Related Experiment Video
Updated: Nov 23, 2025
07:20
Dried Blood Spot Collection of Health Biomarkers to Maximize Participation in Population Studies
Published on: January 28, 2014
37.0K
Research on Finger Vein Image Segmentation and Blood Sampling Point Location in Automatic Blood Collection.
Xi Li1,2, Zhangyong Li3, Dewei Yang1
1School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China.
Sensors (Basel, Switzerland)
|December 31, 2020
Summary
Accurate fingertip blood sampling requires precise vein localization. This study introduces a novel finger vein segmentation method using Gabor transform and Gaussian mixed model (GMM) for improved accuracy in automated blood collection.
Area of Science:
- Biomedical Engineering
- Medical Imaging
- Computer Vision
Background:
- Accurate localization of fingertip veins is crucial for automated blood sampling to prevent excessive bleeding.
- Existing methods may struggle with precise vein identification, impacting sample volume and patient comfort.
Discussion:
- A new finger vein image segmentation approach is proposed, integrating Gabor transform and Gaussian mixed model (GMM).
- Adaptive Gabor filter parameterization and Local Binary Pattern (LBP) fusion enhance feature extraction.
- Segmentation is optimized using max flow min cut based on relative entropy, followed by corner detection for precise blood sampling point localization.
Key Insights:
- The proposed method achieves an average segmentation accuracy of 91.6% for finger vein images.
- Adaptive feature extraction and optimized segmentation significantly improve vein localization precision.
- This technique offers a robust solution for accurate blood sampling point identification in automated systems.
Outlook:
- Potential for integration into next-generation automated blood collection devices.
- Further research could explore real-time application and adaptation to different imaging conditions.
- Validation across diverse populations and fingertip variations is recommended for broader clinical adoption.
