Retinopathy Grading Research Articles

Retinal image quality assessment in diabetic-retinopathy screening: Real world evidence from a lower-middle income country

Objectives: Sight loss due to diabetic retinopathy (DR) is preventable by early detection and treatment. Digital retinal imaging is the most widely practiced method of screening of DR. Poor quality of images is a major hinder to implement systematic DR screening using digital retinal imaging in low- and middle-income countries with a high prevalence of lens opacities. We aimed to identify the proportion of ungradable images using hand-held retinal imaging and predictors of image gradability in a DR screening feasibility study conducted in Sri Lanka. Material and Methods: The present study is a retrospective data analysis of a validation study conducted to assess the diagnostic test accuracy of a held-held digital retinal imaging model integrated into a tertiary level medical clinic. Two selected physician graders underwent formal training to assess retinal image quality using a “four-quadrant method of assessing gradability.” The procedure was a subjective image quality assessment performed by the physician graders manually, that is, images with more than 50% of the field with poor clarity and not suitable for retinopathy grading were classified as ungradable. Two-field (Field-1: macula centered, Field-2: disc centered) non-mydriatic and mydriatic retinal images were captured (Type of camera: Zeiss-Visuscout 100®, Germany) in a consecutive sample of people with diabetes attended for routine medical care and assessed for quality and graded by two independent physician graders on-site. The reference test was a mydriatic biomicroscopic examination conducted at a separate eye clinic by an experienced specialist retinologist. Mixed model regression analysis was conducted to assess the predictors of gradability. Results: A total of 700 individuals (5508 gradability data points) were included in the study. The proportion of ungradable images in non-mydriatic imaging was 30% for Grader 1 and 24% for Grader 2 and went down to 7% and 5%, respectively, for Grader 1 and 2 after dilating the pupils using mydriatic agents. Non-mydriatic images had almost 20 times higher odds (odds ratio [OR] 19.7, 95% confidence interval [CI] 15.1–25.8) of being rated as ungradable compared to mydriatic images. With the increase of each year in age, the odds of having ungradable digital retinal images in a patient increased by 7% (OR 1.07, 95% CI 1.05–1.09). For visual acuity increase in each level of the Log-MAR scale, the odds of having ungradable images increased by 40% (OR 1.40, 95% CI 1.30–1.51). In lens opacity, 54% higher odds of ungradability were observed when present nuclear opalescence (OR 1.54, 95% CI 1.39–1.70) and posterior subcapsular opacity (OR 1.54, 95% CI 1.24–1.92). Conclusion: Non-mydriatic methods may not be suitable as a primary DR screening strategy in countries with a high prevalence of cataracts. Increasing age, poor visual acuity, and the presence of lens opacity are factors that would affect image quality. The capacity to deliver services for managing cataracts may be an important determinant in achieving effective coverage of digital surveillance of DR in low- and middle-income countries.

Clinical characteristics and associated factors of diabetic retinopathy in Sudanese patients

Background: Diabetic retinopathy (DR) is a leading cause of vision loss and preventable blindness, particularly in middle- and high-income countries.Aim: This study aimed to assess the characteristics and associated factors of DR in Sudanese patients with type 2 diabetes mellitus (DM).Setting: A cross-sectional hospital-based study included 119 patients with type 2 DM was conducted at EL-Walidain Eye Hospital, Khartoum, Sudan, from February 2021 to May 2021.Methods: A non-mydriatic digital fundus camera was used for fundus photography, and Scottish Diabetic Retinopathy Grading Scheme was used for the final classification of DR.Results: Females comprised 53.8% of patients, mean age of 58.8 ± 8.5 years. Uncontrolled DM was found in 37.8% with a mean duration of 10.5 ± 4.8 years. Myopia was the most common refractive error found in 57.8% of the uncontrolled group, followed by emmetropia (17.8%). Whereas in the controlled group, 47.3% were hyperopic and 39.2% were myopic. Maculopathy grade M1 was found to be 37.8% in the uncontrolled group and 18.9% in the controlled group, with no significant difference (p = 0.361). Conversely, the difference in retinopathy between the two groups was statistically significant (p = 0.043).Conclusion: Decreased vision and increased retinopathy were associated with an increase in patients’ age. More widespread diabetes awareness and screening programmes to improve diabetes management and control in Sudan and other developing countries.Contribution: This study observed maculopathy grade M1 and myopia were more common in uncontrolled DM, and vision function was inversely correlated with age (p 0.05).

