Ultra-widefield Imaging Research Articles

COMBINED APPLICATION OF B-SCAN ULTRASONOGRAPHY AND EYE-STEERING ULTRAWIDE FIELD IMAGING TO IMPROVE THE DETECTION OF RETINAL TEARS BEFORE CATARACT SURGERY.

To investigate the efficacy of combined application of B-scan ultrasonography (US) and ultrawide field imaging (UWFI) in detecting retinal tears before cataract surgery. Of 1,277 cataract patients, 2,552 eyes were enrolled and received both B-scan US and UWFI examinations preoperatively. Three types of combination were applied: type 1 (union, B-scan US or centered UWFI), type 2 (intersection, B-scan US and centered UWFI), and type 3 (B-scan US and eye-steering UWFI). Sensitivity and specificity of detecting retinal tears by different methods were assessed. Totally 4.55% (116/2,552) of eyes were presented with retinal tears. The sensitivity of B-scan US and UWFI was 87.93% and 84.48%, and specificity was 95.16% and 99.79%, respectively. By applying type 1 and type 2 combination, the sensitivity was 98.28% and 74.14%, and specificity was 95.03% and 99.92%, respectively. By type 3 combination, the sensitivity increased to 95.69% and specificity to 99.88%, both of which were comparable to indirect ophthalmoscopy regardless of the number, type, and location of tears ( P > 0.05). In eyes with any cataract type or axial length, type 3 combination also gained comparable performance to indirect ophthalmoscopy. Combined application of B-scan US and eye-steering UWFI presented satisfactory performance in detecting retinal tears before cataract surgery.

Peripheral retinal lesions in diabetic retinopathy on ultra-widefield imaging.

Peripheral retinal imaging plays a crucial role in the diagnosis, management, and prognosis of diabetic retinopathy (DR). Traditional fundus imaging techniques have limited coverage of the retina, resulting in missed peripheral lesions. The advent of ultra-widefield (UWF) imaging has revolutionized the assessment of the peripheral retina. UWF imaging modalities provide comprehensive visualization of the retina, enabling the detection of peripheral lesions without the need for mydriasis. Integration of UWF imaging with other modalities, including fluorescein angiography (FA), indocyanine green angiography, pseudocolor imaging, and fundus autofluorescence, further enhances our understanding of peripheral retinal lesions. UWF imaging has demonstrated improved detection of DR lesions and presumably more accurate management of DR compared to traditional fundus photography and dilated fundus examination. UWF-FA and UWF-optical coherence tomography angiography have emerged as valuable tools for assessing retinal and choroidal vascular abnormalities, nonperfusion areas, neovascularization, and microvascular abnormalities. The presence and increasing extent of predominantly peripheral lesions detected using UWF FA are associated with a higher risk of DR progression and proliferative DR. UWF imaging provides a comprehensive evaluation of DR severity, aiding in more accurate risk stratification and treatment decision-making. Overall, UWF imaging modalities have significantly advanced our understanding of peripheral retinal lesions in DR, facilitating early detection and targeted management for better visual outcomes.

Silicone oil emulsification: A literature review and role of widefield imaging and ultra-widefield imaging with navigated central and peripheral optical coherence tomography technology.

Silicone oil (SO) emulsification is a significant concern in vitreoretinal surgery, leading to various complications. Despite the high prevalence of SO emulsification within the eye, there is currently no standardized method for its early detection. The recent introduction of widefield (WF) imaging and ultra-WF (UWF) imaging with navigated central and peripheral optical coherence tomography (OCT) techniques have shown promising results in providing high-resolution images of the peripheral vitreous, vitreoretinal interface, retina, and choroid. This enhanced visualization capability enables the early identification of emulsified SO droplets, facilitating a proactive therapeutic approach, and mitigating associated adverse events. This comprehensive literature review aims to provide an updated overview of the topic, focusing on the role of WFimaging and UWF imaging and navigated central and peripheral swept-source OCT (SS-OCT) in the early detection and management of SO emulsification. The review discusses the current understanding of SO emulsification, its associated complications, and the limitations of existing detection methods. In addition, it highlights the potential of WF and UWF imaging and peripheral OCT as advanced imaging modalities for improved visualization of SO emulsification. This review serves as a valuable resource for clinicians and researchers, providing insights into the latest advancements in the field of vitreoretinal surgery and the promising role of WF imaging and UWF imaging and navigated central and peripheral SS-OCT in the management of SO.

