Vein segmentation and visualization of upper and lower extremities using convolution neural network

• Published in: Biomedizinische Technik Vol. 69; no. 5; pp. 455 - 464
• Main Authors: Laddi, Amit, Goyal, Shivalika, Himani, Savlania, Ajay
• Format: Journal Article
• Language: English
• Published: Germany De Gruyter 28.10.2024; Walter de Gruyter GmbH
• Subjects: Algorithms; Artificial neural networks; CNN architecture; Convolution; Datasets; Deep Learning; Extremities; Heart diseases; Humans; Image acquisition; Image processing; Image Processing, Computer-Assisted - methods; Image segmentation; Infrared imaging; Localization; Lower Extremity - blood supply; Lower Extremity - diagnostic imaging; Machine learning; near infrared imaging; Neural networks; Neural Networks, Computer; Parameterization; Phlebotomy; Real time; Segmentation; Upper Extremity - blood supply; Upper Extremity - diagnostic imaging; vein visualization; Veins; Veins & arteries; Veins - diagnostic imaging; Visualization; CNN architecture; deep learning; image segmentation; vein visualization; near infrared imaging
• ISSN: 0013-5585; 1862-278X; 1862-278X
• DOI: 10.1515/bmt-2023-0331

The study focused on developing a reliable real-time venous localization, identification, and visualization framework based upon deep learning (DL) self-parametrized Convolution Neural Network (CNN) algorithm for segmentation of the venous map for both lower and upper limb dataset acquired under unconstrained conditions using near-infrared (NIR) imaging setup, specifically to assist vascular surgeons during venipuncture, vascular surgeries, or Chronic Venous Disease (CVD) treatments. A portable image acquisition setup has been designed to collect venous data (upper and lower extremities) from 72 subjects. A manually annotated image dataset was used to train and compare the performance of existing well-known CNN-based architectures such as ResNet and VGGNet with self-parameterized U-Net, improving automated vein segmentation and visualization. Experimental results indicated that self-parameterized U-Net performs better at segmenting the unconstrained dataset in comparison with conventional CNN feature-based learning models, with a Dice score of 0.58 and displaying 96.7 % accuracy for real-time vein visualization, making it appropriate to locate veins in real-time under unconstrained conditions. Self-parameterized U-Net for vein segmentation and visualization has the potential to reduce risks associated with traditional venipuncture or CVD treatments by outperforming conventional CNN architectures, providing vascular assistance, and improving patient care and treatment outcomes.