Diabetic Retinal Grading Using Attention-Based Bilinear Convolutional Neural Network and Complement Cross Entropy

Abstract

Diabetic retinopathy (DR) is a common complication of diabetes mellitus (DM), and it is necessary to diagnose DR in the early stages of treatment. With the rapid development of convolutional neural networks in the field of image processing, deep learning methods have achieved great success in the field of medical image processing. Various medical lesion detection systems have been proposed to detect fundus lesions. At present, in the image classification process of diabetic retinopathy, the fine-grained properties of the diseased image are ignored and most of the retinopathy image data sets have serious uneven distribution problems, which limits the ability of the network to predict the classification of lesions to a large extent. We propose a new non-homologous bilinear pooling convolutional neural network model and combine it with the attention mechanism to further improve the network’s ability to extract specific features of the image. The experimental results show that, compared with the most popular fundus image classification models, the network model we proposed can greatly improve the prediction accuracy of the network while maintaining computational efficiency.

DFTSA-Net: Deep Feature Transfer-Based Stacked Autoencoder Network for DME Diagnosis

“…However, the performance of conventional classifiers is highly impacted by the selection of hand-crafted features. On the other hand, deep neural networks require larger training datasets, extract features automatically, and generally provide a higher classification performance at a higher computational cost [15][16][17][18][19][20]. The classical SVM classifier was used by Shengchun et al [21] for automatic hard exudate classification.…”

LuNet-LightGBM: An Effective Hybrid Approach for Lesion Segmentation and DR Grading

“…In addition, as compared to neural network models, the majority of these conventional techniques display poor generalization [17,18]. This limits the scope of these engineering methods' usefulness in a therapeutic setting.…”

Classification of Diabetic Retinopathy Severity Based on GCA Attention Mechanism

“…This requires the network to introduce a proper attention mechanism, which makes the model adaptively enhance the perception of useful information. However, the previous researches on this issue are relatively few; Runze Fan et al [11] combined an attention model in the feature fusion stage of DR classification model to adaptively update the weights of each feature block, and Liu et al [12] At the stage of DR detection using traditional machine learning techniques, researchers need to have some medical background and manual extraction of lesion features from the image dataset, eventually the extracted lesion features are fed into a classification model to complete the detection of DR. Nguyen et al [13] proposed a multilayer feedforward neural network with strong robustness for DR severity classification. Zhang et al [14] used support vector machine (SVM) to classify preprocessed bright non-lesion areas, exudates and cotton wool spots.…”

“…Fan et al [31] proposed a residual convolutional block attention model (RCAM) and used it in a multi-feature fusion technique with adaptive weights, which was combined with MobileNetV3 network for DR severity classification. Liu et al [12] considered DR severity classification as a fine-grained classification problem and proposed a compact bilinear pooling network model based on the attention mechanism for DR severity classification, which both improved the prediction accuracy and maintained the computational efficiency of the model. Ramasamy et al [32] extracted and fused ophthalmic features from retinal images, which are based on texture gray level features.…”