Impact of Ultra-widefield Imaging on Understanding the Pathophysiology of Peripheral Retinal Degeneration

Ultra-widefield imaging has revolutionized the understanding of the retinal periphery in clinical practice. Since the publication of the study “Comparison of image-assisted versus traditional fundus examination,” the practice of using fundus imaging to guide retinal examination has become ubiquitous, not only by experienced doctors in practice, but in the training of new clinicians.1 Quick capture and review of micron-level views with various wavelengths allows clinicians to plan their fundus exam, focus their time on areas of suspicion and better understand the health of the entire retina more effectively and efficiently. Image-assisted ophthalmoscopy has excellent agreement for a variety of lesion types but confers an advantage in optic nerve assessment, review of small lesions in the mid and far periphery and in visualization of vitreous lesions. Increasing resolution, integration of new wavelengths, auto-montage, optical coherence tomography and artificial intelligence have improved the sensitivity of this tool - thus improving the overall efficiency of a fundus exam, improving outcomes and leading to more accurate assessment of risk. Advances in ultra-widefield imaging technology have allowed us to better understand the anatomy and physiology of the peripheral retina and its relation to sight threatening disease; making this an invaluable tool to educate our students in training as well as our patients with and without retinal pathology in the clinic. When an image or lesion of concern can be viewed simultaneously by student and instructor or by doctor and patient, the learning and understanding from this ultra-widefield perspective are invaluable. In this review, we explore the technological advances and the impact on detection, diagnosis and prognostic potential in a variety of retinal degenerations. In addition, we examine remaining challenges of the technology.

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.

Ultra-wide field and new wide field composite retinal image registration with AI-enabled pipeline and 3D distortion correction algorithm

PurposeThis study aimed to compare a new Artificial Intelligence (AI) method to conventional mathematical warping in accurately overlaying peripheral retinal vessels from two different imaging devices: confocal scanning laser ophthalmoscope (cSLO) wide-field images and SLO ultra-wide field images.MethodsImages were captured using the Heidelberg Spectralis 55-degree field-of-view and Optos ultra-wide field. The conventional mathematical warping was performed using Random Sample Consensus—Sample and Consensus sets (RANSAC-SC). This was compared to an AI alignment algorithm based on a one-way forward registration procedure consisting of full Convolutional Neural Networks (CNNs) with Outlier Rejection (OR CNN), as well as an iterative 3D camera pose optimization process (OR CNN + Distortion Correction [DC]). Images were provided in a checkerboard pattern, and peripheral vessels were graded in four quadrants based on alignment to the adjacent box.ResultsA total of 660 boxes were analysed from 55 eyes. Dice scores were compared between the three methods (RANSAC-SC/OR CNN/OR CNN + DC): 0.3341/0.4665/4784 for fold 1-2 and 0.3315/0.4494/4596 for fold 2-1 in composite images. The images composed using the OR CNN + DC have a median rating of 4 (out of 5) versus 2 using RANSAC-SC. The odds of getting a higher grading level are 4.8 times higher using our OR CNN + DC than RANSAC-SC (p < 0.0001).ConclusionPeripheral retinal vessel alignment performed better using our AI algorithm than RANSAC-SC. This may help improve co-localizing retinal anatomy and pathology with our algorithm.

Development and validation of a deep learning model to predict axial length from ultra-wide field images.

