Ultra-widefield Fundus Images Research Articles

Deep learning system for screening AIDS-related cytomegalovirus retinitis with ultra-wide-field fundus images

BackgroundOphthalmological screening for cytomegalovirus retinitis (CMVR) for HIV/AIDS patients is important to prevent lifelong blindness. Previous studies have shown good properties of automated CMVR screening using digital fundus images. However, the application of a deep learning (DL) system to CMVR with ultra-wide-field (UWF) fundus images has not been studied, and the feasibility and efficiency of this method are uncertain. MethodsIn this study, we developed, internally validated, externally validated, and prospectively validated a DL system to detect AIDS-related from UWF fundus images from different clinical datasets. We independently used the InceptionResnetV2 network to develop and internally validate a DL system for identifying active CMVR, inactive CMVR, and non-CMVR in 6960 UWF fundus images from 862 AIDS patients and validated the system in a prospective and an external validation data set using the area under the curve (AUC), accuracy, sensitivity, and specificity. A heat map identified the most important area (lesions) used by the DL system for differentiating CMVR. ResultsThe DL system showed AUCs of 0.945 (95 % confidence interval [CI]: 0.929, 0.962), 0.964 (95 % CI: 0.870, 0.999) and 0.968 (95 % CI: 0.860, 1.000) for detecting active CMVR from non-CMVR and 0.923 (95 % CI: 0.908, 0.938), 0.902 (0.857, 0.948) and 0.884 (0.851, 0.917) for detecting active CMVR from non-CMVR in the internal cross-validation, external validation, and prospective validation, respectively. Deep learning performed promisingly in screening CMVR. It also showed the ability to differentiate active CMVR from non-CMVR and inactive CMVR as well as to identify active CMVR and inactive CMVR from non-CMVR (all AUCs in the three independent data sets >0.900). The heat maps successfully highlighted lesion locations. ConclusionsOur UWF fundus image-based DL system showed reliable performance for screening AIDS-related CMVR showing its potential for screening CMVR in HIV/AIDS patients, especially in the absence of ophthalmic resources.

Reliability of Clinical Grading of Diabetic Retinopathy Compared to Grading of Ultra-widefield Images

Purpose: Evaluate the reliability of clinical grading of diabetic retinopathy (DR) severity compared to grading on ultra-widefield pseudocolor fundus (UWF-CF) and ultra-widefield fluorescein angiography (UWF-FA) images and their relative detection of sight-threatening DR (STDR) and referable DR. Methods: We analyzed 184 diabetic eyes. UWF-CF and UWF-FA images were graded based on the International Clinical Diabetic Retinopathy (ICDR) severity scale. Agreement between clinical and UWF-based severity grading was evaluated using Cohen’s kappa coefficient. The rate of STDR and referable DR was evaluated for each grading method. Results: We found moderate agreement between clinical grading and UWF-CF (k=.456, p <.001) and between UWF-CF and UWF-FA (k= .443, p <.001). Agreement between clinical grading and UWF-FA was fair (k=.397, p <.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 to clinical grading, UWF-FA detected a higher rate of STDR (44%; 81/184 vs 22.3%; 41/184), while UWF-CF detected more referable eyes (58.1%;107/184 vs 45.65%;84/184). Conclusion: UWF-CF is a valuable tool for identifying referable eyes and can be a useful, non-invasive adjunct to clinical grading. Our results suggest that UWF-FA is particularly useful for detecting unsuspected STDR in eyes with clinically referable DR.

Generalizability of Deep Neural Networks for Vertical Cup-to-Disc Ratio Estimation in Ultra-Widefield and Smartphone-Based Fundus Images.

