Color Fundus Images Research Articles

Deep Learning to Discriminate Arteritic From Nonarteritic Ischemic Optic Neuropathy on Color Images

Prompt and accurate diagnosis of arteritic anterior ischemic optic neuropathy (AAION) from giant cell arteritis and other systemic vasculitis can contribute to preventing irreversible vision loss from these conditions. Its clinical distinction from nonarteritic anterior ischemic optic neuropathy (NAION) can be challenging, especially when systemic symptoms are lacking or laboratory markers of the disease are not reliable. To develop, train, and test a deep learning system (DLS) to discriminate AAION from NAION on color fundus images during the acute phase. This was an international study including color fundus images of 961 eyes of 802 patients with confirmed AAION and NAION. Training was performed using images from 21 expert neuro-ophthalmology centers in 16 countries, while external testing was performed in a cohort from 5 expert neuro-ophthalmology centers in the US and Europe. Data for training and external testing were collected from August 2018 to January 2023. A mix of deidentified images of 2 fields of view (optic disc centered and macula centered) were used. For training and internal validation, images were from 16 fundus camera models with fields of 30° to 55°. For external testing, images were from 5 fundus cameras with fields of 30° to 50°. Data were analyzed from January 2023 to January 2024. The performance of the DLS was measured using area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy. In the training and validation sets, 374 (54.9%) of patients were female, 301 (44.2%) were male, and 6 (0.9%) were of unknown sex; the median (range) age was 66 (23-96) years. When tested on the external dataset including 121 patients (35 [28.9%] female, 44 [36.4%] male, and 42 [34.7%] of unknown sex; median [range] age, 69 [37-89] years), the DLS achieved an AUC of 0.97 (95% CI, 0.95-0.99), a sensitivity of 91.1% (95% CI, 85.2-96.9), a specificity of 93.4% (95% CI, 91.1-98.2), and an accuracy of 92.6% (95% CI, 90.5-96.6). The accuracy of the 2 experts for classification of the same dataset was 74.3% (95% CI, 66.7-81.9) and 81.6% (95% CI, 74.8-88.4), respectively. A DLS showing disease-specific averaged class-activation maps had greater than 90% accuracy at discriminating between acute AAION from NAION on color fundus images, at the eye level, without any clinical or biomarker information. A DLS that identifies AAION could improve clinical decision-making, potentially reducing the risk of misdiagnosis and improving patient outcomes.

CNN-Based Integrated Framework for Enhanced Diabetic Retinopathy Detection

Background: Diabetic Retinopathy (DR), a significant cause of vision loss globally, is characterized by retinal damage caused by diabetes. Early detection is vital to prevent irreversible blindness, yet challenges remain in accurately identifying DR stages and enhancing blood vessel visibility in fundus images. This paper aims to develop an early detection methodology for DR, addressing the need for early diagnosis and the difficulties in distinguishing DR severity through fundus imaging. The challenges in the early detection of Diabetic Retinopathy (DR) and enhancing blood vessel visibility in fundus images are multifaceted, including issues such as data imbalance, image noise, and complex patterns. By addressing these challenges through advanced ML techniques and image processing methodologies, the proposed methodology in the paper aims to overcome the limitations in early detection and severity assessment of DR, contributing to improved patient outcomes and vision preservation. Methods: This study utilizes Machine Learning (ML) to analyze complex patterns in fundus color images of DR, employing spatial domain filtering to reduce image noise and address data imbalances across DR severity levels through data augmentation. A Convolutional Neural Network (CNN), enhanced with a Gabor filter, is applied for stage-specific DR detection and to pinpoint infected areas. The dataset includes 1000 color fundus images, with a 70:30 split for training and testing, respectively. The adoption of a Gabor filtering technique aims to refine the model’s performance further. Results: The incorporation of a CNN with a Gabor filter has shown outstanding efficacy in detecting DR from fundus color images, achieving a training accuracy of 97.5%, validation accuracy of 96.5%, Cohen kappa score of 89.76%, and testing accuracy of 95.87%. This method effectively illustrates the disease-affected areas in the fundus images provided. Conclusion: Through a comparative analysis of image processing techniques, this research highlights the advantages of using advanced DR analysis for image preprocessing. The proposed CNN-Gabor filter approach demonstrates significant success in identifying diabetic retinopathy in fundus color photographs and accurately delineating the affected regions, contributing valuable insights to the field of medical image processing.

