Diabetic Retinopathy Detection and Classification from Fundus Images Using Deep Learning

Diabetic retinopathy is an ocular disorder that has the potential to result in visual impairment and complete loss of vision in those diagnosed with diabetes. This illness affects the retinal blood vessels inside the light-sensitive tissue layer at the posterior of the eye, known as the retina. This paper presents a complete approach to diagnosing and categorizing diabetic retinopathy using deep learning models. A lightweight Convolutional Neural Network (CNN) is used to detect diabetic retinopathy in fundus images. This CNN has been developed to have fewer parameters and calculations, making it suited for resource-constrained environments while retaining decent performance. The categorization of diabetic retinopathy is carried out with the help of EfficientNet. This model uses an innovative compound scaling approach to strike a balance between the model's depth, width, and resolution. As a result, it maximizes computing efficiency while preserving high accuracy. The proposed detection model obtained an accuracy of 95%, and the classification model produced an accuracy of 84%.

Diabetic retinopathy (DR) is a complication of diabetes that cause retinal damage; therefore, it is a leading cause of blindness. However, early detection of this disease can dramatically reduce the risk of vision loss. The main problem of early DR detection is that the manual diagnosis by ophthalmology is time-consuming, expensive, and prone to misdiagnosis. Deep learning (DL) models have aided in the early diagnosis of DR, and DL is now frequently utilized in DR detection and classification. The main issues with classical DL models is that they are incapable to quantify the uncertainty in the models, thus they are prone to make wrong decisions in complex cases. However, Bayesian deep learning (BDL) models have recently evolved as unified probabilistic framework to integrate DL and Bayesian models to provides an accurate framework to identify all sources of uncertainty in the model. This paper introduces BDL and most recent research that used BDL approaches to treat diabetic retinopathy are reviewed and discussed. A thorough comparison of the existing Bayesian approaches in this topic is also presented. In addition, available datasets for the fundus retina, which is often employed in DR, are provided and reviewed.

ABSTRACTThe early detection of diabetic retinopathy (DR) illnesses improves diagnosis and lowers the risk of permanent blindness. As a result, screening for DR in fundus images is an important method used to diagnose diabetes and other eye diseases. However, detecting diseases manually requires a significant amount of time and work. Deep learning (DL) techniques have produced encouraging results in categorizing fundus images. Still, the multi‐class DR disease remains a difficult task. Thus, the proposed framework adopted a novel Criss‐Cross Attention‐Based Squeeze‐and‐Excitation Assisted Vision Transformer (CCA‐SE‐ViT) model to classify DR from fundus images. Initially, the defective region of the retina is segmented using a novel dilated depth‐wise separable convolutional U‐Net model (dDSC‐UNet). Then, using the segmented regions, the fundus images are classified into multiple classes as age‐related macular degeneration (AMD), DR, glaucoma, cataracts, myopia, hypertension, normal, other abnormalities, and DR cases are classified into no DR, mild, moderate, severe, and proliferative DR, respectively. The retinal fundus images are obtained from publicly available datasets like OIA‐ODIR and APTOS 2019. The proposed methodology for multi‐class categorization of retinal illnesses in the OIA‐ODIR dataset yielded 97.2% accuracy, 96.7% precision, 96.1% recall, 95.9% F1‐score, and 96.4% specificity. The APTOS dataset was used for multi‐class classification of DR illnesses, and the results were 99.68% accuracy, 99.08% precision, 99.31% recall, 99.19% F1‐score, and 99.26% specificity. The results demonstrated that the proposed method accurately identifies DR using retinal fundus images.

