Retinal Image Datasets Research Articles

Multi-class Classification of Retinal Eye Diseases from Ophthalmoscopy Images Using Transfer Learning-Based Vision Transformers.

This study explores a transfer learning approach with vision transformers (ViTs) and convolutional neural networks (CNNs) for classifying retinal diseases, specifically diabetic retinopathy, glaucoma, and cataracts, from ophthalmoscopy images. Using a balanced subset of 4217 images and ophthalmology-specific pretrained ViT backbones, this method demonstrates significant improvements in classification accuracy, offering potential for broader applications in medical imaging. Glaucoma, diabetic retinopathy, and cataracts are common eye diseases that can cause vision loss if not treated. These diseases must be identified in the early stages to prevent eye damage progression. This paper focuses on the accurate identification and analysis of disparate eye diseases, including glaucoma, diabetic retinopathy, and cataracts, using ophthalmoscopy images. Deep learning (DL) has been widely used in image recognition for the early detection and treatment of eye diseases. In this study, ResNet50, DenseNet121, Inception-ResNetV2, and six variations of ViT are employed, and their performance in diagnosing diseases such as glaucoma, cataracts, and diabetic retinopathy is evaluated. In particular, the article uses the vision transformer model as an automated method to diagnose retinal eye diseases, highlighting the accuracy of pre-trained deep transfer learning (DTL) structures. The updated ViT#5 model with the augmented-regularized pre-trained model (AugReg ViT-L/16_224) and learning rate of 0.00002 outperforms the state-of-the-art techniques, obtaining a data-based accuracy score of 98.1% on a publicly accessible retinal ophthalmoscopy image dataset, which includes 4217 images. In most categories, the model outperforms other convolutional-based and ViT models in terms of accuracy, precision, recall, and F1 score. This research contributes significantly to medical image analysis, demonstrating the potential of AI in enhancing the precision of eye disease diagnoses and advocating for the integration of artificial intelligence in medical diagnostics.

Hybrid Statistical-Textural and Intensity Feature-Based Approach for Accurate Discrimination of Retinal Diseases

Retinal imaging features such as fundus lesions are crucial signs of several ocular diseases, such as diabetic retinopathy and age-related macular degeneration. The early and accurate detection of the aforementioned abnormalities is essential for correct diagnosis and therapeutic management. In this study, we propose a new hybrid statistical-textural and intensity feature-based method for the improvement of the discrimination between retinal diseases with machine learning classifiers. By incorporating statistical texture analysis to measure changes in pixel intensity, this method not only extracts features based on intensities of fundus images but also serves as the comparison tool for a complete evaluation of the retina. A robust feature set is developed by merging intensity-based features with several texture descriptors including GLCM, LBP etc. We extract these features and then use them to train and test the performance of a number of classifiers, such as Support Vector Machines (SVM) and Random Forests for more accurate detection. Experiments have been conducted on a public dataset of retinal images and with the support vector machine classifier, encouraging results compared to purely texture-based or intensity-based methods in terms of classification accuracy are obtained. The overall findings of this study underscore the promise of integrating both statistical-textural and intensity features for improved retinal disease detection process, which will serve as an important stepping stone for future investigations and production of automated screening tools in retinal. The method also holds potential to enable early intervention in the development of vision-related diseases and, ultimately, decrease their impact.

Automated analysis of retinal vascular features, age, sex and disc size in a large cohort of primary open angle glaucoma patients

Aims/Purpose: Deriving vascular features of retinal images is a proposed noninvasive method to assess vascular health. Although several studies have linked cardiovascular risk to retinal image features, they often rely on limited datasets or have used deep learning approaches with limited explainability. This research introduces an end‐to‐end method for analyzing the retinal vasculature in large retinal image datasets, applied for primary open angle glaucoma (POAG).Methods: 115,237 retinal images of the UZ Leuven Glaucoma Clinic were extracted. 4858 unique images remain of POAG patients (n = 3010) and controls (n = 1848) after image quality assessment, optic disk detection, region of interest definition, automated segmentation of arterioles and venules (LUNet algorithm), and parametrization of the vascular biomarkers (PVBM toolbox). Analysis of covariance and linear mixed models were used to adjust for age, sex and disc size.Results: Both arteriolar and venular diameter, area, length, tortuosity, branching angle, endpoints, intersection points and both mono‐ as multifractal dimension levels are independently lower in POAG patients and older patients (all p < 0.001), after adjustment for sex, disc size and multiple testing. The multivariate linear mixed models additionally show that the vascular features are significantly influenced by sex and disc size, which necessitates correct adjustment.Conclusions: This is the first time a fully automated end‐to‐end pipeline is published for the analysis of retinal vascular geometry in a large cohort (n = 4858). Given the independent similarities in retinal vascular geometry changes between older age and POAG prevalence the theory of pronounced vascular ageing in POAG patients is proposed. This novel approach enables quantitative analysis of eye vasculature and supports the study of its correlation with specific diseases, facilitating reproducible analysis of large datasets and providing explainability to deep learning approaches.

