Multi-categorical deep learning neural network to classify retinal images: A pilot study employing small database

Abstract

Deep learning emerges as a powerful tool for analyzing medical images. Retinal disease detection by using computer-aided diagnosis from fundus image has emerged as a new method. We applied deep learning convolutional neural network by using MatConvNet for an automated detection of multiple retinal diseases with fundus photographs involved in STructured Analysis of the REtina (STARE) database. Dataset was built by expanding data on 10 categories, including normal retina and nine retinal diseases. The optimal outcomes were acquired by using a random forest transfer learning based on VGG-19 architecture. The classification results depended greatly on the number of categories. As the number of categories increased, the performance of deep learning models was diminished. When all 10 categories were included, we obtained results with an accuracy of 30.5%, relative classifier information (RCI) of 0.052, and Cohen's kappa of 0.224. Considering three integrated normal, background diabetic retinopathy, and dry age-related macular degeneration, the multi-categorical classifier showed accuracy of 72.8%, 0.283 RCI, and 0.577 kappa. In addition, several ensemble classifiers enhanced the multi-categorical classification performance. The transfer learning incorporated with ensemble classifier of clustering and voting approach presented the best performance with accuracy of 36.7%, 0.053 RCI, and 0.225 kappa in the 10 retinal diseases classification problem. First, due to the small size of datasets, the deep learning techniques in this study were ineffective to be applied in clinics where numerous patients suffering from various types of retinal disorders visit for diagnosis and treatment. Second, we found that the transfer learning incorporated with ensemble classifiers can improve the classification performance in order to detect multi-categorical retinal diseases. Further studies should confirm the effectiveness of algorithms with large datasets obtained from hospitals.

Fig 1. Illustration of the proposed procedure in this study.

Fig 2. Performance of deep learning methods with 5-fold cross validation according to the number of categories.

(A) the performance plot of accuracy (B) the performance plot of relative classifier information (C) the performance plot of Kappa. AMD, age-related macular degeneration; BDR, background diabetic retinopathy; PDR, proliferative diabetic retinopathy; RVO, retinal vein occlusion; RAO, retinal artery occlusion; VGG19-TL-RF, transfer learning with random forest based on VGG-19 structure; VGG19-TL-SVM, transfer learning with one-vs-one support vector machine based on VGG-19 structure.

Fig 3. Binary discriminative accuracy between retinal diseases using transfer learning with random forest based on VGG-19 structure.

The number of each pair shows the accuracy of binary classifiers.

Fig 4. Receiver operating characteristic (ROC) curves of transfer learning with random forest based on VGG-19 structure (VGG19-TL-RF), transfer learning with random forest based on VGG-19 structure (VGG19-TL-SVM), and VGG-19, and AlexNet in predicting normal retina or retinal disease status using fundus photographs.

We divided all data set (10,000 images) into training dataset (70%) and test dataset (30%). Retinal disease status includes diabetic retinopathy, age-related macular degeneration, retinal vein occlusion, retinal artery occlusion, hypertensive retinopathy, Coat’s disease, and retinitis.

Fig 5. Comparison of different feature selection methods for 10 multi-categorical retinal image classification problem.

KW, Kruskal-Walis one-way ANOVA; BW, ratio of features between-categories to within-category sum of squares; MRMD, Max-Relevance-Max-Distance.