An Effective Detection Mechanism for Localizing Macular Region and Grading Maculopathy

Area of Science:

• Ophthalmology
• Medical Imaging
• Computer Vision

Background:

• Diabetic retinopathy is a significant cause of vision loss, particularly in diabetic populations.
• Accurate segmentation of retinal features is crucial for grading diabetic retinopathy.
• Existing segmentation methods have limitations in accuracy and computational efficiency.

Purpose of the Study:

• To develop an improved image segmentation method for detecting diabetic retinopathy.
• To enhance the accuracy and efficiency of identifying retinal lesions.
• To create a user-friendly system for ophthalmologists to grade diabetic retinopathy.

Main Methods:

• Utilized K-means clustering for initial coarse segmentation of retinal images.
• Applied mathematical morphology for background feature removal and image reconstruction.
• Developed a graphical user interface for simplified system operation.

Main Results:

• Achieved high accuracy in segmenting hard exudates, outperforming classical approaches.
• Demonstrated a sensitivity of 96.4% and a specificity of 97.2%.
• The developed system effectively divides images into regions of interest with lesions and normal features.

Conclusions:

• The proposed method offers a more accurate and computationally efficient approach to diabetic retinopathy segmentation.
• The integrated graphical user interface facilitates practical application by ophthalmologists.
• This tool can aid in the timely grading and treatment planning for patients with diabetic retinopathy.