WGAN-GP for Synthetic Retinal Image Generation: Enhancing Sensor-Based Medical Imaging for Classification Models.

Accurate synthetic image generation is crucial for addressing data scarcity challenges in medical image classification tasks, particularly in sensor-derived medical imaging. In this work, we propose a novel method using a Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) and nearest-neighbor interpolation to generate high-quality synthetic images for diabetic retinopathy classification. Our approach enhances training datasets by generating realistic retinal images that retain critical pathological features. We evaluated the method across multiple retinal image datasets, including Retinal-Lesions, Fine-Grained Annotated Diabetic Retinopathy (FGADR), Indian Diabetic Retinopathy Image Dataset (IDRiD), and the Kaggle Diabetic Retinopathy dataset. The proposed method outperformed traditional generative models, such as conditional GANs and PathoGAN, achieving the best performance on key metrics: a Fréchet Inception Distance (FID) of 15.21, a Mean Squared Error (MSE) of 0.002025, and a Structural Similarity Index (SSIM) of 0.89 in the Kaggle dataset. Additionally, expert evaluations revealed that only 56.66% of synthetic images could be distinguished from real ones, demonstrating the high fidelity and clinical relevance of the generated data. These results highlight the effectiveness of our approach in improving medical image classification by generating realistic and diverse synthetic datasets.

Development of an Open-Source Dataset of Flat-Mounted Images for the Murine Oxygen-Induced Retinopathy Model of Ischemic Retinopathy.

To describe an open-source dataset of flat-mounted retinal images and vessel segmentations from mice subject to the oxygen-induced retinopathy (OIR) model. Flat-mounted retinal images from mice killed at postnatal days 12 (P12), P17, and P25 used in prior OIR studies were compiled. Mice subjected to normoxic conditions were killed at P12, P17, and P25, and their retinas were flat-mounted for imaging. Major blood vessels from the OIR images were manually segmented by four graders (JSC, HKR, KBL, JM), with cross-validation performed to ensure similar grading. Overall, 1170 images were included in this dataset. Of these images, 111 were of normoxic mice retina, and 1048 were mice subject to OIR. The majority of images from OIR mice were obtained at P17. The 50 images obtained from an external dataset, OIRSeg, did not have age labels. All images were manually segmented and used in the training or testing of a previously published deep learning algorithm. This is the first open-source dataset of original and segmented flat-mounted retinal images. The dataset has potential applications for expanding the development of generalizable and larger-scale artificial intelligence and analyses for OIR. This dataset is published online and publicly available at dx.doi.org/10.6084/m9.figshare.23690973. This open access dataset serves as a source of raw data for future research involving big data and artificial intelligence research concerning oxygen-induced retinopathy.

A Method for Retina Segmentation by Means of U-Net Network

Retinal image segmentation plays a critical role in diagnosing and monitoring ophthalmic diseases such as diabetic retinopathy and age-related macular degeneration. We propose a deep learning-based approach utilizing the U-Net network for the accurate and efficient segmentation of retinal images. U-Net, a convolutional neural network widely used for its performance in medical image segmentation, is employed to segment key retinal structures, including the optic disc and blood vessels. We evaluate the proposed model on a publicly available retinal image dataset, demonstrating interesting performance in automatic retina segmentation, thus showing the effectiveness of the proposed method. Our proposal provides a promising method for automated retinal image analysis, aiding in early disease detection and personalized treatment planning.

