Detection of optic disc in human retinal images utilizing the Bitterling Fish Optimization (BFO) algorithm

Area of Science:

• Ophthalmology and Medical Imaging
• Artificial Intelligence in Healthcare
• Computational Biology

Background:

• Early detection of the optic disc (OD) is crucial for diagnosing ophthalmic conditions such as glaucoma and diabetic retinopathy.
• Conventional OD detection methods face challenges with image noise, illumination variations, and complex image overlaps.
• Automated and precise OD identification is needed to improve diagnostic efficiency and patient outcomes.

Purpose of the Study:

• To develop and evaluate an effective and accurate method for optic disc (OD) detection and delineation in retinal images.
• To enhance the speed and precision of OD imaging processes using the Bitterling Fish Optimization (BFO) algorithm.
• To assess the performance of the BFO algorithm across diverse retinal image datasets.

Main Methods:

• Image enhancement and noise suppression were employed for preprocessing retinal images.
• The Bitterling Fish Optimization (BFO) algorithm was applied to locate and delineate the optic disc region.
• Performance was evaluated using metrics like sensitivity, specificity, accuracy, and DICE coefficient on multiple public datasets (DRIVE, STARE, ORIGA, DRISHTI-GS, DiaRetDB0, DiaRetDB1).

Main Results:

• The BFO-based technique achieved high accuracy rates, including 99.33% (DRIVE), 99.94% (DRISHTI-GS), and 98.22% (DiaRetDB1).
• Demonstrated high overlap and a DICE coefficient of 0.9501 for the DRISHTI-GS database.
• Average processing time per image was under 2.5 seconds, indicating computational efficiency.

Conclusions:

• The BFO algorithm offers an effective, scalable, and automated solution for optic disc detection in retinal images.
• The method exhibits robustness against image quality variations and pathologies, showing potential for clinical application.
• This approach can significantly aid in the early diagnosis and management of ocular diseases.