Non-Contact Cardio-Pulmonary Resuscitation Compression Action Quality Monitoring Based on Depth Camera

• Published in: Proceedings (International Conference on Wearable and Implantable Body Sensor Networks : Print) pp. 1 - 4
• Main Authors: Geng, Fanglin, Zhang, Hao, Yao, Yicheng, Xia, Pan, Wang, Peng, Chen, Xianxiang, Li, Zhenfeng, Du, Lidong, Fang, Zhen
• Format: Conference Proceeding
• Language: English
• Published: IEEE 09.10.2023
• Subjects: Cameras; Cardiac arrest; cardio-pulmonary resuscitation; compression depth; compression frequency; compression quality monitoring; depth camera; Frequency measurement; Image coding; Monitoring; non-contact monitoring; Probes; Quality assessment; Real-time systems; System performance; Time-frequency analysis
• ISSN: 2376-8894
• DOI: 10.1109/BSN58485.2023.10331365

Cardio-pulmonary resuscitation (CPR) is an effective first aid measure to deal with cardiac arrest and is the cornerstone of saving patients ' lives. Chest compression is the most important part of CPR. This work proposed a non-contact CPR compression action quality detection system based on the depth camera in the first time. The RGB camera probe in the system identifies and tracks the position of the rescuer 's hand in the RGB image and maps it to the depth image to obtain the CPR compression curve. Then the CPR parameters such as compression depth and compression frequency are obtained by solving the compression curve. Experiments with different camera deflection angles, different camera measurement distances and different compression frequencies were carried out to evaluate the measurement error of the system. The results showed that under appropriate conditions, the system performance is ideal, and the compression depth and frequency of CPR can be tracked stably.