Diabetic Retinopathy Detection with Deep Learning Techniques on Fundus Photographs: A Review

Diabetic retinopathy (DR) is an ocular condition that can affect individuals with diabetes mellitus (DM) and may lead to reduced vision or even blindness if not detected on time. Delay in diagnosis and disagreement in interpretation of retinal images by different health experts are some of the challenges that can occur during screening for DR. Deep learning (DL) techniques are currently used for classification of images across various domains including the ophthalmic imaging field. The implementation of this cutting edge technology for detection of DR could lead to improvement of existing eye care services for diabetic individuals. This paper discussed the publicly available datasets of retinal images of diabetic individuals used for training DL models. The efficiency of several convolutional neural networks (CNNs) created for the detection of different classes of DR was also reviewed. Furthermore, the achievements and the challenges faced in the application of DL techniques for the DR detection were discussed. Finally, future works that can be performed in this research area has been suggested.

RFMiD: Retinal Image Analysis for multi-Disease Detection challenge

In the last decades, many publicly available large fundus image datasets have been collected for diabetic retinopathy, glaucoma, and age-related macular degeneration, and a few other frequent pathologies. These publicly available datasets were used to develop a computer-aided disease diagnosis system by training deep learning models to detect these frequent pathologies. One challenge limiting the adoption of a such system by the ophthalmologist is, computer-aided disease diagnosis system ignores sight-threatening rare pathologies such as central retinal artery occlusion or anterior ischemic optic neuropathy and others that ophthalmologists currently detect. Aiming to advance the state-of-the-art in automatic ocular disease classification of frequent diseases along with the rare pathologies, a grand challenge on “Retinal Image Analysis for multi-Disease Detection” was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI - 2021). This paper, reports the challenge organization, dataset, top-performing participants solutions, evaluation measures, and results based on a new “Retinal Fundus Multi-disease Image Dataset” (RFMiD). There were two principal sub-challenges: disease screening (i.e. presence versus absence of pathology — a binary classification problem) and disease/pathology classification (a 28-class multi-label classification problem). It received a positive response from the scientific community with 74 submissions by individuals/teams that effectively entered in this challenge. The top-performing methodologies utilized a blend of data-preprocessing, data augmentation, pre-trained model, and model ensembling. This multi-disease (frequent and rare pathologies) detection will enable the development of generalizable models for screening the retina, unlike the previous efforts that focused on the detection of specific diseases.

Retinal fundus image super-resolution based on generative adversarial network guided with vascular structure prior

Many ophthalmic and systemic diseases can be screened by analyzing retinal fundus images. The clarity and resolution of retinal fundus images directly determine the effectiveness of clinical diagnosis. Deep learning methods based on generative adversarial networks are used in various research fields due to their powerful generative capabilities, especially image super-resolution. Although Real-ESRGAN is a recently proposed method that excels in processing real-world degraded images, it suffers from structural distortions when super-resolving retinal fundus images are rich in structural information. To address this shortcoming, we first process the input image using a pre-trained U-Net model to obtain a structural segmentation map of the retinal vessels and use the segmentation map as the structural prior. The spatial feature transform layer is then used to better integrate the structural prior into the generation process of the generator. In addition, we introduce channel and spatial attention modules into the skip connections of the discriminator to emphasize meaningful features and accordingly enhance the discriminative power of the discriminator. Based on the original loss functions, we introduce the L1 loss function to measure the pixel-level differences between the segmentation maps of retinal vascular structures in the high-resolution images and the super-resolution images to further constrain the super-resolution images. Simulation results on retinal image datasets show that our improved algorithm results have a better visual performance by suppressing structural distortions in the super-resolution images.