To validate the feasibility of building a deep learning model to predict axial length (AL) for moderate to high myopic patients from ultra-wide field (UWF) images. This study included 6174 UWF images from 3134 myopic patients during 2014 to 2020 in Eye and ENT Hospital of Fudan University. Of 6174 images, 4939 were used for training, 617 for validation, and 618 for testing. The coefficient of determination (R2), mean absolute error (MAE), and mean squared error (MSE) were used for model performance evaluation. The model predicted AL with high accuracy. Evaluating performance of R2, MSE and MAE were 0.579, 1.419 and 0.9043, respectively. Prediction bias of 64.88% of the tests was under 1-mm error, 76.90% of tests was within the range of 5% error and 97.57% within 10% error. The prediction bias had a strong negative correlation with true AL values and showed significant difference between male and female (P < 0.001). Generated heatmaps demonstrated that the model focused on posterior atrophy changes in pathological fundus and peri-optic zone in normal fundus. In sex-specific models, R2, MSE, and MAE results of the female AL model were 0.411, 1.357, and 0.911 in female dataset and 0.343, 2.428, and 1.264 in male dataset. The corresponding metrics of male AL models were 0.216, 2.900, and 1.352 in male dataset and 0.083, 2.112, and 1.154 in female dataset. It is feasible to utilize deep learning models to predict AL for moderate to high myopic patients with UWF images.

Genetic and clinical characteristics of PROM1-related retinal degeneration in Korean

This scientific report aims to comprehensively describe the genetic and clinical characteristics of PROM1-related retinal degeneration in Korean patients. Medical records of patients diagnosed with retinal dystrophy who underwent comprehensive ophthalmologic examination and genetic testing at Samsung Medical Center between January 2016 and April 2023 were retrospectively reviewed. Genetic testing included targeted gene panel sequencing and Sanger sequencing, with diagnosis based on the presence of a “Likely Pathogenic” or “Pathogenic Variant” in the PROM1 gene, as determined by the ACMG criteria. The study identified seven patients from five unrelated families with PROM1-related retinal degeneration, all carrying the autosomal dominant variant PROM1 p.R373C; no other PROM1 gene variants were detected. All patients exhibited degenerative retinal area within the macula, with peripheral retinal degeneration observed in five patients. Substantial interfamilial and intrafamilial variability was observed in the extent of macular and peripheral degeneration. Ultra-widefield autofluorescence imaging and fluorescein angiography aided in the detection of mild peripheral degeneration in one case. In conclusion, the autosomal dominant variant PROM1 p.R373C constitutes a significant proportion of PROM1-related retinal degeneration cases in the Korean population. The observed clinical heterogeneity may suggests the potential influence of additional genetic, epigenetic, and environmental factors on disease phenotypes.

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.

Evaluation of optical coherence tomography biomarkers to differentiate favourable and unfavourable responders to intravitreal anti-vascular endothelial growth factor treatment in retinopathy of prematurity.

Evaluation of optical coherence tomography biomarkers in predicting treatment response to intravitreal injection of anti-vascular endothelial growth factor (anti-VEGF) Bevacizumab, in aggressive retinopathy of prematurity (A-ROP). Non-contact ultra-widefield (NC-UWF) fundus imaging with integrated UWF guided swept source Optical coherence tomography (SS-OCT) was performed prospectively in preterm babies before and after intravitreal anti-VEGF (Bevacizumab) monotherapy. OCT biomarkers were analysed in eyes that reached complete vascularization versus others. Eyes with retinal vessels reaching near ora serrata were labelled as regressed ROP and vascularised retina (Group1). Eyes with reactivation of ROP needing laser or vitreoretinal surgery or eyes with peripheral avascular retina (PAR) at 16th week post-injection were considered as Group 2. Pre-injection baseline OCT showed a hyperreflectivity of inner retinal layers in 12 out of 46 eyes in Group 1 versus 30 out of 34 eyes in Group 2 (p value 0.002). None of the eyes in Group 1 showed choroidal thinning at posterior pole as compared to 14 out of 34 eyes in Group 2 (p value 0.001). Intraretinal hypo reflective Cystic changes at fovea were seen in 16 out of 46 eyes in Group 1 and 2 out of 34 eyes in Group 2 (p value 0.012). Pre-injection swept source OCT biomarkers could predict the treatment outcomes of anti-VEGF (Bevacizumab) monotherapy in A-ROP eyes. Hyperreflectivity of inner retinal layers and choroidal thinning had poorer and unpredictable response to anti-VEGF injection whereas, cystic changes at fovea predicted favourable response.