This article demonstrates a revolutionary deep- learning approach for automated diabetic retinopathy (DR) detection using convolutional neural networks (CNNs). With the help of datasets from the Diabetic Retinopathy Detection 2015 and APTOS 2019 Blindness Detection competitions, our model attempts to accurately classify retinal images into two categories: No DR and Proliferative DR, based on the severity of macular degeneration. We enhance the quality of the dataset and the robustness of the model by carefully preprocessing and augmenting it using various approaches like rotation, flipping, shearing, rescaling, translation, brightness correction, and normalization. We train two models: a regression model and a deep learning model. With an 88% validation accuracy, we demonstrated that our deep learning technique performs better than the regression model. In order to properly determine and classify DR severity levels, this study presents a robust and noise-tolerant system that evolves the field of automated DR detection. Our approach enables early detection and intervention by combining cutting-edge CNN architectures with sophisticated preprocessing techniques, which may lessen the strain of DR- related blindness on healthcare systems. We also demonstrate out drawbacks and suggest areas for additional investigation, such as analyzing ensemble learning and using a variety of datasets to improve the generalizability of the model. As we wrap up, our study emphasizes the potential of deep learning techniques to transform DR diagnosis and treatment, with the goal of enhancing the quality of care for diabetics globally.

Retinopathy is an eye disease caused by diabetes, and early detection and treatment can potentially reduce the risk of blindness in diabetic retinopathy sufferers. Using retinal Fundus images, diabetic retinopathy can be diagnosed, recognized, and treated. In the current state of the art, sensitivity and specificity are lacking. However, there are still a number of problems to be solved in state-of-the-art techniques like performance, accuracy, and being able to identify DR disease effectively with greater accuracy. In this paper, we have developed a new approach based on a combination of image processing and artificial intelligence that will meet the performance criteria for the detection of disease-causing diabetes retinopathy in Fundus images. Automatic detection of diabetic retinopathy has been proposed and has been carried out in several stages. The analysis was carried out in MATLAB using software-based simulation, and the results were then compared with those of expert ophthalmologists to verify their accuracy. Different types of diabetic retinopathy are represented in the experimental evaluation, including exudates, micro-aneurysms, and retinal hemorrhages. The detection accuracies shown by the experiments are greater than 98.80 percent.

Using Deep Learning Models to Characterize Major Retinal Features on Color Fundus Photographs

Diabetic retinopathy (DR) is a serious complication of diabetes that can result in vision loss if untreated, often progressing silently without warning symptoms. Elevated blood glucose levels damage the retina's microvasculature, initiating the condition. Early detection through retinal fundus imaging, supported by timely analysis and treatment, is critical for managing DR effectively. However, manually inspecting these images is a labour-intensive and time-consuming process, making computer-aided diagnosis (CAD) systems invaluable in supporting ophthalmologists. This research introduces the Fundus Imaging Diabetic Retinopathy Classification using Deep Learning and Fennec Fox Optimization (FIDRC-DLFFO) model, which automates the identification and classification of DR. The model integrates several advanced techniques to enhance performance and accuracy.1.The proposed FIDRC-DLFFO model automates DR detection and classification by combining median filtering for noise reduction, Inception-ResNet-v2 for feature extraction, and a gated recurrent unit (GRU) for classification.2.Fennec Fox Optimization (FFO) fine-tunes the GRU hyperparameters, boosting classification accuracy, with its effectiveness demonstrated on benchmark datasets.3.The results provide insights into the model's effectiveness and potential for real-world application.

In recent times, Internet of Things (IoT) and Deep Learning (DL) models have revolutionized the diagnostic procedures of Diabetic Retinopathy (DR) in its early stages that can save the patient from vision loss. At the same time, the recent advancements made in Machine Learning (ML) and DL models help in developing Computer Aided Diagnosis (CAD) models for DR recognition and grading. In this background, the current research works designs and develops an IoT-enabled Effective Neutrosophic based Segmentation with Optimal Deep Belief Network (ODBN) model i.e., NS-ODBN model for diagnosis of DR. The presented model involves Interval Neutrosophic Set (INS) technique to distinguish the diseased areas in fundus image. In addition, three feature extraction techniques such as histogram features, texture features, and wavelet features are used in this study. Besides, Optimal Deep Belief Network (ODBN) model is utilized as a classification model for DR. ODBN model involves Shuffled Shepherd Optimization (SSO) algorithm to regulate the hyperparameters of DBN technique in an optimal manner. The utilization of SSO algorithm in DBN model helps in increasing the detection performance of the model significantly. The presented technique was experimentally evaluated using benchmark DR dataset and the results were validated under different evaluation metrics. The resultant values infer that the proposed INS-ODBN technique is a promising candidate than other existing techniques.