To develop and validate a deep learning system (DLS) for estimation of vertical cup-to-disc ratio (vCDR) in ultra-widefield (UWF) and smartphone-based fundus images. A DLS consisting of two sequential convolutional neural networks (CNNs) to delineate optic disc (OD) and optic cup (OC) boundaries was developed using 800 standard fundus images from the public REFUGE data set. The CNNs were tested on 400 test images from the REFUGE data set and 296 UWF and 300 smartphone-based images from a teleophthalmology clinic. vCDRs derived from the delineated OD/OC boundaries were compared with optometrists' annotations using mean absolute error (MAE). Subgroup analysis was conducted to study the impact of peripapillary atrophy (PPA), and correlation study was performed to investigate potential correlations between sectoral CDR (sCDR) and retinal nerve fiber layer (RNFL) thickness. The system achieved MAEs of 0.040 (95% CI, 0.037-0.043) in the REFUGE test images, 0.068 (95% CI, 0.061-0.075) in the UWF images, and 0.084 (95% CI, 0.075-0.092) in the smartphone-based images. There was no statistical significance in differences between PPA and non-PPA images. Weak correlation (r = -0.4046, P < 0.05) between sCDR and RNFL thickness was found only in the superior sector. We developed a deep learning system that estimates vCDR from standard, UWF, and smartphone-based images. We also described anatomic peripapillary adversarial lesion and its potential impact on OD/OC delineation. Artificial intelligence can estimate vCDR from different types of fundus images and may be used as a general and interpretable screening tool to improve community reach for diagnosis and management of glaucoma.

Deep Learning-Based Automated Detection of Retinal Breaks and Detachments on Fundus Photography.

The purpose of this study was to develop a deep learning algorithm, to detect retinal breaks and retinal detachments on ultra-widefield fundus (UWF) optos images using artificial intelligence (AI). Optomap UWF images of the database were annotated to four groups by two retina specialists: (1) retinal breaks without detachment, (2) retinal breaks with retinal detachment, (3) retinal detachment without visible retinal breaks, and (4) a combination of groups 1 to 3. The fundus image data set was split into a training set and an independent test set following an 80% to 20% ratio. Image preprocessing methods were applied. An EfficientNet classification model was trained with the training set and evaluated with the test set. A total of 2489 UWF images were included into the dataset, resulting in a training set size of 2008 UWF images and a test set size of 481 images. The classification models achieved an area under the receiver operating characteristic curve (AUC) on the testing set of 0.975 regarding lesion detection, an AUC of 0.972 for retinal detachment and an AUC of 0.913 for retinal breaks. A deep learning system to detect retinal breaks and retinal detachment using UWF images is feasible and has a good specificity. This is relevant for clinical routine as there can be a high rate of missed breaks in clinics. Future clinical studies will be necessary to evaluate the cost-effectiveness of applying such an algorithm as an automated auxiliary tool in a large practices or tertiary referral centers. This study demonstrates the relevance of applying AI in diagnosing peripheral retinal breaks in clinical routine in UWF fundus images.

A new segmentation algorithm for peripapillary atrophy and optic disk from ultra-widefield Photographs

Background and objectiveThe prevalence of myopia and high myopia is increasing globally, underscoring the growing importance of diagnosing high myopia-related pathologies. While existing image segmentation models, such as U-Net, UNet++, ResU-Net, and TransUNet, have achieved significant success in medical image segmentation, they still face challenges when dealing with ultra-widefield (UWF) fundus images. This study introduces a novel automatic segmentation algorithm for the optic disc and peripapillary atrophy (PPA) based on UWF fundus images, aimed at assisting ophthalmologists in more accurately diagnosing high myopia-related diseases. MethodsIn this study, we developed a segmentation model leveraging a Transformer-based network structure, complemented by atrous convolution and selective boundary aggregation modules, to elevate the accuracy of segmenting the optic disc and PPA in UWF photography. The atrous convolution module adeptly manages multi-scale features, catering to the variances in target sizes and expanding the deep network's receptive field. Concurrently, the incorporation of the selective boundary aggregation module in the skip connections of the model significantly improves the differentiation of boundary information between segmentation targets. Moreover, the comparison of our proposed algorithm with classical segmentation models like U-Net, UNet++, ResU-Net, and TransUNet highlights its considerable advantages in processing UWF photographs. ResultsThe experimental results show that, compared to the other four models, our algorithm demonstrates substantial improvements in segmenting the optic disc and PPA in UWF photographs. In PPA segmentation, our algorithm improves by 0.8% in Dice, 1.8% in sensitivity, and 1.3% in intersection over union (IOU). In optic disc segmentation, our algorithm improves by 0.3% in Dice, 0.6% in precision, and 0.4% in IOU. ConclusionOur proposed method improves the segmentation accuracy of PPA and optic disks based on UWF photographs, which is valuable for diagnosing high myopia-related diseases in ophthalmology clinics.

