Enhanced Vein Detection for Venipuncture Using Near-Infrared Illumination and CLAHE

The use of venous blood for diagnostic tests is a common yet critical procedure in medical practice. However, locating suitable veins for cannulation or phlebotomy remains a significant challenge, especially in patients with factors like obesity, dark skin tone, or age-related venous changes, leading to high pre-analytical error rates and patient discomfort. This paper proposes a low-cost, non-invasive vein detection system utilizing Near-Infrared (NIR) illumination and real-time image processing. The system employs an infrared dome camera with an NIR LED ring to capture subcutaneous vein patterns. The captured video stream is processed on a Raspberry Pi using Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance vein contrast against the surrounding tissue. The processed image is then displayed on an LCD screen, providing a clear, real-time map of the patient's vasculature. Experimental results demonstrate that the system effectively visualizes veins that are not visible to the naked eye, offering a portable and affordable solution to improve the success rate of first-attempt venipuncture, thereby reducing patient pain and associated complications.