Deep learning is currently the state-of-the-art for automated detection of referable diabetic retinopathy (DR) from color fundus photographs (CFP). While the general interest is put on improving results through methodological innovations, it is not clear how good these approaches perform compared to standard deep classification models trained with the appropriate settings. In this paper we propose to model a strong baseline for this task based on a simple and standard ResNet-18 architecture. To this end, we built on top of prior art by training the model with a standard preprocessing strategy but using images from several public sources and an empirically calibrated data augmentation setting. To evaluate its performance, we covered multiple clinically relevant perspectives, including image and patient level DR screening, discriminating responses by input quality and DR grade, assessing model uncertainties and analyzing its results in a qualitative manner. With no other methodological innovation than a carefully designed training, our ResNet model achieved an AUC = 0.955 (0.953 - 0.956) on a combined test set of 61007 test images from different public datasets, which is in line or even better than what other more complex deep learning models reported in the literature. Similar AUC values were obtained in 480 images from two separate in-house databases specially prepared for this study, which emphasizes its generalization ability. This confirms that standard networks can still be strong baselines for this task if properly trained.

Diabetic Retinopathy (DR) is a major cause of vision loss globally, necessitating early and accurate diagnosis. This paper evaluates the effectiveness of various deep learning models such as InceptionV3, DenseNet121, DenseNet169, DenseNet201, and ResNet50 in classifying DR stages using retinal fundus images and the efficacy of a custom CNN binary classification model for DR detection. Two publicly available datasets, APTOS 2019 and DR Detection 2015, were used. The study explores the impact of dataset balance, volume, and hyperparameter optimization on model performance. InceptionV3 and DenseNet201 emerged as top performers among the pretrained models used for multi-class classification, with InceptionV3 achieving 75% accuracy on APTOS and DenseNet201 matching it. Meanwhile, the custom CNN binary classification model achieved an accuracy of 91.45%. The findings demonstrate the feasibility of using transfer learning for automated DR stage screening and the efficacy of Convolutional Neural Networks in identifying Diabetic Retinopathy while highlighting the significance of dataset quality and class balance.

To develop a deep learning (DL) model to detect morphologic patterns of diabetic macular edema (DME) based on optical coherence tomography (OCT) images. In the training set, 12,365 OCT images were extracted from a public data set and an ophthalmic center. A total of 656 OCT images were extracted from another ophthalmic center for external validation. The presence or absence of three OCT patterns of DME, including diffused retinal thickening, cystoid macular edema, and serous retinal detachment, was labeled with 1 or 0, respectively. A DL model was trained to detect three OCT patterns of DME. The occlusion test was applied for the visualization of the DL model. Applying 5-fold cross-validation method in internal validation, the area under the receiver operating characteristic curve for the detection of three OCT patterns (i.e., diffused retinal thickening, cystoid macular edema, and serous retinal detachment) was 0.971, 0.974, and 0.994, respectively, with an accuracy of 93.0%, 95.1%, and 98.8%, respectively, a sensitivity of 93.5%, 94.5%, and 96.7%, respectively, and a specificity of 92.3%, 95.6%, and 99.3%, respectively. In external validation, the area under the receiver operating characteristic curve was 0.970, 0.997, and 0.997, respectively, with an accuracy of 90.2%, 95.4%, and 95.9%, respectively, a sensitivity of 80.1%, 93.4%, and 94.9%, respectively, and a specificity of 97.6%, 97.2%, and 96.5%, respectively. The occlusion test showed that the DL model could successfully identify the pathologic regions most critical for detection. Our DL model demonstrated high accuracy and transparency in the detection of OCT patterns of DME. These results emphasized the potential of artificial intelligence in assisting clinical decision-making processes in patients with DME.