MediDRNet: Tackling category imbalance in diabetic retinopathy classification with dual-branch learning and prototypical contrastive learning

Background and Objective:The classification of diabetic retinopathy (DR) aims to utilize the implicit information in images for early diagnosis, to prevent and mitigate the further worsening of the condition. However, existing methods are often limited by the need to operate within large, annotated datasets to show significant advantages. Additionally, the number of samples for different categories within the dataset needs to be evenly distributed, because the characteristic of sample imbalance distribution can lead to an excessive focus on high-frequency disease categories, while neglecting the less common but equally important disease categories. Therefore, there is an urgent need to develop a new classification method that can effectively alleviate the issue of sample distribution imbalance, thereby enhancing the accuracy of diabetic retinopathy classification. Methods:In this work, we propose MediDRNet, a dual-branch network model based on prototypical contrastive learning. This model adopts prototype contrastive learning, creating prototypes for different levels of lesions, ensuring they represent the core features of each lesion level. It classifies by comparing the similarity between data points and their category prototypes. Our dual-branch network structure effectively resolves the issue of category imbalance and improves classification accuracy by emphasizing subtle differences in retinal lesions. Moreover, our approach combines a dual-branch network with specific lesion-level prototypes for core feature representation and incorporates the convolutional block attention module for enhanced lesion feature identification. Results:Our experiments using both the Kaggle and UWF classification datasets have demonstrated that MediDRNet exhibits exceptional performance compared to other advanced models in the industry, especially on the UWF DR classification dataset where it achieved state-of-the-art performance across all metrics. On the Kaggle DR classification dataset, it achieved the highest average classification accuracy (0.6327) and Macro-F1 score (0.6361). Particularly in the classification tasks for minority categories of diabetic retinopathy on the Kaggle dataset (Grades 1, 2, 3, and 4), the model reached high classification accuracies of 58.08%, 55.32%, 69.73%, and 90.21%, respectively. In the ablation study, the MediDRNet model proved to be more effective in feature extraction from diabetic retinal fundus images compared to other feature extraction methods. Conclusions:This study employed prototype contrastive learning and bidirectional branch learning strategies, successfully constructing a grading system for diabetic retinopathy lesions within imbalanced diabetic retinopathy datasets. Through a dual-branch network, the feature learning branch effectively facilitated a smooth transition of features from the grading network to the classification learning branch, accurately identifying minority sample categories. This method not only effectively resolved the issue of sample imbalance but also provided strong support for the precise grading and early diagnosis of diabetic retinopathy in clinical applications, showcasing exceptional performance in handling complex diabetic retinopathy datasets. Moreover, this research significantly improved the efficiency of prevention and management of disease progression in diabetic retinopathy patients within medical practice. We encourage the use and modification of our code, which is publicly accessible on GitHub: https://github.com/ReinforceLove/MediDRNet.

Cross-modality transfer learning with knowledge infusion for diabetic retinopathy grading.

Ultra-wide-field (UWF) fundus photography represents an emerging retinal imaging technique offering a broader field of view, thus enhancing its utility in screening and diagnosing various eye diseases, notably diabetic retinopathy (DR). However, the application of computer-aided diagnosis for DR using UWF images confronts two major challenges. The first challenge arises from the limited availability of labeled UWF data, making it daunting to train diagnostic models due to the high cost associated with manual annotation of medical images. Secondly, existing models' performance requires enhancement due to the absence of prior knowledge to guide the learning process. By leveraging extensively annotated datasets within the field, which encompass large-scale, high-quality color fundus image datasets annotated at either image-level or pixel-level, our objective is to transfer knowledge from these datasets to our target domain through unsupervised domain adaptation. Our approach presents a robust model for assessing the severity of diabetic retinopathy (DR) by leveraging unsupervised lesion-aware domain adaptation in ultra-wide-field (UWF) images. Furthermore, to harness the wealth of detailed annotations in publicly available color fundus image datasets, we integrate an adversarial lesion map generator. This generator supplements the grading model by incorporating auxiliary lesion information, drawing inspiration from the clinical methodology of evaluating DR severity by identifying and quantifying associated lesions. We conducted both quantitative and qualitative evaluations of our proposed method. In particular, among the six representative DR grading methods, our approach achieved an accuracy (ACC) of 68.18% and a precision (pre) of 67.43%. Additionally, we conducted extensive experiments in ablation studies to validate the effectiveness of each component of our proposed method. In conclusion, our method not only improves the accuracy of DR grading, but also enhances the interpretability of the results, providing clinicians with a reliable DR grading scheme.