RetNet30: A Novel Stacked Convolution Neural Network Model for Automated Retinal Disease Diagnosis

ABSTRACTAutomated diagnosis of retinal diseases holds significant promise in enhancing healthcare efficiency and patient outcomes. However, existing methods often lack the accuracy and efficiency required for timely disease detection. To address this gap, we introduce RetNet30, a novel stacked convolutional neural network (CNN) designed to revolutionize automated retinal disease diagnosis. RetNet30 combines a custom‐built 30‐layer CNN with a fine‐tuned Inception V3 model, integrating these sub‐models through logistic regression to achieve superior classification performance. Extensive evaluations on retinal image datasets such as DRIVE, STARE, CHASE_DB1, and HRF demonstrate significant improvements in accuracy, sensitivity, specificity, and area under the ROC curve (AUROC) when compared to conventional approaches. By leveraging advanced deep learning architectures, RetNet30 not only enhances diagnostic precision but also generalizes effectively across diverse datasets, establishing a new benchmark in retinal disease classification. This novel approach offers a highly efficient and reliable solution for early disease detection and patient management, addressing the limitations of manual examination methods. Through rigorous quantitative and qualitative assessments, our proposed method demonstrates its potential to significantly impact medical image analysis and improve healthcare outcomes. RetNet30 marks a major step forward in automated retinal disease diagnosis, showcasing the future of AI‐driven advancements in ophthalmology.

Artificial intelligence-enabled profiling of overlapping retinal disease distribution for ocular diagnosis

<p>Eyesight, an invaluable gift profoundly impacts our daily lives. In a rapidly evolving healthcare landscape, the preservation and enhancement of ocular health stand as critical objectives. This research endeavors to analyze the two retinal fundus multi-disease image datasets (RFMiD) one containing 3200 images and the other containing 860 fundus images. The primary objective of this study is to scrutinize these datasets, discern variations in the frequency of labeled diseases within and across them, and explore common combinations of labels. These findings hold important implications for the field of retinal image analysis, as they provide valuable insights into the distribution and co-occurrence of defects.</p>

Hybrid convolutional neural network optimized with an artificial algae algorithm for glaucoma screening using fundus images.

We developed an optimized decision support system for retinal fundus image-based glaucoma screening. We combined computer vision algorithms with a convolutional network for fundus images and applied a faster region-based convolutional neural network (FRCNN) and artificial algae algorithm with support vector machine (AAASVM) classifiers. Optic boundary detection, optic cup, and optic disc segmentations were conducted using TernausNet. Glaucoma screening was performed using the optimized FRCNN. The Softmax layer was replaced with an SVM classifier layer and optimized with an AAA to attain enhanced accuracy. Using three retinal fundus image datasets (G1020, digital retinal images vessel extraction, and high-resolution fundus), we obtained accuracy of 95.11%, 92.87%, and 93.7%, respectively. Framework accuracy was amplified with an adaptive gradient algorithm optimizer FRCNN (AFRCNN), which achieved average accuracy 94.06%, sensitivity 93.353%, and specificity 94.706%. AAASVM obtained average accuracy of 96.52%, which was 3% ahead of the FRCNN classifier. These classifiers had areas under the curve of 0.9, 0.85, and 0.87, respectively. Based on statistical Friedman evaluation, AAASVM was the best glaucoma screening model. Segmented and classified images can be directed to the health care system to assess patients' progress. This computer-aided decision support system will be useful for optometrists.

Optimized automated detection of diabetic retinopathy severity: integrating improved multithresholding tunicate swarm algorithm and improved hybrid butterfly optimization.

Diabetic retinopathy, a complication of diabetes, damages the retina due to prolonged high blood sugar levels, leading to vision impairment and blindness. Early detection through regular eye exams and proper diabetes management are crucial in preventing vision loss. DR is categorized into five classes based on severity, ranging from no retinopathy to proliferative diabetic retinopathy. This study proposes an automated detection method using fundus images. Image segmentation divides fundus images into homogeneous regions, facilitating feature extraction. Feature selection aims to reduce computational costs and improve classification accuracy by selecting relevant features. The proposed algorithm integrates an Improved Tunicate Swarm Algorithm (ITSA) with Renyi's entropy for enhanced adaptability in the initial and final stages. An Improved Hybrid Butterfly Optimization (IHBO) Algorithm is also introduced for feature selection. The effectiveness of the proposed method is demonstrated using retinal fundus image datasets, achieving promising results in DR severity classification. For the IDRiD dataset, the proposed model achieves a segmentation Dice coefficient of 98.06% and classification accuracy of 98.21%. In contrast, the E-Optha dataset attains a segmentation Dice coefficient of 97.95% and classification accuracy of 99.96%. Experimental results indicate the algorithm's ability to accurately classify DR severity levels, highlighting its potential for early detection and prevention of diabetes-related blindness.