Rheumatic fever and long-term use of benzathine penicillin as possible risk factors for extensive macular atrophy with pseudodrusen in a Brazilian cohort

BackgroundAlthough there has been a large increase in the number of extensive macular atrophy with pseudodrusen (EMAP) cases, the basic aspects of this disease remain unknown. Brazilian patients have a common past history of rheumatic fever (RF) and/or benzathine penicillin (BP) treatment possibly related to the disease. We analyzed how RF and BP might be correlated with EMAP in Brazilian patients.DesignObservational, retrospective, case-control study.MethodsThe databases of three private eye clinics in Brazil were searched for patients with an EMAP-like appearance. Each patient was asked about a previous history of RF and/or long-term use of BP. Patients underwent best-corrected visual acuity (BCVA) measurement, color fundus imaging, fundus autofluorescence (FAF) imaging, optical coherence tomography (OCT) imaging, and electroretinography (ERG). The following characteristics were analyzed: subretinal drusenoid deposits (SDD), pigment mottling, retinal pigment epithelial/basement membrane (RPE/BM) separation, outer retinal or RPE atrophy, and identification of a paving stone-like appearance. The choroidal thickness was measured using enhanced depth imaging OCT. The central atrophic area was measured manually on ultra-wide-field FAF.ResultsA total of 154 eyes of 77 patients (women, 66.2%; mean age, 58.6 years) with EMAP were included; 90.9% of patients were diagnosed with RF; 94.8% had been treated with BP and treatment was started at an average age of 7.3 years (mean duration, 11.8 years). The treatment duration was significant for the area of atrophy (P = 0.027) in which each 1-year increase in treatment duration led to an average reduction of 6.91 mm2 in area. The age at diagnosis of RF was significant (P = 0.026) for SDD. The increase of 1 year in the diagnosis of RF (late disease) led to a reduction of 24% in the chance of central SDD being present. On OCT, 65.5% eyes had SDD and more than 70% had a split RPE/BM and outer retinal or RPE atrophy. The choroidal thickness in patients with EMAP was significantly (P < 0.001) thinner than the control group. The ERG was abnormal in all eyes.ConclusionThese findings may suggest a relation between RF and EMAP in Brazilian patients. Patients with EMAP should be questioned about a history of RF.

Spectrum of retinal microvascular ischemia in patients with COVID-19 based on multimodal imaging

PurposeTo describe and evaluate the multimodal imaging findings in retinal microvascular ischemia associated with COVID-19 infection. MethodsPatients with COVID-19 associated retinal microvascular ischemia and visiting the outpatient Department of Ophthalmology, Shanghai General Hospital from December 2022, to February 2023, were documented and their multimodal images were retrospectively reviewed. Retinal microvascular ischemia was defined as the presence of isolated or multiple focal retinal whitening(s) on color fundus images. Patients with retinal vessel occlusion or retinopathies secondary to systematic disorders diagnosed before infection were excluded. ResultsA total of 32 eyes from 21 patients were included, 24 (75.00 %) eyes with multiple retinal whitenings, while 8 (25.00 %) eyes with isolated lesions. When divided by the types of ischemia, 9 (28.13 %) eyes had only inner retinal involvement (known as cotton wool spot, CWS), 4 (12.50 %) eyes had only middle retinal involvement (known as paracentral acute middle maculopathy, PAMM), and 19 (59.38 %) eyes had both. In addition, 4 (12.50 %) eyes had coincident angular sign of Henle fiber layer hyperreflectivity (ASHH). Patients with hypertension tended to have multiple lesions rather than isolated lesion of retinal microvascular ischemia (P = 0.008). Transient uncontrolled high blood pressure or acute kidney injury was simultaneously detected in some cases. ConclusionsOcular manifestation of COVID-19 associated microvascular ischemia can be variable, including CWS, PAMM and ASHH. Multimodal fundus imaging technologies are useful tools to reveal involved retinal layers, extent, and severity. Moreover, ocular manifestations may serve as a window of COVID-19 related microcirculation in other systems throughout the body.

Generating Synthesized Fluorescein Angiography Images From Color Fundus Images by Generative Adversarial Networks for Macular Edema Assessment.

