Tracking blood vessels in human forearms using visual servoing

Drawing an average of more than 2 blood sample per Danish citizen per year increases the demand for an automatic blood sampling method. This paper presents a proof of concept to one of the main challenges in making a fully automated blood sampling procedure, namely: the patient movement compensation. By using images taken with near-infrared light to locate the blood vessels in a human forearm and using the same images to detects movements of the arm, this paper shows that it is possible make a robot arm, potentially equipped with a needle for drawing the blood, compensate for the movements of the patient in real time.