Artificial intelligence based glaucoma and diabetic retinopathy detection using MATLAB - retrained AlexNet convolutional neural network

Abstract

Background: Glaucoma and diabetic retinopathy (DR) are the leading causes of irreversible retinal damage leading to blindness. Early detection of these diseases through regular screening is especially important to prevent progression. Retinal fundus imaging serves as the principal method for diagnosing glaucoma and DR. Consequently, automated detection of eye diseases represents a significant application of retinal image analysis. Compared with classical diagnostic techniques, image classification by convolutional neural networks (CNN) exhibits potential for effective eye disease detection.

Methods: This paper proposes the use of MATLAB - retrained AlexNet CNN for computerized eye diseases identification, particularly glaucoma and diabetic retinopathy, by employing retinal fundus images. The acquisition of the database was carried out through free access databases and access upon request. A transfer learning technique was employed to retrain the AlexNet CNN for non-disease (Non_D), glaucoma (Sus_G) and diabetic retinopathy (Sus_R) classification. Moreover, model benchmarking was conducted using ResNet50 and GoogLeNet architectures. A Grad-CAM analysis is also incorporated for each eye condition examined.

Results: Metrics for validation accuracy, false positives, false negatives, precision, and recall were reported. Validation accuracies for the NetTransfer (I-V) and netAlexNet ranged from 89.7% to 94.3%, demonstrating varied effectiveness in identifying Non_D, Sus_G, and Sus_R categories, with netAlexNet achieving a 93.2% accuracy in the benchmarking of models against netResNet50 at 93.8% and netGoogLeNet at 90.4%.

Conclusions: This study demonstrates the efficacy of using a MATLAB-retrained AlexNet CNN for detecting glaucoma and diabetic retinopathy. It emphasizes the need for automated early detection tools, proposing CNNs as accessible solutions without replacing existing technologies.

Keywords: AlexNet; Classification; Convolutional Neural Network (CNN); Diabetic Retinopathy; Glaucoma.

Figure 1.. Retraining and benchmarking of AlexNet for non-disease (Non_D), glaucoma (Sus_G) and diabetic retinopathy (Sus_R) detection.

Retinal fundus images retrieved from High-Resolution Fundus (HRF) Image Database.

Figure 2.. Proposed system for glaucoma and diabetic retinopathy detection using AlexNet.

Figure 3.. Data Organization generated by MATLAB for Training and Validation Sets.

Glossary: Non-disease (Non_D), glaucoma (Sus_G) and diabetic retinopathy (Sus_R).

Figure 4.. Proposed neural network architecture for eye diseases detection based on AlexNet.

Figure 5.. Confusion matrix for accuracy of the retrained AlexNet convolutional neural network on all datasets for the eye disease detection.

Glossary: False Negative (FN), False Positive (FP), Non-disease (Non_D), glaucoma (Sus_G) and diabetic retinopathy (Sus_R).

Figure 6.. Confusion matrix for accuracy of the retrained AlexNet, ResNet50 & GoogLeNet convolutional neural networks for the eye disease detection.

Glossary: False Negative (FN), False Positive (FP), Non-disease (Non_D), glaucoma (Sus_G) and diabetic retinopathy (Sus_R).

Figure 7.. Performance Evolution and Training Metrics of netAlexNet, netResNet50, and GoogleNet Models.

Figure 8.. Grad-Cam for AlexNet, ResNet50 & GoogLeNet convolutional neural networks for the eye disease detection.

Glossary: Non-disease (Non_D), glaucoma (Sus_G) and diabetic retinopathy (Sus_R).