To assess the feasibility of generating synthetic fluorescein angiography (FA) images from color fundus (CF) images using pixel-to-pixel generative adversarial network (pix2pixGANs) for clinical applications. Research questions addressed image realism to retinal specialists and utility for assessing macular edema (ME) in Retinal Vein Occlusion (RVO) eyes. We used a registration-guided pix2pixGANs method trained on the CF-FA dataset from Kham Eye Centre, Kandze Prefecture People's Hospital. A visual Turing test confirmed the realism of synthetic images without novel artifacts. We then assessed the synthetic FA images for assessing ME. Finally, we quantitatively evaluated the synthetic images using Fréchet Inception distance (FID) and structural similarity measures (SSIM). The raw development dataset had 881 image pairs from 349 subjects. Our approach is capable of generating realistic FA images because small vessels are clearly visible and sharp within one optic disc diameter around the macula. Two retinal specialists agreed that more than 85% of synthetic FA images have good or excellent image quality. For ME detection, accuracy was similar for real and synthetic images. FID demonstrated a 38.9% improvement over the previous state-of-the-art (SOTA), and SSIM reached 0.78 compared to the previous SOTA's 0.67. We developed a pix2pixGANs model translating FA images from label-free CF images, yielding reliable synthetic FA images. This suggests potential for noninvasive evaluation of ME in RVO eyes using pix2pix GANs techniques. Pix2pixGANs techniques have the potential to assist in the noninvasive clinical assessment of ME in RVO eyes.

A Meta-Learning Approach for Classifying Multimodal Retinal Images of Retinal Vein Occlusion With Limited Data.

To propose and validate a meta-learning approach for detecting retinal vein occlusion (RVO) from multimodal images with only a few samples. In this cross-sectional study, we formulate the problem as meta-learning. The meta-training dataset consists of 1254 color fundus (CF) images from 39 different fundus diseases. Two meta-testing datasets include a public domain dataset and an independent dataset from Kandze Prefecture People's Hospital. The proposed meta-learning models comprise two modules: the feature extraction networks and the prototypical networks (PNs). We use two deep learning models (the ResNet and the Contrastive Language-Image Pre-Training networks [CLIP]) for feature extraction. We evaluate the performance of the algorithms using accuracy, area under the receiver operating characteristic curve (AUCROC), F1-score, and recall. CLIP-based PNs outperform across all meta-testing datasets. For the public APTOS dataset, meta-learning algorithms achieve good results with an accuracy of 86.06% and an AUCROC of 0.87 with only 16 training images. In the hospital datasets, meta-learning algorithms show excellent diagnostic capability for detecting RVO with a very low number of shots (AUCROC above 0.99 for n = 4, 8, and 16, respectively). Notably, even though the meta-training dataset does not include fluorescein angiography (FA) images, meta-learning algorithms also have excellent diagnostic capability for detecting RVO from images with a different modality (AUCROC above 0.93 for n = 4, 8, and 16, respectively). The proposed meta-learning models excel in detecting RVO, not only on CF images but also on FA images from a different imaging modality. The proposed meta-learning models could be useful in automatically detecting RVO on CF and FA images.

Application of Artificial Intelligence in Ophthalmology: An Updated Comprehensive Review.

Artificial intelligence (AI) holds immense promise for transforming ophthalmic care through automated screening, precision diagnostics, and optimized treatment planning. This paper reviews recent advances and challenges in applying AI techniques such as machine learning and deep learning to major eye diseases. In diabetic retinopathy, AI algorithms analyze retinal images to accurately identify lesions, which helps clinicians in ophthalmology practice. Systems like IDx-DR (IDx Technologies Inc, USA) are FDA-approved for autonomous detection of referable diabetic retinopathy. For glaucoma, deep learning models assess optic nerve head morphology in fundus photographs to detect damage. In age-related macular degeneration, AI can quantify drusen and diagnose disease severity from both color fundus and optical coherence tomography images. AI has also been used in screening for retinopathy of prematurity, keratoconus, and dry eye disease. Beyond screening, AI can aid treatment decisions by forecasting disease progression and anti-VEGF response. However, potential limitations such as the quality and diversity of training data, lack of rigorous clinical validation, and challenges in regulatory approval and clinician trust must be addressed for the widespread adoption of AI. Two other significant hurdles include the integration of AI into existing clinical workflows and ensuring transparency in AI decision-making processes. With continued research to address these limitations, AI promises to enable earlier diagnosis, optimized resource allocation, personalized treatment, and improved patient outcomes. Besides, synergistic human-AI systems could set a new standard for evidence-based, precise ophthalmic care.

