Identification of Diabetic Retinopathy Using Weighted Fusion Deep Learning Based on Dual-Channel Fundus Scans

Abstract

It is a well-known fact that diabetic retinopathy (DR) is one of the most common causes of visual impairment between the ages of 25 and 74 around the globe. Diabetes is caused by persistently high blood glucose levels, which leads to blood vessel aggravations and vision loss. Early diagnosis can minimise the risk of proliferated diabetic retinopathy, which is the advanced level of this disease, and having higher risk of severe impairment. Therefore, it becomes important to classify DR stages. To this effect, this paper presents a weighted fusion deep learning network (WFDLN) to automatically extract features and classify DR stages from fundus scans. The proposed framework aims to treat the issue of low quality and identify retinopathy symptoms in fundus images. Two channels of fundus images, namely, the contrast-limited adaptive histogram equalization (CLAHE) fundus images and the contrast-enhanced canny edge detection (CECED) fundus images are processed by WFDLN. Fundus-related features of CLAHE images are extracted by fine-tuned Inception V3, whereas the features of CECED fundus images are extracted using fine-tuned VGG-16. Both channels’ outputs are merged in a weighted approach, and softmax classification is used to determine the final recognition result. Experimental results show that the proposed network can identify the DR stages with high accuracy. The proposed method tested on the Messidor dataset reports an accuracy level of 98.5%, sensitivity of 98.9%, and specificity of 98.0%, whereas on the Kaggle dataset, the proposed model reports an accuracy level of 98.0%, sensitivity of 98.7%, and specificity of 97.8%. Compared with other models, our proposed network achieves comparable performance.

Deep learning for automatic diagnosis of fundus diseases on optical coherence tomography images

“…A large amount of data often leads to higher accuracy 21 . Previous researchers have often used datasets from open-source web platforms to train models 22,23 . This study utilized the opensource dataset provided on Kaggle to obtain a large amount of image data and achieved an accuracy rate of 96% in test 1.…”

Deep learning in the grading of diabetic retinopathy: A review

“…The authors of Ref. [24,25,29,34,36] implement Inception V3 while those of Ref. [33,40] used Inception V4.…”

“…Some of the more commonly used are stochastic gradient descent which is applied by Ref. [5, 34], Adaptive Moment Estimation Algorithm (ADAM) [33] and RMSprop [22]. Shankar [44] mentioned that the Bayesian optimization model can be used to tune a model as it analyses the previous validation outcome in which it utilises to create a probabilistic model, which will map the hyperparameters to a probability score.…”

“…The highest accuracy is by Nneji et al. [34] with 98.50% when the features are extracted using Inception V3. There are 3 models that achieved a specificity of 98% which is Nneji et al.…”

COVID-19 Pneumonia Classification Based on NeuroWavelet Capsule Network

“…Both RT-PCR tests and CT scans are relatively expensive [ 11 , 12 ] and quite a number of nations are mandated to conduct limited testing for only risk-prone populations due to excessive demand. CXR imaging is a comparably low-cost method of detecting lung infections, and it can also be used to detect COVID-19 [ 13 , 14 , 15 , 16 , 17 ]. As a result, automated algorithms that can accurately classify COVID-19 from CXR exams are extremely useful in the fight against the pandemic.…”