Automatic Optic Disc Segmentation Based on Modified Local Image Fitting Model with Shape Prior Information.

Area of Science:

• Ophthalmology
• Medical Imaging
• Computer Vision

Background:

• Accurate optic disc (OD) detection is crucial for diagnosing retinal diseases.
• Current methods often require manual initialization or struggle with image quality variations.

Purpose of the Study:

• To develop a fully automatic optic disc boundary segmentation approach.
• To enhance the accuracy and robustness of optic disc detection in retinal images.

Main Methods:

• Proposed a novel approach for automatic optic disc segmentation using saliency detection and thresholding for adaptive initialization.
• Introduced a modified Level Set (LS) approach incorporating shape prior information to handle unreliable image intensity data.

Main Results:

• The developed method achieved accurate optic disc boundary segmentation without manual intervention.
• Demonstrated superior performance compared to existing methods on the DIARETDB0 database.
• Achieved high accuracy rates and average overlapping ratios.

Conclusions:

• The proposed fully automatic double boundary extraction method is effective for optic disc segmentation.
• The approach offers improved accuracy and robustness, particularly in abnormal retinal images.
• This technique can aid in more reliable and efficient retinal disease diagnosis.