Comparison of Standard 7-Field, Clarus, and Optos Ultrawidefield Imaging Systems for Diabetic Retinopathy (COCO Study)

The purpose of this study was to compare diabetic retinopathy (DR) severity levels assessed from 7 standard-field stereoscopic color photographs on a 35° fundus camera to both Clarus and Optos ultrawidefield color images. Cross-sectional, comparative imaging study. Participants with DR imaged at a single-center retina practice. Participants were imaged on 3 cameras at a single visit with the Topcon 35° fundus camera, Clarus, and Optos. The DR Severity Scale (DRSS) level was determined within the 7-field (7F) area of each image set using the ETDRS scale. An additional global DRSS was assigned for both Clarus and Optos images using the entire visible retina. Weighted kappa (wκ) measured the agreement between cameras. The primary outcome was a 3-way comparison of DRSS level within the 7F area imaged on the 3 cameras. Secondary outcomes included a comparison of the DRSS obtained with standard 7F imaging to the global DRSS of Clarus and Optos and a comparison of the global DRSS between Clarus and Optos only. Ninety-seven eyes (50 participants) were evaluated. Agreement within 1-step of ETDRS levels between standard 7F imaging and Clarus 7F was 90.1% (wκ=0.65), and with Optos 7F in 85.9%, (wκ=0.58). Agreement within 1-step between standard 7F imaging and Clarus global was 88.9% of eyes (wκ=0.63), and Optos global was 85.7%, (wκ=0.54). Agreement between Clarus and Optos global DR level within 1-step was 89.1% (wκ=0.68). Intergrader agreement for the 7F ETDRS level was 96% for standard 7F imaging, 98% for Clarus, and 95.5% for Optos. These findings suggest that when evaluating the 7F area on Clarus and Optos, DR severity grades are comparable to standard 7F imaging. However, it is important to understand the unique attributes and differences of each fundus camera when changing the type of system used in a clinical setting due to upgrading equipment. Additionally, if the facility has access to > 1 device, there should not be an exchange between cameras for the same patient. Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Purtscher-like retinopathy induced by diffuse cutaneous systemic scleroderma.

Abstract We report a rare case of Purtscher-like retinopathy related to systemic sclerosis, presented unprecedently an ultra-widefield optical coherence tomography angiography image for this combination, and showed not only retinal but also choroidal vascular nonperfusion.

Vascular abnormalities in fellow eyes of patients with unilateral coats’ disease

Coats’ disease is an idiopathic retinal vascular disorder, known to usually occur unilaterally; however, recent studies have highlighted vascular abnormalities in the fellow unaffected eyes. This retrospective study investigated the peripheral vascular features and macular vascular structure of unaffected fellow eyes in patients with unilateral Coats’ disease using multimodal imaging tools. We analysed images of patients, including bilateral ultra-widefield imaging, fluorescein angiography (FA), ultra-widefield FA, or standard fundus photography. Available bilateral optical coherence tomography angiography (OCT-A) images were used for macular vascular structure analysis. OCT-A parameters, including foveal avascular zone (FAZ), perfusion index, and vessel density (VD) in the superficial and deep capillary plexuses (SCP, DCP), were calculated using Image J software. The mean age at diagnosis was 34.5 ± 17.9 years. The mean final best-corrected visual acuity of the affected eyes was logMAR 0.78 ± 0.79, while that of the fellow eyes was logMAR 0.04 ± 0.12. Ten fellow eyes had microaneurysms (47.6%), two had tortuous vessel abnormalities (9.5%), and 11(52.4%) had abnormal vascular findings on FA. Although there was a trend towards larger DCP FAZ (1.201 ± 0.086 vs. 1.072 ± 0.226), and lower DCP VD (8.593 ± 1.583 vs. 10.827 ± 3.392) in the affected eyes as measured by the Cirrus machine, the difference was not statistically significant between affected and fellow eyes when measured using the Zeiss Cirrus machine (P = 0.686, P = 0.343, respectively). However, when measured with the Spectralis machine, DCP FAZ was larger in affected eyes (0.828 ± 0.426 vs. 0.254 ± 0.092, P = 0.002) and DCP VD was lower in affected eyes (6.901 ± 2.634 vs. 17.451 ± 7.207, P = 0.002) compared to the fellow eyes, while other parameters showed no significant variations. These findings indicate that there may be subtle vascular abnormalities primarily located in the peripheral regions of the unaffected fellow eyes in patients with unilateral Coats’ disease, while the macular microvasculature remains unaffected.

