Respiratory rate detection algorithm based on RGB‐D camera: theoretical background and experimental results

Abstract

Both the theoretical background and the experimental results of an algorithm developed to perform human respiratory rate measurements without any physical contact are presented. Based on depth image sensing techniques, the respiratory rate is derived by measuring morphological changes of the chest wall. The algorithm identifies the human chest, computes its distance from the camera and compares this value with the instantaneous distance, discerning if it is due to the respiratory act or due to a limited movement of the person being monitored. To experimentally validate the proposed algorithm, the respiratory rate measurements coming from a spirometer were taken as a benchmark and compared with those estimated by the algorithm. Five tests were performed, with five different persons sat in front of the camera. The first test aimed to choose the suitable sampling frequency. The second test was conducted to compare the performances of the proposed system with respect to the gold standard in ideal conditions of light, orientation and clothing. The third, fourth and fifth tests evaluated the algorithm performances under different operating conditions. The experimental results showed that the system can correctly measure the respiratory rate, and it is a viable alternative to monitor the respiratory activity of a person without using invasive sensors.

Remote, Depth-Based Lung Function Assessment

“…The best coefficient of determination (R 2 ) between the spirometer signal and the estimated airflow signal was reported as 0.93. Benetazzo et al [12] detected respiratory rates by applying a weighted averaging filter to the chest region pixels segmented by using the first generation Kinect skeleton's shoulder and torso joint positions. Their breathing rate results were evaluated against a spirometer, with an outcome of 0.98 correlation.…”

“…Chest-averaging-based volume estimation-Similar to previous approaches [6], [12], [15], [17], [18], we also estimated the uncalibrated chest volume at time point t by computing the average distance of each pixel located in the chest region. Chest-averaging is simple and fast to compute.…”

“…This correlation is less strong for the TV measure error and tidal volume scaling factor error. The reason for this is that tidal volume scaling factors are computed using all data points of tidal volume part of volume-time curve and TV measure itself is computed using group keypoints F and G (12). However, FVC measure and main effort scaling factors are both computed using the same keypoints A and B.…”

3D imaging system for respiratory monitoring in pediatric intensive care environment

“…Due to these constraints, previous works on contactless RR and Vt assessment were limited to proofs of concept. They were conducted for a general use case, and are thus inappropriate for Pediatric Intensive Care Environment (PICE) (Benetazzo et al, 2014;Ostadabbas et al, 2014;Katashev et al, 2015;Harte et al, 2016;Sharp et al, 2017).…”

Markerless Active Trunk Shape Modelling for Motion Tolerant Remote Respiratory Assessment

“…The medical achievements of our remote approaches were reported in [6,7]. Apart from our works [3][4][5][6][7], we know of only [8,9] which also performed remote respiratory assessment, rather than just breathing rate estimation or respiration monitoring [10][11][12][13][14][15]. In [8,9], Ostadabbas et al detected airway obstruction as mild, moderate and severe, and computed only FEV1 in [8].…”