Extended-period AOSLO imaging in the living human retina without pupil dilation: a feasibility study.

In vivo imaging using an adaptive optics scanning laser ophthalmoscope (AOSLO) is challenging, especially over extended periods. Pharmacological agents, administered as eye drops, are commonly used to dilate the pupil and paralyse accommodation, to improve image quality. However, they are contraindicated in some scenarios. Here, we evaluate the feasibility and reproducibility of performing AOSLO imaging without pharmacological pupil dilation over 1.5 hours with visual stimulation. Through statistical analysis and theoretical modelling using a dataset of retinal and pupil images collected from six healthy, young, near-emmetropic participants between the ages of 20-30 years, we validate that the retinal image quality does not change significantly with time in the experimental session (p = 0.33), and that pupil size has a strong effect on image quality but is not the only contributing factor.

Retinal Image Dataset of Infants and Retinopathy of Prematurity

Retinopathy of prematurity (ROP) represents a vasoproliferative disease, especially in newborns and infants, which can potentially affect and damage the vision. Despite recent advances in neonatal care and medical guidelines, ROP still remains one of the leading causes of worldwide childhood blindness. The paper presents a unique dataset of 6,004 retinal images of 188 newborns, most of whom are premature infants. The dataset is accompanied by the anonymized patients’ information from the ROP screening acquired at the University Hospital Ostrava, Czech Republic. Three digital retinal imaging camera systems are used in the study: Clarity RetCam 3, Natus RetCam Envision, and Phoenix ICON. The study is enriched by the software tool ReLeSeT which is aimed at automatic retinal lesion segmentation and extraction from retinal images. Consequently, this tool enables computing geometric and intensity features of retinal lesions. Also, we publish a set of pre-processing tools for feature boosting of retinal lesions and retinal blood vessels for building classification and segmentation models in ROP analysis.

Transforming the future of ophthalmology: artificial intelligence and robotics' breakthrough role in surgical and medical retina advances: a mini review.

Over the past decade, artificial intelligence (AI) and its subfields, deep learning and machine learning, have become integral parts of ophthalmology, particularly in the field of ophthalmic imaging. A diverse array of algorithms has emerged to facilitate the automated diagnosis of numerous medical and surgical retinal conditions. The development of these algorithms necessitates extensive training using large datasets of retinal images. This approach has demonstrated a promising impact, especially in increasing accuracy of diagnosis for unspecialized clinicians for various diseases and in the area of telemedicine, where access to ophthalmological care is restricted. In parallel, robotic technology has made significant inroads into the medical field, including ophthalmology. The vast majority of research in the field of robotic surgery has been focused on anterior segment and vitreoretinal surgery. These systems offer potential improvements in accuracy and address issues such as hand tremors. However, widespread adoption faces hurdles, including the substantial costs associated with these systems and the steep learning curve for surgeons. These challenges currently constrain the broader implementation of robotic surgical systems in ophthalmology. This mini review discusses the current research and challenges, underscoring the limited yet growing implementation of AI and robotic systems in the field of retinal conditions.

LMBiS-Net: A lightweight bidirectional skip connection based multipath CNN for retinal blood vessel segmentation

Blinding eye diseases are often related to changes in retinal structure, which can be detected by analysing retinal blood vessels in fundus images. However, existing techniques struggle to accurately segment these delicate vessels. Although deep learning has shown promise in medical image segmentation, its reliance on specific operations can limit its ability to capture crucial details such as the edges of the vessel. This paper introduces LMBiS-Net, a lightweight convolutional neural network designed for the segmentation of retinal vessels. LMBiS-Net achieves exceptional performance with a remarkably low number of learnable parameters (only 0.172 million). The network used multipath feature extraction blocks and incorporates bidirectional skip connections for the information flow between the encoder and decoder. In addition, we have optimised the efficiency of the model by carefully selecting the number of filters to avoid filter overlap. This optimisation significantly reduces training time and improves computational efficiency. To assess LMBiS-Net’s robustness and ability to generalise to unseen data, we conducted comprehensive evaluations on four publicly available datasets: DRIVE, STARE, CHASE_DB1, and HRF The proposed LMBiS-Net achieves significant performance metrics in various datasets. It obtains sensitivity values of 83.60%, 84.37%, 86.05%, and 83.48%, specificity values of 98.83%, 98.77%, 98.96%, and 98.77%, accuracy (acc) scores of 97.08%, 97.69%, 97.75%, and 96.90%, and AUC values of 98.80%, 98.82%, 98.71%, and 88.77% on the DRIVE, STARE, CHEASE_DB, and HRF datasets, respectively. In addition, it records F1 scores of 83.43%, 84.44%, 83.54%, and 78.73% on the same datasets. Our evaluations demonstrate that LMBiS-Net achieves high segmentation accuracy (acc) while exhibiting both robustness and generalisability across various retinal image datasets. This combination of qualities makes LMBiS-Net a promising tool for various clinical applications.

