Hypermixed Convolutional Neural Network for Retinal Vein Occlusion Classification

Abstract

Retinal vein occlusion (RVO) is one of the most common retinal vascular diseases leading to vision loss if not diagnosed and treated in time. RVO can be classified into two types: CRVO (blockage of the main retinal veins) and BRVO (blockage of one of the smaller branch veins). Automated diagnosis of RVO can improve clinical workflow and optimize treatment strategies. However, to the best of our knowledge, there are few reported methods for automated identification of different RVO types. In this study, we propose a new hypermixed convolutional neural network (CNN) model, namely, the VGG-CAM network, that can classify the two types of RVOs based on retinal fundus images and detect lesion areas using an unsupervised learning method. The image data used in this study is collected and labeled by three senior ophthalmologists in Shanxi Eye Hospital, China. The proposed network is validated to accurately classify RVO diseases and detect lesions. It can potentially assist in further investigating the association between RVO and brain vascular diseases and evaluating the optimal treatments for RVO.

Figure 1

(a) Normal retina images, (b) CRVO retina images, and (c) BRVO retina images.

Figure 2

A simple CNN framework containing input, convolutional, pooling (subsampling), and fully connected layers (Heung-II, [36]).

Figure 3

Framework of VGG-CAM model for RVO classification and lesion detection.

Figure 4

Examples of average pooling and GAP.

Figure 5

RVO lesion detection by CAM.

Figure 6

Input images before and after preprocessing.

Figure 7

Lesion detection by VGG-CAM network on problematic areas of BRVO and CRVO.

Figure 8

Confusion matrix of VGG-CAM network on validation set.

Figure 9

ROC curve of VGG-CAM network performance on RVO classification.