Automatic retinal image analysis methods using colour fundus images for screening glaucomatous optic neuropathy

ObjectivesTrain an automatic retinal image analysis (ARIA) method to screen glaucomatous optic neuropathy (GON) on non-mydriatic retinal images labelled with the additional results of optical coherence tomography (OCT) and assess...

Eye Deep-Net: a deep neural network-based multi-class retinal disease diagnostic

Ophthalmologists rely heavily on retinal pictures to diagnose a wide range of eye conditions. Numerous retinal disorders may lead to microvascular alterations in the retina, and a number of studies have been conducted on the early identification of medical pictures to enable prompt and appropriate treatment. In order to identify various eye illnesses using color fundus pictures, this study develops a non-invasive, automated deep learning system. A multiclass ocular illness A productive diagnostic approach was created using the Remind dataset. A variety of augmentation strategies were used to make the structure robust in real-time after multi-class fundus pictures were collected from a multi-label dataset. Low computational demand images were processed in accordance with the network. The fundamental convolutional neural network (CNN) extracts appropriate characteristics from the input color fundus image dataset, and then processed characteristics were employed to make predictive diagnoses. This multi-layer neural network, called Eye Deep-Net, has been developed for training and evaluating images for the recognition of various eye problems. The performance of the suggested model is determined to be much better than numerous baseline state-of-the-art models. The strength from the Eye Deep-Net is assessed using different statistical metrics. The suggested methodology's effectiveness in classifying and identifying diseases using digital fundus pictures is shown by a thorough comparison with the most modern techniques.

A Cascade U‐Net With Transformer for Retinal Multi‐Lesion Segmentation

ABSTRACTDiabetic retinopathy (DR) is an important cause of blindness. If not diagnosed and treated in a timely manner, it can lead to irreversible vision loss. The diagnosis of DR relies heavily on specialized ophthalmologists. In recent years, with the development of artificial intelligence a number of diagnostics using this technique have begun to appear. One method for diagnosing diseases in this field is to segment four common kinds of lesions from color fundus images, including: exudates (EX), soft exudates (SE), hemorrhages (HE), and microaneurysms (MA). In this paper, we propose a segmentation model for DR based on deep learning. The main part of the model consists of two layers of improved U‐Net network based on transformer, corresponding to the two stages of coarse segmentation and fine segmentation, respectively. The model can segment four common kinds of lesions from the input color fundus image at the same time. To validate the performance of our proposed model, we test our model on three public datasets: IDRiD, DDR, and DIARETDB1. The test results show that our proposed model achieves competitive results compared with the existing methods in terms of PR‐AUC, ROC‐AUC, Dice, and IoU, especially for lesions segmentation of SE and MA.

Fine-Tuning SSL-Model to Enhance Detection of Cilioretinal Arteries on Colored Fundus Images.

Cilioretinal arteries are a common congenital anomaly of retinal blood supply. This paper presents a deep learning-based approach for the automated detection of a CRA from color fundus images. Leveraging the Vision Transformer architecture, a pre-trained model from RETFound was fine-tuned to transfer knowledge from a broader dataset to our specific task. An initial dataset of 85 was expanded to 170 images through data augmentation using self-supervised learning-driven techniques. To address the imbalance in the dataset and prevent overfitting, Focal Loss and Early Stopping were implemented. The model's performance was evaluated using a 70-30 split of the dataset for training and validation. The results showcase the potential of ophthalmic foundation models in enhancing detection of CRAs and reducing the effort required for labeling by retinal experts, as promising results could be achieved with only a small amount of training data through fine-tuning.

