Near-Infrared Spectroscopy for Vein Visualization

Near-Infrared spectroscopy is used for better vein visualization to make the venipuncture process more efficient. While there exist a few models which use the said mechanism, these models are costly, have accuracy issues, and are limited only to certain types of skin tones. Some of the available devices use image-guided venipuncture technique and the others use projection.

We propose a low-cost mechanism of obtaining near-infrared spectroscopy by using the image-guided technique. We decided on using this technique after assessing both the available techniques. The low-cost is achieved by optimizing the image processing algorithms and adjusting the illumination method. We have tested and optimized the algorithms accordingly.

We use near-infrared LEDs as the source of illumination, and a CMOS camera for image acquisition. Images are processed using OpenCV, and Histogram equalization and CLAHE algorithms are used in preprocessing. Initially, we processed the still images and later on developed the model to process the live video stream and display the processed video footage that visualizes the veins in real-time. We display the vein map on a 7 inch IPS LCD screen.

We have tested the prototype using different combinations of light sources with different intensities and have analyzed the results. We have also analyzed how the results vary based on body fat. In order to quantitatively analyze, we have obtained a count of the number of visible veins and depicted the comparison in a graph. We have concluded that a higher intensity does not always increase the visibility of veins. Our plan is to conduct a clinical trial and test the device on human subjects and get the feedback from both the patients and phlebotomists and improve the model so that those final users are satisfied.