Latest advances in white spot syndromes: New findings and interpretations.

White spot syndromes (WSS) pose challenges in the field of ophthalmology, particularly in terms of accurate diagnosis and effective management. However, recent advancements in multimodal imaging (MMI) have significantly contributed to our understanding of WSS, allowing for improved characterization of these inflammatory chorioretinopathies. By employing various imaging modalities, including fundus fluorescein angiography, indocyanine green angiography, fundus autofluorescence, optical coherence tomography (OCT), ultra-widefield imaging, and OCT angiography, researchers and clinicians have gained valuable insights into the underlying pathophysiological changes and clinical progression of WSS. Furthermore, MMI has unveiled novel and atypical variants within the spectrum of WSS, expanding our knowledge in this field. Notably, the identification of secondary forms of WSS occurring concurrently with unrelated chorioretinal disorders has suggested a potential autoimmune mechanism underlying these conditions. The introduction of MMI has also facilitated a more comprehensive evaluation of previously ill-defined entities, such as acute zonal occult outer retinopathy, leading to improved diagnostic criteria and enhanced recognition of distinct features. This review paper provides a comprehensive overview of the latest advances and interpretations in WSS. By integrating MMI into the diagnosis and management of these conditions, this review aims to enhance patient outcomes and provide valuable insights into the complexities surrounding WSS.

Assessment of early diabetic retinopathy severity using ultra-widefield Clarus versus conventional five-field and ultra-widefield Optos fundus imaging

To compare early diabetic retinopathy (DR) severity level and the abilities in detecting early DR lesions among conventional five-field, ultrawide-field (UWF) Optos, and UWF Clarus fundus imaging methods. This was a single-center, prospective, clinic-based, and comparative study. In total, 157 consecutive patients with diabetes mellitus were enrolled in this study. All patients underwent comprehensive ophthalmological examinations. Following mydriasis, each eye was examined with conventional five-field, UWF Optos, and UWF Clarus fundus imaging methods. The initial UWF images were overlaid with a template mask that obscured the retina, which created a five-field view from UWF images (covered UWF images). The covered UWF images were then graded, after which the template mask was removed, and the original UWF images were also evaluated. All images were graded using the International Clinical DR severity scale. DR grades were compared and analyzed by weighted kappa statistics among the three fundus imaging methods. In total, 157 consecutive patients with diabetes (302 eyes) were enrolled in this study. Weighted kappa statistics for agreement were 0.471 (five-field vs. covered Optos), 0.809 (five-field vs. covered Clarus), 0.396 (covered Optos vs. covered Clarus), 0.463 (five-field vs. Optos), 0.521 (five-field vs. Clarus 133°), 0.500 (five-field vs. Clarus 200°), 0.323 (Optos vs. Clarus 133°), and 0.349 (Optos vs. Clarus 200°). The area under curve of covered Clarus images was higher than that of conventional five-field images at three different thresholds. Compared with conventional five-field and Optos fundus imaging methods, Clarus fundus imaging methods exhibited excellent performance in assessing early DR severity. Thus, Clarus fundus imaging methods were superior for early detection of DR.

