Multi-Modal Imaging-Guided Robotic System for Automated Venipuncture

Abstract

This paper presents a collaborative robotic system designed for automated venipuncture. The system integrates a 6 -axis collaborative robot with a custom multifunctional tool equipped with a near-infrared (NIR) camera, an RGB-D camera, and dedicated LED and IR illumination arrays. This multi-modal imaging setup enables robust vein detection by combining color, depth, and NIR data to accurately localize superficial veins and estimate their depth. A gripper mechanism mounted on the robot holds the tool, which includes a retractable needle for safe and precise insertion. The system is designed for use in clinical environments, aiming to reduce the failure rate of manual venipuncture and enhance patient safety. The collaborative nature of the robot ensures safe interaction with medical staff and patients. Initial tests on phantom arms demonstrate the feasibility and effectiveness of the approach, suggesting its potential for future deployment in hospitals and blood collection centers. © 2025 IEEE.