Artificial Intelligence for Retinal Detachment Detection: A Systematic Review of Current Applications

Abstract&#x0D; Introduction &amp; Objectives : Recent development in artificial intelligence has facilitated efficient and automatic detection of retinal detachment, leading to the potential of early intervention and better patient outcomes. This review aims to evaluate the latest advancements in detecting retinal detachment using artificial intelligence, specifically deep learning (DL) systems.&#x0D; Methods : A systematic literature search was performed on five databases: PubMed, Embase, Scopus, ProQuest, and Cochrane. Original research investigations that evaluate deep learning in detecting retinal detachment or features of retinal detachment until February 2023 were included in the study. The outcome of interest were measures of diagnostic accuracy. The review was conducted based on the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.&#x0D; Results : A total of six retrospective studies were included in this review. All studies applied deep learning in ultra-widefield fundus images for automated detection of retinal detachment compared to analysis and diagnosis by human graders (ophthalmologist/retinal specialists) as a reference standard. The deep learning model showed high sensitivity (87.5% - 99.0%), specificity (96.5% - 100%), area under the receiver operating characteristic curve (0.988 - 1.00), and accuracy (97.82% - 99.50%).&#x0D; Conclusion : Artificial intelligence using deep learning systems provided a high diagnostic value in retinal detachment detection. Further integration of artificial intelligence is expected to offer new possibilities in early diagnosis and treatment of retinal detachment.

Artificial intelligence-assisted management of retinal detachment from ultra-widefield fundus images based on weakly-supervised approach.

Retinal detachment (RD) is a common sight-threatening condition in the emergency department. Early postural intervention based on detachment regions can improve visual prognosis. We developed a weakly supervised model with 24,208 ultra-widefield fundus images to localize and coarsely outline the anatomical RD regions. The customized preoperative postural guidance was generated for patients accordingly. The localization performance was then compared with the baseline model and an ophthalmologist according to the reference standard established by the retina experts. In the 48-partition lesion detection, our proposed model reached an 86.42% (95% confidence interval (CI): 85.81-87.01%) precision and an 83.27% (95%CI: 82.62-83.90%) recall with an average precision (PA) of 0.9132. In contrast, the baseline model achieved a 92.67% (95%CI: 92.11-93.19%) precision and limited recall of 68.07% (95%CI: 67.25-68.88%). Our holistic lesion localization performance was comparable to the ophthalmologist's 89.16% (95%CI: 88.75-89.55%) precision and 83.38% (95%CI: 82.91-83.84%) recall. As to the performance of four-zone anatomical localization, compared with the ground truth, the un-weighted Cohen's κ coefficients were 0.710(95%CI: 0.659-0.761) and 0.753(95%CI: 0.702-0.804) for the weakly-supervised model and the general ophthalmologist, respectively. The proposed weakly-supervised deep learning model showed outstanding performance comparable to that of the general ophthalmologist in localizing and outlining the RD regions. Hopefully, it would greatly facilitate managing RD patients, especially for medical referral and patient education.

Deep Learning Detection of Early Retinal Peripheral Degeneration From Ultra-Widefield Fundus Photographs of Asymptomatic Young Adult (17-19 Years) Candidates to Airforce Cadets.

