Optimizing vision care: Dual path network model in eye disease classification.

Area of Science:

• Ophthalmology
• Medical Imaging
• Artificial Intelligence

Background:

• Global rise in eye diseases necessitates advanced diagnostic tools.
• Early detection of conditions like cataracts, glaucoma, and diabetic retinopathy is vital for preventing vision loss.
• Current diagnostic methods require enhancement for efficiency and accuracy.

Purpose of the Study:

• To develop and evaluate an innovative AI model for automated eye disease classification.
• To leverage the Dual Path Network (DPN) for enhanced feature extraction from retinal images.
• To improve the accuracy and efficiency of early eye disease detection.

Main Methods:

• Utilized a dataset of retinal images from Kaggle, categorized into Normal, Diabetic Retinopathy, Cataract, and Glaucoma.
• Implemented a Dual Path Network (DPN) model for simultaneous capture of global and local image features.
• Evaluated model performance using standard classification metrics.

Main Results:

• The DPN model achieved a classification accuracy of 95.32%.
• The model demonstrated a strong ability to reduce false positives while accurately identifying positive cases.
• The approach proved effective in comprehensive information extraction from retinal images.

Conclusions:

• The proposed DPN model is a proficient tool for accurate eye disease classification.
• This AI-driven approach shows significant potential for early detection and diagnosis of major ocular conditions.
• The model can assist healthcare professionals in making informed decisions, improving patient outcomes and reducing the burden of eye diseases.