NON-CONTACT TECHNIQUES FOR HUMAN VITAL SIGN DETECTION AND GAIT ANALYSIS

Human vital signs including respiratory rate, heart rate, oxygen saturation, blood pressure, and body temperature are important physiological parameters that are used to track and monitor human health condition. Another important biological parameter of human health is human gait. Human vital sign detection and gait investigations have been attracted many scientists and practitioners in various fields such as sport medicine, geriatric medicine, bio-mechanic and bio-medical engineering and has many biological and medical applications such as diagnosis of health issues and abnormalities, elderly care and health monitoring, athlete performance analysis, and treatment of joint problems. Thoroughly tracking and understanding the normal motion of human limb joints can help to accurately monitor human subjects or patients over time to provide early flags of possible complications in order to aid in a proper diagnosis and development of future comprehensive treatment plans.

With the spread of COVID-19 around the world, it has been getting more important than ever to employ technology that enables us to detect human vital signs in a non-contact way and helps protect both patients and healthcare providers from potentially life-threatening viruses, and have the potential to also provide a convenient way to monitor people health condition, remotely. A popular technique to extract biological parameters from a distance is to use cameras. Radar systems are another attractive solution for non-contact human vital signs monitoring and gait investigation that track and monitor these biological parameters without invading people privacy.

The goal of this research is to develop non-contact methods that is capable of extracting human vital sign parameters and gait features accurately.

To do that, in this work, optical systems including cameras and proper filters have been developed to extract human respiratory rate, heart rate, and oxygen saturation. Feasibility of blood pressure extraction using the developed optical technique has been investigated, too. Moreover, a wideband and low-cost radar system has been implemented to detect single or multiple human subject’s respiration and heart rate in dark or from behind the wall. The performance of the implemented radar system has been enhanced and it has been utilized for non-contact human gait analysis. Along with the hardware, advanced signal processing schemes have been enhanced and applied to the data collected using the aforementioned radar system. The data processing algorithms have been extended for multi-subject scenarios with high accuracy for both human vital sign detection and gait analysis. In addition, different configurations of this and high-performance radar system including mono-static and MIMO have been designed and implemented with great success. Many sets of exhaustive experiments have been conducted using different human subjects and various situations and accurate reference sensors have been used to validate the performance of the developed systems and algorithms.