Artificial intelligence (AI)-assisted ultra-widefield (UWF) fundus photographic interpretation is beneficial to improve the screening of fundus abnormalities. Therefore we constructed an AI machine-learning approach and performed preliminary training and validation. We proposed a two-stage deep learning-based framework to detect early retinal peripheral degeneration using UWF images from the Chinese Air Force cadets' medical selection between February 2016 and June 2022. We developed a detection model for the localization of optic disc and macula, which are used to find the peripheral areas. Then we developed six classification models for the screening of various retinal cases. We also compared our proposed framework with two baseline models reported in the literature. The performance of the screening models was evaluated by area under the receiver operating curve (AUC) with 95% confidence interval. A total of 3911 UWF fundus images were used to develop the deep learning model. The external validation included 760 UWF fundus images. The results of comparison study revealed that our proposed framework achieved competitive performance compared to existing baselines while also demonstrating significantly faster inference time. The developed classification models achieved an average AUC of 0.879 on six different retinal cases in the external validation dataset. Our two-stage deep learning-based framework improved the machine learning efficiency of the AI model for fundus images with high resolution and many interference factors by maximizing the retention of valid information and compressing the image file size. This machine learning model may become a new paradigm for developing UWF fundus photography AI-assisted diagnosis.

Applications of deep learning for detecting ophthalmic diseases with ultrawide-field fundus images.

To summarize the application of deep learning in detecting ophthalmic disease with ultrawide-field fundus images and analyze the advantages, limitations, and possible solutions common to all tasks. We searched three academic databases, including PubMed, Web of Science, and Ovid, with the date of August 2022. We matched and screened according to the target keywords and publication year and retrieved a total of 4358 research papers according to the keywords, of which 23 studies were retrieved on applying deep learning in diagnosing ophthalmic disease with ultrawide-field images. Deep learning in ultrawide-field images can detect various ophthalmic diseases and achieve great performance, including diabetic retinopathy, glaucoma, age-related macular degeneration, retinal vein occlusions, retinal detachment, and other peripheral retinal diseases. Compared to fundus images, the ultrawide-field fundus scanning laser ophthalmoscopy enables the capture of the ocular fundus up to 200° in a single exposure, which can observe more areas of the retina. The combination of ultrawide-field fundus images and artificial intelligence will achieve great performance in diagnosing multiple ophthalmic diseases in the future.

RELATIONSHIP BETWEEN RETINAL HEMORRHAGE ON GREEN AND RED CHANNELS OF ULTRA-WIDEFIELD FUNDUS IMAGES AND RETINAL PERFUSION IN ACUTE BRANCH RETINAL VEIN OCCLUSION.

To explore the relationship between retinal hemorrhage in the green and red channels on ultra-widefield fundus images and the nonperfusion area (NPA) on ultra-widefield fundus fluorescein angiography in patients with acute branch retinal vein occlusion (BRVO). This was a retrospective cross-sectional study with 96 patients, including 46 with ischemic BRVO and 50 with nonischemic BRVO. Correlation analysis between green channel hemorrhage (GCH), red channel hemorrhage (RCH), and NPA was performed. Panretina was divided into posterior and peripheral areas. Ischemic BRVO showed significantly higher GCH% and RCH% than nonischemic BRVO in the peripheral regions (both P < 0.001), whereas no significant differences were observed in the panretinal and posterior areas (all P > 0.05). Significant correlations were found between NPA% in the panretinal and peripheral areas and the corresponding GCH% and RCH% (all P < 0.01). However, no significant correlation was observed between posterior NPA% and posterior GCH% or RCH% (both P > 0.05). In addition, peripheral GCH% and RCH% were related to panretinal NPA% (r = 0.506, P < 0.001; r = 0.558, P < 0.001). Retinal hemorrhage on ultra-widefield fundus image was significantly associated with NPA, providing insights for assessing retinal perfusion status in acute BRVO patients.