Distribution shift detection for the postmarket surveillance of medical AI algorithms: a retrospective simulation study

Distribution shifts remain a problem for the safe application of regulated medical AI systems, and may impact their real-world performance if undetected. Postmarket shifts can occur for example if algorithms developed on data from various acquisition settings and a heterogeneous population are predominantly applied in hospitals with lower quality data acquisition or other centre-specific acquisition factors, or where some ethnicities are over-represented. Therefore, distribution shift detection could be important for monitoring AI-based medical products during postmarket surveillance. We implemented and evaluated three deep-learning based shift detection techniques (classifier-based, deep kernel, and multiple univariate kolmogorov-smirnov tests) on simulated shifts in a dataset of 130’486 retinal images. We trained a deep learning classifier for diabetic retinopathy grading. We then simulated population shifts by changing the prevalence of patients’ sex, ethnicity, and co-morbidities, and example acquisition shifts by changes in image quality. We observed classification subgroup performance disparities w.r.t. image quality, patient sex, ethnicity and co-morbidity presence. The sensitivity at detecting referable diabetic retinopathy ranged from 0.50 to 0.79 for different ethnicities. This motivates the need for detecting shifts after deployment. Classifier-based tests performed best overall, with perfect detection rates for quality and co-morbidity subgroup shifts at a sample size of 1000. It was the only method to detect shifts in patient sex, but required large sample sizes (>30′000\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$> 30^{\\prime} 000$$\\end{document}). All methods identified easier-to-detect out-of-distribution shifts with small (≤300) sample sizes. We conclude that effective tools exist for detecting clinically relevant distribution shifts. In particular classifier-based tests can be easily implemented components in the post-market surveillance strategy of medical device manufacturers.

Pupil cycle time and its significance in patients with diabetic retinopathy

Context: Prolonged pupil cycle time (PCT) is a complication of type II diabetes and becomes more protracted as the disease progresses. PCT can be used as a quick and accurate clinical test to determine how much autonomic neuropathy is influencing a diabetic patient’s pupil. Aims: The aim of the study was to compare the PCT of type II diabetes patients with that of the healthy control group and analyze the impact of several factors, including age, sex, duration of diabetes, and grade of diabetic retinopathy on PCT. Settings and Design: The study was conducted from August 2022 to August 2023. One hundred people were arbitrarily selected, 50 of whom had diabetes and 50 healthy controls. Subjects and Methods: To start rhythmic cycles of contraction and relaxation, a narrow, horizontally aligned beam of fixed proportions was focused on the inferior area of the pupil. The cycle period was then monitored in milliseconds using a handheld stopwatch. The mean PCT of diabetics was compared with nondiabetics and its correlation with different variables such as age, sex, diabetes duration, and grade of retinopathy was done. Results: Fifty diabetic patients (28 females and 22 males) aged 42–74 years and 50 controls (32 females and 18 males) aged 44–70 were included in the study. The mean PCT of the diabetics was 1130 ± 186 ms, whereas it was 940 ± 132 ms in the controls; thus, the mean PCT was longer in the diabetics. It was also found that the mean PCT increased with increasing grades of retinopathy. A positive correlation was also noted between mean PCT and duration of diabetes. Conclusions: PCT is a fast, simple, and reliable clinical test of optic nerve function. Being objective and quantifiable for each eye separately is a huge benefit. The PCT measurements might be used in the clinical evaluation of an autonomic nervous system. It is a simple, inexpensive yet quick, and effective tool to assess pupillary autonomic neuropathy in the diabetics.

SSiT: Saliency-Guided Self-Supervised Image Transformer for Diabetic Retinopathy Grading.

