Dynamic Real-Time Heart Rate and Blood Pressure Detection for Multiple Individuals

• Published in: 2023 International Conference on Networks, Communications and Intelligent Computing (NCIC) pp. 152 - 156
• Main Authors: Guang, Litao, Zou, Jiancheng
• Format: Conference Proceeding
• Language: English
• Published: IEEE 17.11.2023
• Subjects: blood pressure; Face recognition; Head; Heart rate; image Photoplethysmography; pedestrian tracking; Pedestrians; Photoplethysmography; Pose estimation; posture estimation; Real-time systems
• DOI: 10.1109/NCIC61838.2023.00031

Heart rate and blood pressure are important physiological parameters that can to some extent reflect the health status of the human body. The development of contactless testing has made it more convenient, but most of it is aimed at single individuals and is static. This article proposes a real-time multi person dynamic continuous heart rate and blood pressure measurement method. It is based on the principle of image Photoplethysmography, and unlike most non-contact detection methods that rely on facial recognition and tracking, it will achieve the selection of regions of interest through pose estimation combined with hand detection, and use pedestrian tracking to achieve tracking. Compare the detection results of this article with the widely accepted smart bracelet or Omron electronic blood pressure gauge in the market. And conducting Bland Altman analysis, the mean differences in heart rate, systolic blood pressure, and diastolic blood pressure were -0.82, 1.73, and 1.67, respectively. The majority of the points were distributed within a 95% confidence interval, indicating good consistency.