Initiation of China Alliance of Research in High Myopia (CHARM): protocol for an AI-based multimodal high myopia research biobank

IntroductionHigh myopia is a pressing public health concern due to its increasing prevalence, younger trend and the high risk of blindness, particularly in East Asian countries, including China. The China...

Prevalence of peripheral retinal findings in retinal patients using ultra-widefield pseudocolor fundus imaging

Ultra-widefield retinal imaging is increasingly used in ophthalmology and optometry practices to image patients identifying peripheral abnormalities. However, the clinical relevance of these peripheral retinal abnormalities is unclear. This cross-sectional study aims to firstly validate a new grading system, secondly, assess the prevalence of peripheral retinal abnormalities in retinal patients, and finally understand how peripheral findings may associate with retinal disease. Ultra-widefield pseudocolor fundus images were taken from the eyes of clinic patients. Demographic data and clinical diagnosis for each patient was noted. The grading system was validated using masked retinal specialists. Logistic regression identified associations between retinal disease and peripheral retinal findings. Using the grading system, inter-observer agreement was 76.1% with Cohen’s Kappa coefficient 0.542 (p < 0.0001) and the test–retest agreement was 95.1% with Kappa 0.677(p < 0.0001). 971 images were included, with 625 eyes (64.4%) having peripheral abnormalities. Peripheral drusen was the most common abnormality (n = 221, 22.76%) and correlated with age-related macular degeneration (p < 0.001). Novel correlations were also identified between diabetic retinopathy and retinal pigmentation as well as pigmentary degeneration. This study provides a validated system for identifying peripheral abnormalities and adds to literature highlighting peripheral retinal associations with retinal disease which would benefit from further study.

Self-FI: Self-Supervised Learning for Disease Diagnosis in Fundus Images.

Self-supervised learning has been successful in computer vision, and its application to medical imaging has shown great promise. This study proposes a novel self-supervised learning method for medical image classification, specifically targeting ultra-wide-field fundus images (UFI). The proposed method utilizes contrastive learning to pre-train a deep learning model and then fine-tune it with a small set of labeled images. This approach reduces the reliance on labeled data, which is often limited and costly to obtain, and has the potential to improve disease detection in UFI. This method employs two contrastive learning techniques, namely bi-lateral contrastive learning and multi-modality pre-training, to form positive pairs using the data correlation. Bi-lateral learning fuses multiple views of the same patient's images, and multi-modality pre-training leverages the complementary information between UFI and conventional fundus images (CFI) to form positive pairs. The results show that the proposed contrastive learning method achieves state-of-the-art performance with an area under the receiver operating characteristic curve (AUC) score of 86.96, outperforming other approaches. The findings suggest that self-supervised learning is a promising direction for medical image analysis, with potential applications in various clinical settings.

Joint conditional generative adversarial networks for eyelash artifact removal in ultra-wide-field fundus images.

Background: Ultra-Wide-Field (UWF) fundus imaging is an essential diagnostic tool for identifying ophthalmologic diseases, as it captures detailed retinal structures within a wider field of view (FOV). However, the presence of eyelashes along the edge of the eyelids can cast shadows and obscure the view of fundus imaging, which hinders reliable interpretation and subsequent screening of fundus diseases. Despite its limitations, there are currently no effective methods or datasets available for removing eyelash artifacts from UWF fundus images. This research aims to develop an effective approach for eyelash artifact removal and thus improve the visual quality of UWF fundus images for accurate analysis and diagnosis. Methods: To address this issue, we first constructed two UWF fundus datasets: the paired synthetic eyelashes (PSE) dataset and the unpaired real eyelashes (uPRE) dataset. Then we proposed a deep learning architecture called Joint Conditional Generative Adversarial Networks (JcGAN) to remove eyelash artifacts from UWF fundus images. JcGAN employs a shared generator with two discriminators for joint learning of both real and synthetic eyelash artifacts. Furthermore, we designed a background refinement module that refines background information and is trained with the generator in an end-to-end manner. Results: Experimental results on both PSE and uPRE datasets demonstrate the superiority of the proposed JcGAN over several state-of-the-art deep learning approaches. Compared with the best existing method, JcGAN improves PSNR and SSIM by 4.82% and 0.23%, respectively. In addition, we also verified that eyelash artifact removal via JcGAN could significantly improve vessel segmentation performance in UWF fundus images. Assessment via vessel segmentation illustrates that the sensitivity, Dice coefficient and area under curve (AUC) of ResU-Net have respectively increased by 3.64%, 1.54%, and 1.43% after eyelash artifact removal using JcGAN. Conclusion: The proposed JcGAN effectively removes eyelash artifacts in UWF images, resulting in improved visibility of retinal vessels. Our method can facilitate better processing and analysis of retinal vessels and has the potential to improve diagnostic outcomes.

