Noncontact respiratory measurement of volume change using depth camera

Abstract

In this study, a system is developed to measure human chest wall motion for respiratory volume estimation without any physical contact. Based on depth image sensing technique, respiratory volume is estimated by measuring morphological changes of the chest wall. We evaluated the system and compared with a standard reference device, and the results show strong agreement in respiratory volume measurement [correlation coefficient: r=0.966]. The isovolume test presents small variations of the total respiratory volume during the isovolume maneuver (standard deviation<107 ml). Then, a regional pulmonary measurement test is evaluated by a patient, and the results show visibly difference of pulmonary functional between the diseased and the contralateral sides of the thorax after the thoracotomy. This study has big potential for personal health care and preventive medicine as it provides a novel, low-cost, and convenient way to measure user's respiration volume.

Measurement of chest wall motion using a motion capture system with the one-pitch phase analysis method

“…We conceive that our research has the advantage of simplicity because of its non-contacting nature. Yu et al 11 and Sharp et al 12 used only one Kinect depth camera with a non-contacting manner in capturing chest wall motion in their study, and they showed a high correlation of respiratory volume measured by between spirometry and Kinect-based method. There are several reasons for the discrepancy between these studies and our study in terms of the correlation of two analyzing methods.…”

“…Although time-of-flight depth camera sensor also can measure thoracic wall motion 10 – 13 , some of these methods apply marker attachment or use multiple cameras. Yu et al and Sharp et al described thoracic volume analysis using Kinect with a single camera and a non-contact manner 11 , 12 . However, structured pattern projection methods used in Kinect rely on the accuracy of the pixel unit.…”

Detection of respiratory activity in newborn infants using a noncontact vision‐based monitor

“…This device has been used for respiratory monitoring research in infants, children, and adults. 14,[23][24][25][26][27][28] Produced in both structured light plethysmography (SLP) and TOF versions, in both cases with an RGB camera, the older SLP and newer TOF devices differ in data performance characteristics (resolution, working distance, etc). The SLP version of the Kinect has been used to accurately measure tidal volume in adults, [23][24][25] and was found to be potentially feasible for RR monitoring.…”

“…14,[23][24][25][26][27][28] Produced in both structured light plethysmography (SLP) and TOF versions, in both cases with an RGB camera, the older SLP and newer TOF devices differ in data performance characteristics (resolution, working distance, etc). The SLP version of the Kinect has been used to accurately measure tidal volume in adults, [23][24][25] and was found to be potentially feasible for RR monitoring. 28 The TOF version of the Kinect has also been used to detect respiration in children, with good correlation with tidal volume measured with a piezo respiratory belt transducer, 14 or a ventilator, 26 in the latter case with a multicamera system monitoring a baby mannequin.…”

Noncontact Respiratory Monitoring Using Depth Sensing Cameras: A Review of Current Literature

“…Other studies include those by Harte et al [44] who studied respiratory volume in groups of healthy and cystic fibrosis patients; Samir et al [45] who compared respiratory volume measurement using both a Kinect V1 and Kinect V2; Ostadabbas et al [46] who investigated respiratory volume and airway resistance in a study of the correlation of FEV1 derived from a spirometer and depth sensing system; Ernst et al [47] who completed a study on respiratory motion tracking with a depth camera; Yang et al [48] who included sleep patients in a study of sleep event detection and respiratory volume; Shan et al [49] who also investigated respiratory volume signals in the context of stress classification; Kempfle and Van Laerhoven [50] who propose modelling chest elevation to robustly monitor a user's respiration, whenever users are sitting or standing or the view is occasionally blocked; some smaller scale proof of principle studies by Yu et al [51], Prochazka et al [52], and Aoki et al [53].…”