Real-world evaluation of RetCAD deep-learning system for the detection of referable diabetic retinopathy and age-related macular degeneration

ABSTRACT Clinical Relevance The challenges of establishing retinal screening programs in rural settings may be mitigated by the emergence of deep-learning systems for early disease detection. Background Deep-learning systems have demonstrated promising results in retinal disease detection and may be particularly useful in rural settings where accessibility remains a barrier to equitable service provision. This study aims to evaluate the real-world performance of Thirona RetCAD for the detection of referable diabetic retinopathy and age-related macular degeneration in a rural Australian population. Methods Colour fundus images from participants with known diabetic retinopathy or age-related macular degeneration were randomly selected from ophthalmology clinics in four rural Australian centres. Grading was confirmed retrospectively by two retinal specialists. RetCAD produced a quantitative measure (0–100) for DR and AMD severity. The area under the ROC curve (AUC) was calculated. Sensitivity, specificity, and positive and negative predictive values were calculated at a pre-defined cut-point of ≥50. Results A total of 150 images from 82 participants were included. The mean age (SD) was 64.0 (12.8) years. Seventy-nine (52.7%) eyes had evidence of referable DR, while 54 (36.0%) had evidence of referable AMD. The AUC for referable DR detection was 0.971 (95% CI 0.950–0.936) with a sensitivity of 86.1% (76.8%–92.0%) and a specificity of 91.6% (82.8%−96.1%) at the pre-defined cut-point. Using the Youden Index method, the optimal cut-point was 41.2 (sensitivity 93.7%, specificity 90.1%). The AUC for the detection of referable AMD was 0.880 (0.824–0.936). At the pre-defined cut-point sensitivity was 88.9% (77.8%–94.8%) and specificity was 66.7% (56.8%–75.3%). The optimal cut-point was 52.6 (sensitivity 87.0%, specificity 75.0%). Conclusion RetCAD is comparable with but does not outperform equivalent deep-learning systems for retinal disease detection. RetCAD may be suitable as an automated screening tool in a rural Australian setting.

Artificial intelligence-based classification of cardiac autonomic neuropathy from retinal fundus images in patients with diabetes: The Silesia Diabetes Heart Study

BackgroundCardiac autonomic neuropathy (CAN) in diabetes mellitus (DM) is independently associated with cardiovascular (CV) events and CV death. Diagnosis of this complication of DM is time-consuming and not routinely performed in the clinical practice, in contrast to fundus retinal imaging which is accessible and routinely performed. Whether artificial intelligence (AI) utilizing retinal images collected through diabetic eye screening can provide an efficient diagnostic method for CAN is unknown.MethodsThis was a single center, observational study in a cohort of patients with DM as a part of the Cardiovascular Disease in Patients with Diabetes: The Silesia Diabetes-Heart Project (NCT05626413). To diagnose CAN, we used standard CV autonomic reflex tests. In this analysis we implemented AI-based deep learning techniques with non-mydriatic 5-field color fundus imaging to identify patients with CAN. Two experiments have been developed utilizing Multiple Instance Learning and primarily ResNet 18 as the backbone network. Models underwent training and validation prior to testing on an unseen image set.ResultsIn an analysis of 2275 retinal images from 229 patients, the ResNet 18 backbone model demonstrated robust diagnostic capabilities in the binary classification of CAN, correctly identifying 93% of CAN cases and 89% of non-CAN cases within the test set. The model achieved an area under the receiver operating characteristic curve (AUCROC) of 0.87 (95% CI 0.74–0.97). For distinguishing between definite or severe stages of CAN (dsCAN), the ResNet 18 model accurately classified 78% of dsCAN cases and 93% of cases without dsCAN, with an AUCROC of 0.94 (95% CI 0.86–1.00). An alternate backbone model, ResWide 50, showed enhanced sensitivity at 89% for dsCAN, but with a marginally lower AUCROC of 0.91 (95% CI 0.73–1.00).ConclusionsAI-based algorithms utilising retinal images can differentiate with high accuracy patients with CAN. AI analysis of fundus images to detect CAN may be implemented in routine clinical practice to identify patients at the highest CV risk.Trial registrationThis is a part of the Silesia Diabetes-Heart Project (Clinical-Trials.gov Identifier: NCT05626413).Graphical

A Deep Learning Approach for Accurate Discrimination Between Optic Disc Drusen and Papilledema on Fundus Photographs.

