Coinciding Diabetic Retinopathy and Diabetic Macular Edema Grading With Rat Swarm Optimization Algorithm for Enhanced...

Area of Science:

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

Background:

• Diabetic retinopathy (DR) and diabetic macular edema (DME) are leading causes of vision impairment and blindness in the working-age population globally.
• Diabetes prevalence is increasing, leading to a higher incidence of diabetes-related eye conditions.
• Early diagnosis and grading of DR and DME are critical for timely intervention and vision preservation.

Purpose of the Study:

• To propose an enhanced capsule generation adversarial network (ECGAN) optimized with rat swarm optimization (RSO) for accurate grading of DR and DME.
• To evaluate the performance of the proposed DR-DME-ECGAN-RSO-ISBI 2018 IDRiD model using the ISBI 2018 unbalanced DR grading dataset.
• To compare the proposed model's accuracy against existing state-of-the-art methods for DR and DME grading.

Main Methods:

• Image preprocessing using the Savitzky-Golay (SG) filter to reduce noise in fundus images.
• Feature extraction from preprocessed images using discrete shearlet transform (DST).
• Classification and grading of DR and DME using the ECGAN-RSO algorithm.

Main Results:

• The proposed DR-DME-ECGAN-RSO-ISBI 2018 IDRiD model achieved superior accuracy compared to existing methods.
• Accuracy improvements were noted as 7.94%, 36.66%, and 4.88% over specific comparative models.
• The model demonstrated effectiveness in grading DR and DME from fundus images.

Conclusions:

• The developed DR-DME-ECGAN-RSO-ISBI 2018 IDRiD approach offers a promising advancement in the automated grading of diabetic eye diseases.
• The integration of ECGAN with RSO provides a robust framework for enhancing diagnostic accuracy in ophthalmology.
• This AI-driven method has the potential to aid clinicians in early detection and management of DR and DME, thereby preventing vision loss.