ACCURATE REGISTRATION BETWEEN ULTRA-WIDE-FIELD AND NARROW ANGLE RETINA IMAGES WITH 3D EYEBALL SHAPE OPTIMIZATION.

The Ultra-Wide-Field (UWF) retina images have attracted wide attentions in recent years in the study of retina. However, accurate registration between the UWF images and the other types of retina images could be challenging due to the distortion in the peripheral areas of an UWF image, which a 2D warping can not handle. In this paper, we propose a novel 3D distortion correction method which sets up a 3D projection model and optimizes a dense 3D retina mesh to correct the distortion in the UWF image. The corrected UWF image can then be accurately aligned to the target image using 2D alignment methods. The experimental results show that our proposed method outperforms the state-of-the-art method by 30%.

Ultra-Widefield Imaging as a Teleophthalmology Screening Tool for Ocular Pathology.

Prior studies have validated ultra-widefield imaging as a remote screening tool for diabetic retinopathy. The aim of this study was to determine its use in screening for any fundus pathology in a routine patient population. In this prospective randomized study, patients underwent both slit lamp indirect ophthalmoscopy and ultra-widefield imaging. Ultra-widefield images were independently reviewed by two optometrists, and discrepancies were adjudicated by a retina specialist. Clinical findings from slit-lamp examiners and image-reviewers were coded into themes and clinically meaningful findings were extracted. Cohen's kappa was used to estimate agreement for these findings between the two image-reviewers and between the image-reviewers and slit-lamp examiners. Nine-hundred eyes of 450 patients were examined and imaged, of which 616 eyes were analyzed. At least one abnormal fundus finding was present on ophthalmoscopy in 71 eyes (11%) and on adjudicated image interpretation in 166 eyes (27%). Agreement between the two image-reviewers was moderate to substantial for most clinically meaningful findings, including optic disc hemorrhage (κ = 0.8), macular exudates (κ = 0.7), and macular pigmentary changes (κ = 0.7). Agreement between examiners and image-reviewers was moderate to substantial for optic disc hemorrhage (κ = 1), indistinct optic disc margins (κ = 0.5), drusen (κ = 0.4), pigmentary changes (κ = 0.4), and hemorrhage (κ = 0.8). A total of 187 findings were detected by imaging but not examination, compared with 42 that were detected on examination but not imaging. In a routine patient population, ultra-widefield imaging agreed with standard-of-care slit-lamp examinations and detected more fundus findings.

Ultra-widefield fundus image of an oculocutaneous albinism patient with rhegmatogenous retinal detachment

Ultra-widefield fundus image of an oculocutaneous albinism patient with rhegmatogenous retinal detachment

Discriminative-Region Multi-Label Classification of Ultra-Widefield Fundus Images.

Ultra-widefield fundus image (UFI) has become a crucial tool for ophthalmologists in diagnosing ocular diseases because of its ability to capture a wide field of the retina. Nevertheless, detecting and classifying multiple diseases within this imaging modality continues to pose a significant challenge for ophthalmologists. An automated disease classification system for UFI can support ophthalmologists in making faster and more precise diagnoses. However, existing works for UFI classification often focus on a single disease or assume each image only contains one disease when tackling multi-disease issues. Furthermore, the distinctive characteristics of each disease are typically not utilized to improve the performance of the classification systems. To address these limitations, we propose a novel approach that leverages disease-specific regions of interest for the multi-label classification of UFI. Our method uses three regions, including the optic disc area, the macula area, and the entire UFI, which serve as the most informative regions for diagnosing one or multiple ocular diseases. Experimental results on a dataset comprising 5930 UFIs with six common ocular diseases showcase that our proposed approach attains exceptional performance, with the area under the receiver operating characteristic curve scores for each class spanning from 95.07% to 99.14%. These results not only surpass existing state-of-the-art methods but also exhibit significant enhancements, with improvements of up to 5.29%. These results demonstrate the potential of our method to provide ophthalmologists with valuable information for early and accurate diagnosis of ocular diseases, ultimately leading to improved patient outcomes.