Remote Bio Vision: Perfusion Imaging Based Non-Contact Autonomic Biosignal Measurement Algorithm

Abstract

Due to the outbreak of new viruses in recent years, the need for contactless medical systems has significantly increased for monitoring, treating, managing, and preventing diseases. Systems are constantly being developed to replace conventional biosignal measuring methods with contactless medical systems. Several systems that substitute attachable sensors in conventional systems have been developed to measure diverse biosignals using images captured with visible cameras. Most non-contact biometric systems have certain limitations, for example, they can only measure a single biosignal, such as the heart rate, oxygen saturation, or blood pressure. They are also easily affected by external factors such as lighting conditions and noises. Furthermore, a blood pressure measuring system based on the pulse transit time requires physical contact and accuracy depends on the measurement location. This study examined a system that provides highly accurate biosignals to overcome these limitations of conventional biosignal measurement systems. This study proposes a biosignal measurement system that uses an infra-red camera to reduce visible light noise. The proposed system uses light sources outside the visible range and measures the amplified reflected light using a near infra-red camera to calculate the heart rate and oxygen saturation. The blood flow value is calculated by the measured heart rate and oxygen saturation. Based on measured data, blood pressure is estimated without contact with the body. We compared the proposed system with an existing non-contact blood pressure measurement system. The proposed system showed an average error rate of 2.07% in blood pressure measurement, which increased by 3.86% compared to the conventional system. Also, this study enabled the early detection of pressure ulcers, which is difficult to measure quantitatively.