An enhanced deep image model for glaucoma diagnosis using feature-based detection in retinal fundus

Area of Science:

• Ophthalmology
• Medical Imaging
• Artificial Intelligence

Background:

• Glaucoma is a leading cause of irreversible blindness.
• Early diagnosis is crucial for effective treatment and vision preservation.
• Retinal fundus imaging is a key diagnostic tool.

Purpose of the Study:

• To develop and validate a deep image analysis model for glaucoma detection.
• To improve the accuracy and efficiency of glaucoma diagnosis.
• To assist ophthalmologists in clinical decision-making.

Main Methods:

• A deep image analysis model was developed using retinal fundus images.
• Key features extracted include optic disc and cup parameters.
• The model integrated four machine learning algorithms for classification.

Main Results:

• The proposed model achieved a classification accuracy of 98.60%.
• This surpasses accuracies of Support Vector Machine (97.61%), K-Nearest Neighbors (90.47%), and Naïve Bayes (95.23%).
• The model demonstrated superior performance in glaucoma prediction.

Conclusions:

• The deep image analysis model offers a highly accurate method for early glaucoma diagnosis.
• This approach aids ophthalmologists in making informed decisions for patient eye care.
• The model represents a significant advancement in automated glaucoma screening.