Self-supervised Learning (SSL) has been widely applied to learn image representations through exploiting unlabeled images. However, it has not been fully explored in the medical image analysis field. In this work, Saliency-guided Self-Supervised image Transformer (SSiT) is proposed for Diabetic Retinopathy (DR) grading from fundus images. We novelly introduce saliency maps into SSL, with a goal of guiding self-supervised pre-training with domain-specific prior knowledge. Specifically, two saliency-guided learning tasks are employed in SSiT: 1) Saliency-guided contrastive learning is conducted based on the momentum contrast, wherein fundus images' saliency maps are utilized to remove trivial patches from the input sequences of the momentum-updated key encoder. Thus, the key encoder is constrained to provide target representations focusing on salient regions, guiding the query encoder to capture salient features. 2) The query encoder is trained to predict the saliency segmentation, encouraging the preservation of fine-grained information in the learned representations. To assess our proposed method, four publicly-accessible fundus image datasets are adopted. One dataset is employed for pre-training, while the three others are used to evaluate the pre-trained models' performance on downstream DR grading. The proposed SSiT significantly outperforms other representative state-of-the-art SSL methods on all downstream datasets and under various evaluation settings. For example, SSiT achieves a Kappa score of 81.88% on the DDR dataset under fine-tuning evaluation, outperforming all other ViT-based SSL methods by at least 9.48%.

RELIABILITY OF CLINICAL GRADING OF DIABETIC RETINOPATHY COMPARED WITH GRADING OF ULTRA-WIDEFIELD IMAGES.

To evaluate the reliability of clinical grading of diabetic retinopathy (DR) severity compared with grading on ultra-widefield pseudocolor fundus (UWF-CF) and ultra-widefield fluorescein angiography (UWF-FA) images and their relative detection of sight-threatening DR and referable DR. A total of 184 diabetic eyes were analyzed. UWF-CF and UWF-FA images were graded based on the International Clinical Diabetic Retinopathy severity scale. Agreement between clinical and UWF-based severity grading was evaluated using Cohen's kappa coefficient. The rate of sight-threatening DR and referable DR was evaluated for each grading method. Moderate agreement was found between clinical grading and UWF-CF (k = 0.456, P < 0.001) and between UWF-CF and UWF-FA (k = 0.443, P < 0.001). The agreement between clinical grading and UWF-FA was fair (k = 0.397, P < 0.001). UWF-based grading identified a higher DR grade in 56 eyes (30%) on UWF-CF and 85 eyes (46.2%) on UWF-FA. Compared with clinical grading, UWF-FA detected a higher rate of sight-threatening DR (44%; 81/184 vs. 22.3%; 41/184), while UWF-CF detected more referable eyes (58.1%; 107/184 vs. 45.65%; 84/184). Ultra-widefield pseudocolor fundus is a valuable tool for identifying referable eyes and can be a useful, noninvasive adjunct to clinical grading. The results suggest that UWF-FA is particularly useful for detecting unsuspected sight-threatening DR in eyes with clinically referable DR.

Arithmetic Optimization for Coinciding Diabetic Retinopathy and Diabetic Macular Edema Grading based on Self-Attention Convolutional Neural Network

This paper proposes a Self-Attention Convolutional Neural Network (SACNN) optimized with Arithmetic Optimization Algorithm (AOA) for coinciding Diabetic Retinopathy (DR) and Diabetic Macular Edema Grading (DMEG) (SACNN-AOA-DR-DMEG). Initially, the input image is collected from 2 openly available benchmark datasets, namely Messidor and ISBI 2018 IDRiD. Then the input image is pre-processing using Altered Phase Preserving Dynamic Range Compression (APPDRC) for reducing noise from the imageries. SACNN receives the pre-processed imageries. The SACNN has three modules: (i) plane attention module, (ii) depth attention module, (iii) Attention Fusion Module. DR and DME features are extracted by plane attention module and depth attention module of SACNN. Attention Fusion Module receives extracted characteristics for categorizing and grading DR and DME disorders. SACNN does not adopt any optimization techniques to guarantee accurate DR and DME grading disorders. That’s why, Arithmetic Optimization Algorithm (AOA) is deemed to optimize the SACNN weight parameters. The proposed technique is implemented in Python. The proposed SACNN-AOA-DR-DMEG method provides 11.18%, 18.99% and 23.76% higher accuracy for diabetic retinopathy grading; 11.52%, 29.62% and 20.38% higher accuracy for DMEG; 33.39%, 22%, 39.26% lower computation time on Messidor data compared with the existing methods, such as AMGNN-DR-DMEG, LCNN-DR-DMEG, and FFN-DR-DMEG respectively.