Deep learning model using classification for diabetic retinopathy detection: an overview

Early detection of diabetic retinopathy is a serious disease for diabetics to minimize their sightlessness risks. The different approaches take a much longer time for a very large training dataset. In classifying to better the accuracy of diabetic retinopathy, a novel technique called MAP Concordance Regressive Camargo’s Index-Based Deep Multilayer Perceptive Learning Classification (MAPCRCI-DMPLC) has been introduced with minimum time consumption. The novel model of MAPCRCI-DMPLC comprises the input layer, hidden layers, and output layer for detecting diabetic retinopathy at an early stage through high accuracy and less moment consumption. The proposed MAPCRCI-DMPLC model collected the retinal fundus images from the dataset as input. After that, we carried out image preprocessing using the MAP-estimated local region filtering-based preprocessing technique in the first hidden layer. In the second hidden layer, Camargo’s index-based ROI extraction is performed to identify the infected region. Then, Concordance Correlative Regression is applied for texture feature extraction. Then the color feature is extracted, beginning the image. The features extracted to the output layer to classify the different levels of DR using the swish activation function through higher accuracy. An investigational assessment using a retinal image dataset on factors such as peak signal-to-noise ratio (PSNR), disease detection accuracy (DDA), false-positive rate (FPR), and disease detection time (DDT), regarding the quantity of retinal fundus images and image dimension. The quantitative and qualitatively analyzed outcome shows a better presentation of our proposed MAPCRCI-DMPLC technique when compared through the five state-of-the-art approaches.

PAM-UNet: Enhanced Retinal Vessel Segmentation Using a Novel Plenary Attention Mechanism

Retinal vessel segmentation is critical for diagnosing related diseases in the medical field. However, the complex structure and variable size and shape of retinal vessels make segmentation challenging. To enhance feature extraction capabilities in existing algorithms, we propose PAM-UNet, a U-shaped network architecture incorporating a novel Plenary Attention Mechanism (PAM). In the BottleNeck stage of the network, PAM identifies key channels and embeds positional information, allowing spatial features within significant channels to receive more focus. We also propose a new regularization method, DropBlock_Diagonal, which discards diagonal regions of the feature map to prevent overfitting and enhance vessel feature learning. Within the decoder stage of the network, features from each stage are merged to enhance the segmentation accuracy of the final vessel. Experimental validation on two retinal image datasets, DRIVE and CHASE_DB1, shows that PAM-UNet achieves 97.15%, 83.16%, 98.45%, 83.15%, 98.66% and 97.64%, 85.82%, 98.46%, 82.56%, 98.95% on Acc, Se, Sp, F1, AUC, respectively, outperforming UNet and most other retinal vessel segmentation algorithms.

Analisis Retinal Optical Coherence Tomography (OCT) Untuk Deteksi Kerusakan Retina Menggunakan Metode Machine Learning

This study attempts to use support vector machine and otsu thresholding as proposed algorithm models to classify Retinal optical coherence tomography (OCT) images. In this study, there are two types implemented in classifying retinal image datasets. The first scenario is to classify using the support vector machine algorithm without the otsu thresholding method and the second scenario is to classify using the support vector machine algorithm with the otsu thresholding method with various parameter values. Based on the experimental results, classification of retina image datasets using the support vector machine algorithm without the otsu thresholding method obtained an accuracy of 63.00% while classification using the support vector machine algorithm with the otsu thresholding method with parameter values (0, 255), (50, 255), (100, 255), (150, 255) obtained an accuracy of 59.30%.