Optic disc drusen (ODD) represent an important differential diagnosis of papilledema caused by intracranial hypertension, but their distinction may be difficult in clinical practice. The aim of this study was to train, validate, and test a dedicated deep learning system (DLS) for binary classification of ODD vs papilledema (including various subgroups within each category), on conventional mydriatic digital ocular fundus photographs collected in a large international multiethnic population. This retrospective study included 4,508 color fundus images in 2,180 patients from 30 neuro-ophthalmology centers (19 countries) participating in the Brain and Optic Nerve Study with Artificial Intelligence (BONSAI) Group. For training and internal validation, we used 857 ODD images and 3,230 papilledema images, in 1,959 patients. External testing was performed on an independent data set (221 patients), including 207 images with ODD (96 visible and 111 buried), provided by 3 centers of the Optic Disc Drusen Studies Consortium, and 214 images of papilledema (92 mild-to-moderate and 122 severe) from a previously validated study. The DLS could accurately distinguish between all ODD and papilledema (all severities included): area under the receiver operating characteristic curve (AUC) 0.97 (95% confidence interval [CI], 0.96-0.98), accuracy 90.5% (95% CI, 88.0%-92.9%), sensitivity 86.0% (95% CI, 82.1%-90.1%), and specificity 94.9% (95% CI, 92.3%-97.6%). The performance of the DLS remained high for discrimination of buried ODD from mild-to-moderate papilledema: AUC 0.93 (95% CI, 0.90-0.96), accuracy 84.2% (95% CI, 80.2%-88.6%), sensitivity 78.4% (95% CI, 72.2%-84.7%), and specificity 91.3% (95% CI, 87.0%-96.4%). A dedicated DLS can accurately distinguish between ODD and papilledema caused by intracranial hypertension, even when considering buried ODD vs mild-to-moderate papilledema.

Artificial intelligence in the classification and segmentation of fundus images with choroidal nevi

ObjectiveThe purpose of this study is to summarize the results from 3 experimental studies into the use of artificial intelligence to classify and segment colour fundus images with choroidal nevi. Study DesignThis study is based on a secondary analysis of colour fundus images taken of patients receiving usual clinical care from the Alberta Ocular Brachytherapy Program. MethodsHigh-resolution colour fundus images were labeled by experienced ocular oncologists. In experimental study 1, four pre-trained models (ResNet 50, VGG-19, VGG-16, and AlexNet) were evaluated for their ability to classify images based on the presence of choroidal nevi. In experimental study 2, the performance of 3 patch-based models to classify images based on the presence of choroidal nevi were compared. In experimental study 3, four convolutional neural network models were developed to segment the images. In experimental studies 1 and 2, performance was measured using accuracy, precision, recall, F1 score, and AUC. In experimental study 3, IoU and Dice measures were used to evaluate performance. ResultsA total of 591 labelled colour fundus images were used for analysis. In experimental study 1, VGG-16 showed the best accuracy, AUC, and recall, but lower precision in classifying images. In experimental study 2, the patched approached enhanced with artifact and contrast outperformed the others in classifying images. In experimental study 3, a voting-based Ensemble model excelled in segmenting the part of images with nevi. ConclusionsIt is feasible to train AI models to identify choroidal nevi in colour fundus images.

Correlation of posterior segment lesions with anterior segment biometric parameters and FBN1 genotype in patients with Marfan syndrome