Smart grading of diabetic retinopathy: an intelligent recommendation-based fine-tuned EfficientNetB0 framework.

Diabetic retinopathy is a condition that affects the retina and causes vision loss due to blood vessel destruction. The retina is the layer of the eye responsible for visual processing and nerve signaling. Diabetic retinopathy causes vision loss, floaters, and sometimes blindness; however, it often shows no warning signals in the early stages. Deep learning-based techniques have emerged as viable options for automated illness classification as large-scale medical imaging datasets have become more widely available. To adapt to medical image analysis tasks, transfer learning makes use of pre-trained models to extract high-level characteristics from natural images. In this research, an intelligent recommendation-based fine-tuned EfficientNetB0 model has been proposed for quick and precise assessment for the diagnosis of diabetic retinopathy from fundus images, which will help ophthalmologists in early diagnosis and detection. The proposed EfficientNetB0 model is compared with three transfer learning-based models, namely, ResNet152, VGG16, and DenseNet169. The experimental work is carried out using publicly available datasets from Kaggle consisting of 3,200 fundus images. Out of all the transfer learning models, the EfficientNetB0 model has outperformed with an accuracy of 0.91, followed by DenseNet169 with an accuracy of 0.90. In comparison to other approaches, the proposed intelligent recommendation-based fine-tuned EfficientNetB0 approach delivers state-of-the-art performance on the accuracy, recall, precision, and F1-score criteria. The system aims to assist ophthalmologists in early detection, potentially alleviating the burden on healthcare units.

CauDR: A causality-inspired domain generalization framework for fundus-based diabetic retinopathy grading

Diabetic retinopathy (DR) is the most common diabetic complication, which usually leads to retinal damage, vision loss, and even blindness. A computer-aided DR grading system has a significant impact on helping ophthalmologists with rapid screening and diagnosis. Recent advances in fundus photography have precipitated the development of novel retinal imaging cameras and their subsequent implementation in clinical practice. However, most deep learning-based algorithms for DR grading demonstrate limited generalization across domains. This inferior performance stems from variance in imaging protocols and devices inducing domain shifts. We posit that declining model performance between domains arises from learning spurious correlations in the data. Incorporating do-operations from causality analysis into model architectures may mitigate this issue and improve generalizability. Specifically, a novel universal structural causal model (SCM) was proposed to analyze spurious correlations in fundus imaging. Building on this, a causality-inspired diabetic retinopathy grading framework named CauDR was developed to eliminate spurious correlations and achieve more generalizable DR diagnostics. Furthermore, existing datasets were reorganized into 4DR benchmark for DG scenario. Results demonstrate the effectiveness and the state-of-the-art (SOTA) performance of CauDR. Diabetic retinopathy (DR) is the most common diabetic complication, which usually leads to retinal damage, vision loss, and even blindness. A computer-aided DR grading system has a significant impact on helping ophthalmologists with rapid screening and diagnosis. Recent advances in fundus photography have precipitated the development of novel retinal imaging cameras and their subsequent implementation in clinical practice. However, most deep learning-based algorithms for DR grading demonstrate limited generalization across domains. This inferior performance stems from variance in imaging protocols and devices inducing domain shifts. We posit that declining model performance between domains arises from learning spurious correlations in the data. Incorporating do-operations from causality analysis into model architectures may mitigate this issue and improve generalizability. Specifically, a novel universal structural causal model (SCM) was proposed to analyze spurious correlations in fundus imaging. Building on this, a causality-inspired diabetic retinopathy grading framework named CauDR was developed to eliminate spurious correlations and achieve more generalizable DR diagnostics. Furthermore, existing datasets were reorganized into 4DR benchmark for DG scenario. Results demonstrate the effectiveness and the state-of-the-art (SOTA) performance of CauDR.

Association of Mean Platelet Volume with Grades of Hypertensive Retinopathy.