In the recent era, the increasing diabetic ratio effects in producing complex retinal diseases. The early diagnosis will avoid the consequences of diabetic-related complications. So, there is an urge to develop a computerized model for identifying the retinopathy features caused due to diabetes while evading the false positives. The aim of this paper is to identify and segment the retinal features like blood vessels, optic disc in diabetic retinopathy (DR) fundus images in order to remove the non-diabetic retinopathy features which makes the detection of DR features (lesions) easier. And this is done by using a hybrid segmentation algorithm called BINI which combines the model of both binary and Niblack’s thresholding. Initially, the input image is standardized by resizing the image using bi-cubic interpolation area method. Then, the fundus image quality is enhanced using preprocessing techniques like green channel extraction, intensity channel extraction, median filtering, contrast limited adaptive histogram equalization and morphological operations. The preprocessed images are segmented to produce retinal features using the BINI algorithm. It is a novel method which hybrids both binary thresholding and Niblack’s thresholding. The performance of the segmentation methods is evaluated using the validation measures like Rand index (Ri) and Jaccard index (Ji), Precision (Pr), Recall (Rc) and F-Measure (Fm). The proposed method for segmenting the retinal features using BINI thresholding has given an accuracy of about 96.48%; it leads to an accuracy of 100% clear segmentation of the lesions of diabetic retinopathy images.

PurposeTo conduct a head-to-head comparison between deep learning (DL) and radiomics models across institutions for predicting microvascular invasion (MVI) in hepatocellular carcinoma (HCC) and to investigate the model robustness and generalizability through rigorous internal and external validation.MethodsThis retrospective study included 2304 preoperative images of 576 HCC lesions from two centers, with MVI status determined by postoperative histopathology. We developed DL and radiomics models for predicting the presence of MVI using B-mode ultrasound, contrast-enhanced ultrasound (CEUS) at the arterial, portal, and delayed phases, and a combined modality (B + CEUS). For radiomics, we constructed models with enlarged vs. original regions of interest (ROIs). A cross-validation approach was performed by training models on one center’s dataset and validating the other, and vice versa. This allowed assessment of the validity of different ultrasound modalities and the cross-center robustness of the models. The optimal model combined with alpha-fetoprotein (AFP) was also validated. The head-to-head comparison was based on the area under the receiver operating characteristic curve (AUC).ResultsThirteen DL models and 25 radiomics models using different ultrasound modalities were constructed and compared. B + CEUS was the optimal modality for both DL and radiomics models. The DL model achieved AUCs of 0.802–0.818 internally and 0.667–0.688 externally across the two centers, whereas radiomics achieved AUCs of 0.749–0.869 internally and 0.646–0.697 externally. The radiomics models showed overall improvement with enlarged ROIs (P < 0.05 for both CEUS and B + CEUS modalities). The DL models showed good cross-institutional robustness (P > 0.05 for all modalities, 1.6–2.1% differences in AUC for the optimal modality), whereas the radiomics models had relatively limited robustness across the two centers (12% drop-off in AUC for the optimal modality). Adding AFP improved the DL models (P < 0.05 externally) and well maintained the robustness, but did not benefit the radiomics model (P > 0.05).ConclusionCross-institutional validation indicated that DL demonstrated better robustness than radiomics for preoperative MVI prediction in patients with HCC, representing a promising solution to non-standardized ultrasound examination procedures.

The primary source of vision loss in patients is mainly due to Diabetic retinopathy (DR), caused due to diabetes mellitus. It has become a significant reason for visual impairment among people within 25-74 years of age. If timely medical attention is provided to DR patients, over 90% of people can be saved from vision loss. It's crucial for the early diagnosis of the disease and provide the necessary treatment. The symptoms are more prevalent in type 2 diabetics than associated with type 1 diabetics. Unlike computer-aided diagnosis systems, the traditional procedures of DR detection using fundus photography are both time and cost-consuming. Among the numerous methods for screening and detecting DR, Convolutional Neural Networks are considered extensively in Deep Learning (DL) methods. This review article illustrates the different datasets, pre-processing steps, and DL techniques used in the fundus images for efficient DR detection at an early stage. The main motive of this review article is to provide the research community with an insight into the various pre-processing steps, Public datasets, DL models in DR detection, and some future research directions in this field.

Deep Learning Model to Detect Diabetic Retinopathy in 45° Images Using Ground Truth from Ultra-Widefield Imaging.