Objective: To investigate the characteristics of posterior segment lesions in Marfan syndrome (MFS) patients and their relationship with anterior segment biometric parameters and FBN1 genotype. Methods: A cross-sectional study was conducted. A total of 121 MFS patients, 76 males and 45 females, with an average age of (11.72±11.66) years, who visited the Department of Ophthalmology, Eye & ENT Hospital of Fudan University from January 2013 to March 2023 were included. The presence of posterior scleral staphyloma was observed using B-mode ultrasound, and macular lesions were identified and classified using the atrophy-traction-neovascularization system based on ultra-widefield fundus images, color fundus images, and optical coherence tomography scans. Anterior segment biometric parameters, including axial length of the eye, average corneal curvature, corneal astigmatism, horizontal corneal diameter, anterior chamber depth, and lens thickness, were collected, and the direction and extent of lens dislocation were observed. Molecular genetic analysis of FBN1 gene mutations in patients was performed using next-generation sequencing based on a panel of ocular genetic diseases, and the impact of the genotype and anterior segment biometric parameters on the posterior segment manifestations was analyzed. Results: Sixty patients exhibited posterior segment lesions, including retinal detachment (4 cases, 3.31%), macular lesions (47 cases, 38.84%), and posterior scleral staphyloma (54 cases, 44.63%). There was statistically significant difference in axial length of the eye between patients with and without posterior scleral staphyloma [23.09 (22.24, 24.43) and 27.04 (25.44, 28.88) mm], between patients with and without macular lesions [23.16 (22.24, 24.61) and 27.04 (25.74, 28.78) mm], and between patients with and without atrophic macular lesions [23.16 (22.24, 24.61) and 27.04 (25.74, 28.79) mm] (all P<0.001). There was statistically significant difference in anterior chamber depth between patients with and without macular lesions [3.11 (2.75, 3.30) and 3.34 (3.09, 3.60) mm] (P<0.05). There was also statistically significant difference in corneal astigmatism between patients with and without posterior scleral staphyloma [2.15 (1.20, 2.93) and 1.40 (1.00, 2.20) diopters] (P<0.05). The location and region of the FBN1 gene mutation not only showed statistically significant difference from the positive rates of posterior scleral staphyloma and macular lesions (all P<0.05), but also influenced the occurrence of atrophic macular lesions (both P<0.05). Patients with FBN1 mutations located in the transforming growth factor β regulatory sequence had the highest proportion of posterior scleral staphyloma and macular lesions (both 10/11). Conclusions: Posterior scleral staphyloma and macular lesions have a relatively high incidence in MFS patients and tend to progress to more severe grades. The age, axial length of the eye, anterior chamber depth, corneal astigmatism, and location and region of the FBN1 gene mutation are factors affecting the posterior segment lesions in MFS patients.

Multi-dimensional dense attention network for pixel-wise segmentation of optic disc in colour fundus images.

Segmentation of retinal fragments like blood vessels, Optic Disc (OD), and Optic Cup (OC) enables the early detection of different retinal pathologies like Diabetic Retinopathy (DR), Glaucoma, etc. Accurate segmentation of OD remains challenging due to blurred boundaries, vessel occlusion, and other distractions and limitations. These days, deep learning is rapidly progressing in the segmentation of image pixels, and a number of network models have been proposed for end-to-end image segmentation. However, there are still certain limitations, such as limited ability to represent context, inadequate feature processing, limited receptive field, etc., which lead to the loss of local details and blurred boundaries. A multi-dimensional dense attention network, or MDDA-Net, is proposed for pixel-wise segmentation of OD in retinal images in order to address the aforementioned issues and produce more thorough and accurate segmentation results. In order to acquire powerful contexts when faced with limited context representation capabilities, a dense attention block is recommended. A triple-attention (TA) block is introduced in order to better extract the relationship between pixels and obtain more comprehensive information, with the goal of addressing the insufficient feature processing. In the meantime, a multi-scale context fusion (MCF) is suggested for acquiring the multi-scale contexts through context improvement. Specifically, we provide a thorough assessment of the suggested approach on three difficult datasets. In the MESSIDOR and ORIGA data sets, the suggested MDDA-NET approach obtains accuracy levels of 99.28% and 98.95%, respectively. The experimental results show that the MDDA-Net can obtain better performance than state-of-the-art deep learning models under the same environmental conditions.

Deep Learning for Early Detection of Diabetic Retinopathy: A Colour Fundus Image-based Approach with VGG16 and VGG19

Diabetic retinopathy (DR) is a prevalent complication of diabetes, posing a significant threat to vision. Early detection and intervention are critical to prevent vision loss. However, manual screening of retinal images for DR is time-consuming and susceptible to human error. This study investigates the application of deep learning models, specifically VGG16 and VGG19, for automated DR diagnosis in Colour Fundus images. We evaluate their performance in classifying images as either Diabetic Retinopathy or No Diabetic Retinopathy. Our results demonstrate high testing accuracy with VGG16 achieving 93.96%, suggesting a promising approach for automated DR screening and aiding healthcare professionals in early and accurate diagnosis. Keywords : Convolutional Neural Network, Diabetic Retinopathy, Visual Geometry Group 16 (VGG16), Visual Geometry Group 19 (VGG19), Transfer Learning, Deep Learning.

