Related Experiment Video
Updated: Oct 4, 2025
Non-invasive Assessment of Microvascular and Endothelial Function
Published on: January 29, 2013
EVA: Fully automatic hemodynamics assessment system for the bulbar conjunctival microvascular network
Zhaoqiang Yun1, Qing Xu2, Gengyuan Wang3
1School of Biomedical Engineering and Guangdong Provincial Key Laboratory of Medical Image Processing, Southern Medical University, Guangzhou, China; Guangdong Province Engineering Laboratory for Medical Imaging and Diagnostic Technology, Southern Medical University, Guangzhou, China.
This study introduces an EVA system using deep learning for continuous conjunctival microvessel segmentation and automatic blood velocity measurement. The system offers a reliable method for assessing microvascular hemodynamics in systemic disorders.
Area of Science:
- Ophthalmology
- Biomedical Engineering
- Medical Imaging
Background:
- Conjunctival microcirculation analysis aids in assessing systemic disorders.
- Red blood cell clustering causes segmentation discontinuities, hindering automatic blood velocity measurement.
- Accurate hemodynamic assessment requires continuous vessel segmentation and precise velocity measurements.
Purpose of the Study:
- To develop an Eye Vascular Analysis (EVA) system using deep learning for continuous conjunctival microvessel segmentation.
- To enable automatic and reliable measurement of blood velocity in conjunctival microvessels.
- To quantitatively assess microvascular hemodynamics in the context of systemic disorders.
Main Methods:
- Developed an EVA system integrating image registration, segmentation, diameter, and velocity measurements.
- Utilized a U-Net model with a connectivity-preserving loss function to overcome segmentation discontinuities.
- Implemented an automatic blood velocity measurement algorithm based on line segment detection.
Main Results:
- The U-Net model achieved superior vessel segment length (158.03 µm) compared to other methods.
- The proposed method demonstrated higher consistency with manual measurements for blood velocity (bias: -0.04, SD: 0.14) than cross-correlation (bias: 0.36, SD: 0.32).
- The EVA system successfully assessed hemodynamic parameters from conjunctival microcirculation videos.
Conclusions:
- The EVA system provides an automatic and reliable solution for quantitative hemodynamic assessment in conjunctival microvascular images.
- The system's potential application extends to other microcirculation imaging, such as hypoglossal microcirculation.
- This technology can advance the understanding and monitoring of systemic disorders through microvascular analysis.