Automatic Detection of Peripheral Retinal Lesions From Ultrawide-Field Fundus Images Using Deep Learning.

To establish a multilabel-based deep learning (DL) algorithm for automatic detection and categorization of clinically significant peripheral retinal lesions using ultrawide-field fundus images. A total of 5958 ultrawide-field fundus images from 3740 patients were randomly split into a training set, validation set, and test set. A multilabel classifier was developed to detect rhegmatogenous retinal detachment, cystic retinal tuft, lattice degeneration, and retinal breaks. Referral decision was automatically generated based on the results of each disease class. t -distributed stochastic neighbor embedding heatmaps were used to visualize the features extracted by the neural networks. Gradient-weighted class activation mapping and guided backpropagation heatmaps were generated to investigate the image locations for decision-making by the DL models. The performance of the classifier(s) was evaluated by sensitivity, specificity, accuracy, F 1 score, area under receiver operating characteristic curve (AUROC) with 95% CI, and area under the precision-recall curve. In the test set, all categories achieved a sensitivity of 0.836-0.918, a specificity of 0.858-0.989, an accuracy of 0.854-0.977, an F 1 score of 0.400-0.931, an AUROC of 0.9205-0.9882, and an area under the precision-recall curve of 0.6723-0.9745. The referral decisions achieved an AUROC of 0.9758 (95% CI= 0.9648-0.9869). The multilabel classifier had significantly better performance in cystic retinal tuft detection than the binary classifier (AUROC= 0.9781 vs 0.6112, P < 0.001). The model showed comparable performance with human experts. This new DL model of a multilabel classifier is capable of automatic, accurate, and early detection of clinically significant peripheral retinal lesions with various sample sizes. It can be applied in peripheral retinal screening in clinics.

Is Tumour Thickness Measurement Required for MOLES Scoring of Melanocytic Choroidal Tumours?

Introduction: It can be challenging to distinguish between choroidal naevi and melanomas in the community setting, particularly without access to ultrasonography (US), required to measure the thickness of melanocytic choroidal tumours. We aimed to determine whether thickness measurement is required for MOLES scoring of melanocytic choroidal tumours. Methods: The dataset of a recent MOLES evaluation was reviewed. Patients were selected for the present study if their MOLES tumour size category was determined by tumour thickness measured with US. The largest basal tumour diameter and tumour thickness were then measured from ultra-widefield fundus images and optical coherence tomography (OCT) images, respectively. Results: The tumour size category was determined by tumour diameter in 203/222 (91.4%) with no influence of tumour thickness. The tumour thickness influenced the MOLES score in 19/222 (8.6%) patients. In 11/19 patients with OCT measurements of tumour thickness, the US measurement exceeded the OCT by more than 25% in 5 patients, more than 50% in 2 patients, and more than 75% in 1 patient. As a result, the revised tumour thickness based on OCT determined the size category in 4/216 (1.8%) patients. The ultra-widefield fundus images measurements increased the diameter score by 1 in 5 patients. As a result, the revised tumour thickness determined the size category in 4/216 (1.8%) patients. If both the revised diameter and thickness scores were considered, the MOLES score reduced in 4 patients. If both the diameter and thickness scores were considered, the MOLES score reduced in 5 and increased in 1. Only 0.94% (2/211) of melanocytic choroidal tumours assessed with MOLES when using Optos ultra-widefield fundus images diameter and OCT to measure tumour diameter and thickness, respectively, required a change in management from a reduction in MOLES score from 1 to 0. Discussion/Conclusion: This study suggests that the MOLES category for size is influenced more by the tumour diameter, if it can be measured accurately, than by the thickness. This study suggests ignoring tumour thickness if this cannot be measured accurately with OCT, unless the tumour has a mushroom shape.

