clinical performance of the proposed sensor was verified by thirty human respiratory rate measurements. It showed an overall mean absolute error (MAE) of 0.43 breaths/min. Respiratory rate measurements also showed a strong and positive relationship with the clinical reference method (r =0.99, p <0.001) and Bland-Altman analysis revealed a bias of -0.18 breaths/min. Besides, the accuracy of the sensor is well suited, on average, for clinical measurements over the skin reflectivity range of 20-80%. The findings of this work provide a potentially simple, low-cost solution for remote displacement and vibration measurement, making it a promising tool for remote health monitoring in clinical settings.