Development of Automatic 3D Blood Vessel Search and Automatic Blood Sampling System by Using Hybrid Stereo-Autofocus Method

Abstract

We developed an accurate three-dimensional blood vessel search (3D BVS) system and an automatic blood sampling system. They were implemented into a point-of-care system designed for medical care, installed in a portable self-monitoring blood glucose (SMBG) device. The system solves problems of human error caused by complicated manual operations of conventional SMBG devices. We evaluated its accuracy of blood-vessel position detection. The 3D BVS system uses near-infrared (NIR) light imaging and the stereo and autofocus hybrid method to determine blood vessel locations accurately in three dimensions. We evaluated the accuracy of our 3D BVS system using a phantom of human skin, blood vessels, and blood. Additionally, we established an automatic blood sampling system for SMBG and assessed its performance in relation to punctures, blood suction, transport, and discharge on an enzyme sensor. The 3D BVS and automatic blood sampling system are adequate for use in a portable SMBG device.