Abstract
This project presents a novel approach for vein recognition using Convolutional Neural Networks (CNN). The motivation behind this research is to address the challenges associated with vein detection in healthcare settings, such as the manual and error-prone process of locating veins for procedures like blood draws and IV insertions. The methodology employed in this project involved collecting a dataset of forearm near-infrared (NIR) images, preprocessing the data, and training a CNN model to accurately identify and classify veins. The findings of this study demonstrate the effectiveness and novelty of using CNNs for vein recognition, achieving high accuracy rates in identifying veins in various settings. The actual work performed included data collection from the GitHub repository, implementation of the CNN algorithm, training the model, and evaluating its performance. The results obtained showcase the potential of CNN-based vein recognition systems in improving the success rate of vein punctures, enhancing patient safety, and optimizing healthcare practices. This project contributes to the field by providing innovative insights and solutions to vein recognition, opening avenues for further advancements in medical imaging and healthcare technologies.
Metadata
Item Type: | Thesis (Degree) |
---|---|
Creators: | Creators Email / ID Num. Wan Mohd Rohaizan @ Wan Mohd Fakruddin, Wan Nur Nadia Shafiqah 2022937635 |
Contributors: | Contribution Name Email / ID Num. Thesis advisor Fadzal, Ahmad Nazmi UNSPECIFIED |
Subjects: | Q Science > QA Mathematics > Instruments and machines > Electronic Computers. Computer Science > Algorithms |
Divisions: | Universiti Teknologi MARA, Terengganu > Kuala Terengganu Campus |
Programme: | Bachelor of Computer Science (Hons) |
Keywords: | Vein Recognition, Convolutional Neural Network (CNN) |
Date: | 2024 |
URI: | https://ir.uitm.edu.my/id/eprint/96289 |
Download
96289.pdf
Download (90kB)