Hypertension is the most prevalent non communicable disease causing deaths even before evidence of target organ damage. Platelet activation is associated with risk of vascular events. Identification of mean platelet volume (MPV) values in patients with hypertensive retinopathy (HR) can help in prognostication and risk stratification of possible complications. Patients presenting to the outpatient department over a period of 18 months underwent clinical examination, blood sampling, direct ophthalmoscopic examination, and two-dimensional echocardiography. Out of 104 subjects, there were a total of 25 females and 79 males. Patients with grade 1HR had a mean MPV value of 9.4, those with grade 2 HR had a mean MPV of 9.8, among people with grade 3 HR the mean MPV was 10.1, and those with grade 4 HR had a mean MPV of 10.9. In our study, we have seen that MPV increases with increasing severity of HR, thus implying that MPV probably plays a role in HR. Testing of MPV can be used to decide whether preventative treatments should be instituted to the prevent the severe complications of retinopathy and thus decrease the risk and occurrence of cardiovascular and cerebrovascular events.

Attention-based deep learning framework for automatic fundus image processing to aid in diabetic retinopathy grading

Background and objectiveEarly detection and grading of Diabetic Retinopathy (DR) is essential to determine an adequate treatment and prevent severe vision loss. However, the manual analysis of fundus images is time consuming and DR screening programs are challenged by the availability of human graders. Current automatic approaches for DR grading attempt the joint detection of all signs at the same time. However, the classification can be optimized if red lesions and bright lesions are independently processed since the task gets divided and simplified. Furthermore, clinicians would greatly benefit from explainable artificial intelligence (XAI) to support the automatic model predictions, especially when the type of lesion is specified. As a novelty, we propose an end-to-end deep learning framework for automatic DR grading (5 severity degrees) based on separating the attention of the dark structures from the bright structures of the retina. As the main contribution, this approach allowed us to generate independent interpretable attention maps for red lesions, such as microaneurysms and hemorrhages, and bright lesions, such as hard exudates, while using image-level labels only. MethodsOur approach is based on a novel attention mechanism which focuses separately on the dark and the bright structures of the retina by performing a previous image decomposition. This mechanism can be seen as a XAI approach which generates independent attention maps for red lesions and bright lesions. The framework includes an image quality assessment stage and deep learning-related techniques, such as data augmentation, transfer learning and fine-tuning. We used the architecture Xception as a feature extractor and the focal loss function to deal with data imbalance. ResultsThe Kaggle DR detection dataset was used for method development and validation. The proposed approach achieved 83.7 % accuracy and a Quadratic Weighted Kappa of 0.78 to classify DR among 5 severity degrees, which outperforms several state-of-the-art approaches. Nevertheless, the main result of this work is the generated attention maps, which reveal the pathological regions on the image distinguishing the red lesions and the bright lesions. These maps provide explainability to the model predictions. ConclusionsOur results suggest that our framework is effective to automatically grade DR. The separate attention approach has proven useful for optimizing the classification. On top of that, the obtained attention maps facilitate visual interpretation for clinicians. Therefore, the proposed method could be a diagnostic aid for the early detection and grading of DR.

Association of Inflammatory Markers and Diabetic Retinopathy

Abstract Introduction: Diabetic retinopathy is a micro-angiopathy, the pathogenesis of which is manifold. Chronic inflammation releasing various mediators has been proposed to cause micro-angiopathy in long-term diabetics. These markers can be easily measured in serum and may serve as a risk factor for progression as well as allow newer modalities of management aimed at this pathology. Aims and Methodology: The aim of the study was to determine the association between different inflammatory markers and diabetic retinopathy. A total of 100 eyes in diabetic patients, 20 each with no retinopathy, mild NPDR, moderate NPDR, severe NPDR, and PDR were included. The serum levels of C-reactive protein (CRP) and absolute neutrophil count were recorded and compared across each of these groups. Results: The mean ANC among these groups, respectively, was 5525, 5162, 5374, 5239, and 5313 (per microlitre), and neither were these raised significantly nor was the difference between groups statistically significant. Mean CRP levels were within normal limits in all groups (2.56, 2.87, 4.06, 5.15, 6.97 mg/dl), although there was a steady increase with each grade of retinopathy, and this was found statistically significant. Conclusion: There was no association between absolute neutrophil count and diabetic retinopathy. Serum CRP levels may guide progression of retinopathy and can be monitored to help management protocols.

