Related Experiment Video
Updated: Jun 15, 2026
10:11
Fundus Photography as a Convenient Tool to Study Microvascular Responses to Cardiovascular Disease Risk Factors in Epidemiological Studies
Published on: October 22, 2014
A study on hemorrhage detection using hybrid method in fundus images.
Jang Pyo Bae1, Kwang Gi Kim, Ho Chul Kang
1Biomedical Engineering Branch, Division of Basic & Applied Sciences, National Cancer Center, 111 Jungbalsan-ro, Ilsandong-gu, Goyang-si, Gyeonggi-do, South Korea.
Journal of Digital Imaging
|February 24, 2010
Summary
This study introduces a novel image processing technique for early diabetic retinopathy detection. The method effectively extracts retinal hemorrhages using template matching, aiding in disease progression monitoring.
Area of Science:
- Ophthalmology
- Medical Image Analysis
- Computer Vision
Background:
- Diabetic retinopathy is a leading cause of vision loss.
- Early detection and monitoring are crucial for managing diabetic retinopathy.
- Automated image analysis of fundus images can aid in early detection.
Purpose of the Study:
- To develop and evaluate an automated method for extracting retinal hemorrhages from fundus images.
- To improve the accuracy and efficiency of diabetic retinopathy detection using image processing techniques.
- To assess the potential of the developed method for early detection of diabetic retinopathy progression, particularly in telemedicine.
Main Methods:
- Applied hue saturation value brightness correction and contrast-limited adaptive histogram equalization to fundus images.
- Utilized template matching with normalized cross-correlation for candidate hemorrhage extraction.
- Employed region growing to reconstruct hemorrhage shapes and calculate their sizes.
- Implemented false positive reduction techniques including compactness, bounding box ratios, kernel values, and a foveal filter.
Main Results:
- Achieved a sensitivity of 85% with 4.0 false positives per image using a combination of template matching methods.
- Successfully extracted and quantified retinal hemorrhages.
- Analyzed the causes of false positives and false negatives in hemorrhage detection.
Conclusions:
- The developed image processing program shows promise for assisting clinicians in diagnosing diabetic retinopathy.
- The method can be a valuable tool for the early detection of diabetic retinopathy progression.
- The technique is particularly suitable for telemedicine applications, enhancing remote patient monitoring.