A new ultra-wide-field fundus dataset to diabetic retinopathy grading using hybrid preprocessing methods

Diabetic retinopathy(DR) is a common early diabetic complication and one of the main causes of blindness. In clinical diagnosis and treatment, regular screening with fundus imaging is an effective way to prevent the development of DR. However, the regular fundus images used in most DR screening work have a small imaging range, narrow field of vision, and can not contain more complete lesion information, which leads to less ideal automatic DR grading results. In order to improve the accuracy of DR grading, we establish a dataset containing 101 ultra-wide-field(UWF) DR fundus images and propose a deep learning(DL) automatic classification method based on a new preprocessing method. The emerging UWF fundus images have the advantages of a large imaging range and wide field of vision and contain more information about the lesions. In data preprocessing, we design a data denoising method for UWF images and use data enhancement methods to improve their contrast and brightness to improve the classification effect. In order to verify the efficiency of our dataset and the effectiveness of our preprocessing method, we design a series of experiments including a variety of DL classification models. The experimental results show that we can achieve high classification accuracy by using only the backbone model. The most basic ResNet50 model reaches an average of classification accuracy(ACA) 0.66, Macro F1 0.6559, and Kappa 0.58. The best-performing Swin-S model reaches ACA 0.72, Macro F1 0.7018, and Kappa 0.65. DR grading using UWF images can achieve higher accuracy and efficiency, which has practical significance and value in clinical applications.

Estimation of Visual Function Using Deep Learning From Ultra-Widefield Fundus Images of Eyes With Retinitis Pigmentosa

There is no widespread effective treatment to halt the progression of retinitis pigmentosa. Consequently, adequate assessment and estimation of residual visual function are important clinically. To examine whether deep learning can accurately estimate the visual function of patients with retinitis pigmentosa by using ultra-widefield fundus images obtained on concurrent visits. Data for this multicenter, retrospective, cross-sectional study were collected between January 1, 2012, and December 31, 2018. This study included 695 consecutive patients with retinitis pigmentosa who were examined at 5 institutions. Each of the 3 types of input images-ultra-widefield pseudocolor images, ultra-widefield fundus autofluorescence images, and both ultra-widefield pseudocolor and fundus autofluorescence images-was paired with 1 of the 31 types of ensemble models constructed from 5 deep learning models (Visual Geometry Group-16, Residual Network-50, InceptionV3, DenseNet121, and EfficientNetB0). We used 848, 212, and 214 images for the training, validation, and testing data, respectively. All data from 1 institution were used for the independent testing data. Data analysis was performed from June 7, 2021, to December 5, 2022. The mean deviation on the Humphrey field analyzer, central retinal sensitivity, and best-corrected visual acuity were estimated. The image type-ensemble model combination that yielded the smallest mean absolute error was defined as the model with the best estimation accuracy. After removal of the bias of including both eyes with the generalized linear mixed model, correlations between the actual values of the testing data and the estimated values by the best accuracy model were examined by calculating standardized regression coefficients and P values. The study included 1274 eyes of 695 patients. A total of 385 patients were female (55.4%), and the mean (SD) age was 53.9 (17.2) years. Among the 3 types of images, the model using ultra-widefield fundus autofluorescence images alone provided the best estimation accuracy for mean deviation, central sensitivity, and visual acuity. Standardized regression coefficients were 0.684 (95% CI, 0.567-0.802) for the mean deviation estimation, 0.697 (95% CI, 0.590-0.804) for the central sensitivity estimation, and 0.309 (95% CI, 0.187-0.430) for the visual acuity estimation (all P < .001). Results of this study suggest that the visual function estimation in patients with retinitis pigmentosa from ultra-widefield fundus autofluorescence images using deep learning might help assess disease progression objectively. Findings also suggest that deep learning models might monitor the progression of retinitis pigmentosa efficiently during follow-up.