Leveraging Convolutional Neural Networks for Automated Detection and Grading of Diabetic Retinopathy from Fundus Images

This study addresses the critical challenge of Diabetic Retinopathy (DR) detection and severity grading, aiming to advance the field of medical image analysis. The research problem focuses on the need for an accurate and efficient model to discern DR conditions, thereby facilitating early diagnosis and intervention. Employing a Convolutional Neural Network (CNN), our methodology is developed to strike a balance between precision and computational efficiency, a pivotal aspect in the context of healthcare applications. The research leverages the APTOS 2019 dataset, a comprehensive collection of fundus photographs, to evaluate the efficacy of our proposed model. The dataset allows for a thorough investigation into the model's performance in binary-class and multi-class classifications, providing a robust foundation for analysis. The most important result of our study manifests in the achieved accuracy rates of 98.67% and 87.81% for binary-class and multi-class classifications, respectively. These outcomes underscore the model's reliability and innovation, surpassing established machine learning algorithms and affirming its potential as a valuable tool for early DR detection and severity assessment. In conclusion, the study marks a significant advancement in leveraging deep learning for ophthalmic diagnoses, particularly in the nuanced landscape of Diabetic Retinopathy. The implications of our findings extend to the broader realm of AI-driven healthcare solutions, presenting opportunities for enhanced clinical practices and early intervention strategies. Future research endeavors could explore further refinements to the model, considering additional datasets and collaborating with healthcare professionals for real-world validation, ensuring the continued progress of AI applications in the medical domain.

A graph convolutional network with dynamic weight fusion of multi-scale local features for diabetic retinopathy grading

Diabetic retinopathy (DR) is a serious ocular complication that can pose a serious risk to a patient’s vision and overall health. Currently, the automatic grading of DR is mainly using deep learning techniques. However, the lesion information in DR images is complex, variable in shape and size, and randomly distributed in the images, which leads to some shortcomings of the current research methods, i.e., it is difficult to effectively extract the information of these various features, and it is difficult to establish the connection between the lesion information in different regions. To address these shortcomings, we design a multi-scale dynamic fusion (MSDF) module and combine it with graph convolution operations to propose a multi-scale dynamic graph convolutional network (MDGNet) in this paper. MDGNet firstly uses convolution kernels with different sizes to extract features with different shapes and sizes in the lesion regions, and then automatically learns the corresponding weights for feature fusion according to the contribution of different features to model grading. Finally, the graph convolution operation is used to link the lesion features in different regions. As a result, our proposed method can effectively combine local and global features, which is beneficial for the correct DR grading. We evaluate the effectiveness of method on two publicly available datasets, namely APTOS and DDR. Extensive experiments demonstrate that our proposed MDGNet achieves the best grading results on APTOS and DDR, and is more accurate and diverse for the extraction of lesion information.

A novel approach for intelligent diagnosis and grading of diabetic retinopathy

Diabetic retinopathy (DR) is a severe ocular complication of diabetes that can lead to vision damage and even blindness. Currently, traditional deep convolutional neural networks (CNNs) used for DR grading tasks face two primary challenges: (1) insensitivity to minority classes due to imbalanced data distribution, and (2) neglecting the relationship between the left and right eyes by utilizing the fundus image of only one eye for training without differentiating between them. To tackle these challenges, we proposed the DRGCNN (DR Grading CNN) model. To solve the problem caused by imbalanced data distribution, our model adopts a more balanced strategy by allocating an equal number of channels to feature maps representing various DR categories. Furthermore, we introduce a CAM-EfficientNetV2-M encoder dedicated to encoding input retinal fundus images for feature vector generation. The number of parameters of our encoder is 52.88 M, which is less than RegNet_y_16gf (80.57 M) and EfficientNetB7 (63.79 M), but the corresponding kappa value is higher. Additionally, in order to take advantage of the binocular relationship, we input fundus retinal images from both eyes of the patient into the network for features fusion during the training phase. We achieved a kappa value of 86.62% on the EyePACS dataset and 86.16% on the Messidor-2 dataset. Experimental results on these representative datasets for diabetic retinopathy (DR) demonstrate the exceptional performance of our DRGCNN model, establishing it as a highly competitive intelligent classification model in the field of DR. The code is available for use at https://github.com/Fat-Hai/DRGCNN.

