Detection of Optic Disc Localization from Retinal Fundus Image Using Optimized Color Space

Area of Science:

• Ophthalmology
• Medical Imaging
• Computer Vision

Background:

• Optic disc localization is vital for identifying retinal structures and preventing vision loss from diseases like age-related macular degeneration and diabetic retinopathy.
• Computer-aided diagnosis systems are increasingly important for early detection and intervention.

Purpose of the Study:

• To propose an automated method for accurate optic disc localization in fundus images.
• To enhance the clarity of the optic disc region compared to standard RGB color space.

Main Methods:

• Developed an automated method utilizing an artificial bee colony algorithm to transform fundus images into a new color space.
• Created a feature matrix from image patches and an optimized conversion matrix via the artificial bee colony algorithm.
• Applied thresholding to the transformed images for optic disc localization.

Main Results:

• The proposed method achieved high accuracy on three public datasets: 100% on DRIVE, 96.37% on DRIONS, and 94.42% on MESSIDOR.
• The artificial bee colony algorithm optimized the color space conversion for clearer optic disc identification.
• The automated approach demonstrated robust performance across diverse datasets.

Conclusions:

• The proposed automated method effectively localizes the optic disc with high accuracy.
• This technique shows significant potential for improving early diagnosis of retinal diseases.
• The artificial bee colony algorithm offers a novel approach to enhancing medical image analysis for ophthalmology.