Intelligent Diagnosis of Multiple Peripheral Retinal Lesions in Ultra-widefield Fundus Images Based on Deep Learning.

Compared with traditional fundus examination techniques, ultra-widefield fundus (UWF) images provide 200° panoramic images of the retina, which allows better detection of peripheral retinal lesions. The advent of UWF provides effective solutions only for detection but still lacks efficient diagnostic capabilities. This study proposed a retinal lesion detection model to automatically locate and identify six relatively typical and high-incidence peripheral retinal lesions from UWF images which will enable early screening and rapid diagnosis. A total of 24,602 augmented ultra-widefield fundus images with labels corresponding to 6 peripheral retinal lesions and normal manifestation labelled by 5 ophthalmologists were included in this study. An object detection model named You Only Look Once X (YOLOX) was modified and trained to locate and classify the six peripheral retinal lesions including rhegmatogenous retinal detachment (RRD), retinal breaks (RB), white without pressure (WWOP), cystic retinal tuft (CRT), lattice degeneration (LD), and paving-stone degeneration (PSD). We applied coordinate attention block and generalized intersection over union (GIOU) loss to YOLOX and evaluated it for accuracy, sensitivity, specificity, precision, F1 score, and average precision (AP). This model was able to show the exact location and saliency map of the retinal lesions detected by the model thus contributing to efficient screening and diagnosis. The model reached an average accuracy of 96.64%, sensitivity of 87.97%, specificity of 98.04%, precision of 87.01%, F1 score of 87.39%, and mAP of 86.03% on test dataset 1 including 248 UWF images and reached an average accuracy of 95.04%, sensitivity of 83.90%, specificity of 96.70%, precision of 78.73%, F1 score of 81.96%, and mAP of 80.59% on external test dataset 2 including 586 UWF images, showing this system performs well in distinguishing the six peripheral retinal lesions. Focusing on peripheral retinal lesions, this work proposed a deep learning model, which automatically recognized multiple peripheral retinal lesions from UWF images and localized exact positions of lesions. Therefore, it has certain potential for early screening and intelligent diagnosis of peripheral retinal lesions.

Retinal haemorrhages on ultra-widefield red channel images and perfusion status in central retinal vein occlusion.

To evaluate the relationship between retinal haemorrhages detected on Ultra-widefield (UWF) red channel images and perfusion status in eyes with acute central retinal vein occlusion (CRVO). UWF fundus images were split into green and red channels using ImageJ software. The retinal haemorrhages were calculated quantitatively with both the green and red channel images, resulting in green channel haemorrhages (GCH) and red channel haemorrhages (RCH). The nonperfusion area (NPA) was also calculated from fluorescein angiography in each eye. The relationships between both the GCH and RCH with the NPA were investigated. Thirty-two eyes of 32 patients with acute CRVO (18 men, 14 women) were included. The mean GCH and RCH values were 10.4% ± 8.2% and 1.7% ± 1.7%, respectively. The mean NPA was 39.2% ± 28.8%. Significant correlations were seen between the GCH and NPA (r = 0.38; P = 0.022) and RCH and NPA (r = 0.44; P = 0.010, linear regression analysis). Multivariate analysis suggested that only the RCHs were correlated significantly with the NPA. Retinal haemorrhages detected by UWF red channel imaging were less compared to green channel imaging and associated closely with retinal NPAs in eyes with acute CRVO. UWF red channel imaging allowed us to identify ischaemia-related haemorrhage.