Graph-based multi-level feature fusion network for diabetic retinopathy grading using ultra-wide-field images

Diabetic Retinopathy (DR) grading plays an important role for early intervention in managing the disease. However, conventional grading methods predominantly rely on traditional color fundus photography, which may not fully capture crucial pathological information. In contrast, Ultra-wide-field (UWF) represents a novel non-contact and non-invasive modality, capable of generating images with an unprecedented field of view, ranging from 180∘ to 200∘. This offers distinct advantages over conventional color fundus photography for DR grading. Furthermore, given the subtle distinctions in imaging characteristics among adjacent levels of DR, the grading task becomes exceptionally challenging. Therefore, we propose a new multi-level feature fusion network based on the graph convolution to accomplish the DR grading task using UWF images. Specifically, we first use the progressive multi-level constraint module to iteratively refine features by imposing constraints on model features at different levels. This process serves to minimize the inter-individual distances within the intra-class feature space, ultimately reducing the overlap of feature spaces between adjacent classes. Subsequently, since features at different network levels exhibit distinct semantics and distributions, conventional aggregation techniques, such as splicing and summing, may introduce redundant information and noise interference. To address this issue, we introduce a graph-based feature fusion module, capable of capturing spatial relationships and effectively fusing multi-layered features. Experimental results evaluated on an in-house UWF dataset with 1234 images and a publicly available fundus dataset with 3662 images illustrate that the proposed approach outperforms the state-of-the-art methods and shows promise for DR grading, with a respective accuracy of 0.81 and 0.84.

Dual branch deep learning network for detection and stage grading of diabetic retinopathy

Diabetic retinopathy is a severe complication of diabetes that can lead to permanent blindness if not treated promptly. Early and accurate diagnosis of the disease is essential for successful treatment. This paper introduces a deep learning method for the detection and stage grading of diabetic retinopathy, using a single fundus retinal image. Our model utilizes transfer learning, employing two state-of-the-art pre-trained models as feature extractors and fine-tuning them on a new dataset. The proposed model is trained on a large multi-center dataset, including the APTOS 2019 dataset, obtained from publicly available sources. It achieves remarkable performance in diabetic retinopathy detection and stage classification on the APTOS 2019, outperforming the established literature. For binary classification, the proposed approach achieves an accuracy of 98.50%, a sensitivity of 99.46%, and a specificity of 97.51%. In stage grading, it achieves a quadratic weighted kappa of 93.00%, an accuracy of 89.60%, a sensitivity of 89.60%, and a specificity of 97.72%. The proposed approach serves as a reliable screening and stage grading tool for diabetic retinopathy, offering significant potential to enhance clinical decision-making and patient care.

Handling Imbalance and Limited Data in Thyroid Ultrasound and Diabetic Retinopathy Datasets Using Discrete Levy Flights Grey Wolf Optimizer Based Random Forest for Robust Medical Data Classification

In the field of disease diagnosis, medical image classification faces an inherent challenge due to various factors involving data imbalance, image quality variability, annotation variability, and limited data availability and data representativeness. Such challenges affect the algorithm's classification ability on the medical images in an adverse way, which leads to biased model outcomes and inaccurate interpretations. In this paper, a novel Discrete Levy Flight Grey Wolf Optimizer (DLFGWO) is combined with the Random Forest (RF) classifier to address the above limitations on the biomedical datasets and to achieve better classification rate. The DLFGWO-RF resolves the image quality variability in ultrasound images and limits the inaccuracies on classification using RF by handling the incomplete and noisy data. The sheer focus on the majority class may lead to unequal distribution of classes and thus leads to data imbalance. The DLFGWO balances such distribution by leveraging grey wolves and its exploration and exploitation capabilities are improved using Discrete Levy Flight (DLF). It further optimizes the classifier's performance to achieve balanced classification rate. DLFGWO-RF is designed to perform classification even on limited datasets, thereby the requirement of numerous expert annotations can thus be reduced. In diabetic retinopathy grading, the DLFGWO-RF reduces disagreements in annotation variability using subjective interpretations. However, the representativeness of the diabetic retinopathy dataset fails to capture the entire population diversity, which limits the generalization ability of the proposed DLFGWO-RF. Thus, fine-tuning of RF can robustly adapt to the subgroups in the dataset, enhancing its overall performance. The experiments are conducted on two widely used medical image datasets to test the efficacy of the model. The experimental results show that the DLFGWO-RF classifier achieves improved classification accuracy between 90-95%, which outperforms the existing techniques for various imbalanced datasets.