Blue Wavelength of Scanning Laser Ophthalmoscope Potentially Detects Arteriosclerotic Lesions in Diabetic Retinopathy.

(1) Background: The fundus examination is one of the best and popular methods in the assessment of vascular status in the human body. Direct viewing of retinal vessels by ophthalmoscopy has been utilized in judging hypertensive change or arteriosclerosis. Recently, fundus imaging with the non-mydriatic scanning laser ophthalmoscope (SLO) has been widely used in ophthalmological clinics since it has multimodal functions for optical coherence tomography or angiography with contrast agent dye. The purpose of this study was to examine the utility in detecting arteriosclerosis of retinal vessels in SLO images; (2) Methods: Both color and blue standard field SLO images of eyes with diabetic retinopathy (DR) were examined retrospectively. Retinal arteriosclerosis in color SLO images was graded according to the Scheie classification. Additionally, characteristics of retinal arterioles in blue SLO images were identified and examined for their relevance to arteriosclerosis grades, stages of DR or general complications; (3) Results: Relative to color fundus images, blue SLO images showed distinct hyper-reflective retinal arterioles against a monotone background. Irregularities of retinal arterioles identified in blue SLO images were frequently observed in the eyes of patients with severe arteriosclerosis (Grade 3: 79.0% and Grade 4: 81.8%). Furthermore, the findings on arterioles were more frequently associated with the eyes of DR patients with renal dysfunction (p < 0.05); (4) Conclusions: While color SLO images are equally as useful in assessing retinal arteriosclerosis as photography or ophthalmoscopy, the corresponding blue SLO images show arteriosclerotic lesions with high contrast in a monotone background. Retinal arteriosclerosis in eyes of advanced grades or advanced DR frequently show irregularities of retinal arterioles in the blue images. The findings of low, uneven, or discontinuous attenuation were easier to find in blue than in color SLO images. Consequently, blue SLO images can show pathological micro-sclerosis in retinal arterioles and are potentially one of the safe and practical methods for the vascular assessment of diabetic patients.

Multicentric Longitudinal Prospective Study in a European Cohort of MYO7A Patients: Disease Course and Implications for Gene Therapy.

We investigated the natural history of retinal dystrophy owing to variants in the MYO7A gene. Fifty-three patients (mean age, 33.6 ± 16.7years) with Usher syndrome owing to biallelic, mostly pathogenic, variants in MYO7A underwent baseline and two annual follow-up visits. Best-corrected visual acuity (BCVA), semiautomatic kinetic visual field, full-field electroretinogram, color fundus imaging, microperimetry, spectral-domain optical coherence tomography, and fundus autofluorescence were assessed. At baseline, all patients presented with decreased BCVA (66.4 ± 17.9 Early Treatment Diabetic Retinopathy score and 59.5 ± 21.7 Early Treatment Diabetic Retinopathy score, in the better- and worse-seeing eyes, respectively), restricted semiautomatic kinetic visual field (III4e area, 3365.8 ± 4142.1°2; 4176.4 ± 4400.3°2) and decreased macular sensitivity (9.7 ± 9.9dB; 9.0 ± 10.2dB). Spectral-domain optical coherence tomography revealed reduced central macular thickness (259.6 ± 63.0µm; 250.7 ± 63.3µm) and narrowed ellipsoid zone band width (2807.5 ± 2374.6µm; 2615.5 ± 2370.4µm). Longitudinal analyses (50 patients) showed a significant decrease of BCVA in better-seeing eyes, whereas no changes were observed in worse-seeing eyes for any parameter. BCVA, semiautomatic kinetic visual field (III4e and V4e) and macular sensitivity were related significantly to age at baseline. Hyperautofluorescent foveal patch (16 eyes [31.4%]) and abnormal central hypoautofluorescence (9 eyes [17.6%]) were significantly associated with worse morphological and functional read-outs compared with the hyperautofluorescent ring pattern (22 eyes [43.1%]). Our European multicentric study offers the first prospective longitudinal analysis in one of the largest cohorts of MYO7A patients described to date, confirming the slow disease progression. More important, this study emphasizes the key role of fundus autofluorescence patterns in retinal impairment staging and advocates its adoption as an objective biomarker in patient selection for future gene therapy clinical trials.