Search

In ophthalmic diagnostics, achieving precise segmentation of retinal blood vessels is a critical yet challenging task, primarily due to the complex nature of retinal images. The intricacies of these images often hinder the accuracy and efficiency of segmentation processes. To overcome these challenges, we introduce the cognitive DL retinal blood vessel segmentation (CoDLRBVS), a novel hybrid model that synergistically combines the deep learning capabilities of the U-Net architecture with a suite of advanced image processing techniques. This model uniquely integrates a preprocessing phase using a matched filter (MF) for feature enhancement and a post-processing phase employing morphological techniques (MT) for refining the segmentation output. Also, the model incorporates multi-scale line detection and scale space methods to enhance its segmentation capabilities. Hence, CoDLRBVS leverages the strengths of these combined approaches within the cognitive computing framework, endowing the system with human-like adaptability and reasoning. This strategic integration enables the model to emphasize blood vessels, accurately segment effectively, and proficiently detect vessels of varying sizes. CoDLRBVS achieves a notable mean accuracy of 96.7%, precision of 96.9%, sensitivity of 99.3%, and specificity of 80.4% across all of the studied datasets, including DRIVE, STARE, HRF, retinal blood vessel and Chase-DB1. CoDLRBVS has been compared with different models, and the resulting metrics surpass the compared models and establish a new benchmark in retinal vessel segmentation. The success of CoDLRBVS underscores its significant potential in advancing medical image processing, particularly in the realm of retinal blood vessel segmentation. © 2024. The Author(s).

In ophthalmic diagnostics, achieving precise segmentation of retinal blood vessels is a critical yet challenging task, primarily due to the complex nature of retinal images. The intricacies of these images often hinder the accuracy and efficiency of segmentation processes. To overcome these challenges, we introduce the cognitive DL retinal blood vessel segmentation (CoDLRBVS), a novel hybrid model that synergistically combines the deep learning capabilities of the U-Net architecture with a suite of advanced image processing techniques. This model uniquely integrates a preprocessing phase using a matched filter (MF) for feature enhancement and a post-processing phase employing morphological techniques (MT) for refining the segmentation output. Also, the model incorporates multi-scale line detection and scale space methods to enhance its segmentation capabilities. Hence, CoDLRBVS leverages the strengths of these combined approaches within the cognitive computing framework, endowing the system with human-like adaptability and reasoning. This strategic integration enables the model to emphasize blood vessels, accurately segment effectively, and proficiently detect vessels of varying sizes. CoDLRBVS achieves a notable mean accuracy of 96.7%, precision of 96.9%, sensitivity of 99.3%, and specificity of 80.4% across all of the studied datasets, including DRIVE, STARE, HRF, retinal blood vessel and Chase-DB1. CoDLRBVS has been compared with different models, and the resulting metrics surpass the compared models and establish a new benchmark in retinal vessel segmentation. The success of CoDLRBVS underscores its significant potential in advancing medical image processing, particularly in the realm of retinal blood vessel segmentation. © 2024. The Author(s).

In ophthalmic diagnostics, achieving precise segmentation of retinal blood vessels is a critical yet challenging task, primarily due to the complex nature of retinal images. The intricacies of these images often hinder the accuracy and efficiency of segmentation processes. To overcome these challenges, we introduce the cognitive DL retinal blood vessel segmentation (CoDLRBVS), a novel hybrid model that synergistically combines the deep learning capabilities of the U-Net architecture with a suite of advanced image processing techniques. This model uniquely integrates a preprocessing phase using a matched filter (MF) for feature enhancement and a post-processing phase employing morphological techniques (MT) for refining the segmentation output. Also, the model incorporates multi-scale line detection and scale space methods to enhance its segmentation capabilities. Hence, CoDLRBVS leverages the strengths of these combined approaches within the cognitive computing framework, endowing the system with human-like adaptability and reasoning. This strategic integration enables the model to emphasize blood vessels, accurately segment effectively, and proficiently detect vessels of varying sizes. CoDLRBVS achieves a notable mean accuracy of 96.7%, precision of 96.9%, sensitivity of 99.3%, and specificity of 80.4% across all of the studied datasets, including DRIVE, STARE, HRF, retinal blood vessel and Chase-DB1. CoDLRBVS has been compared with different models, and the resulting metrics surpass the compared models and establish a new benchmark in retinal vessel segmentation. The success of CoDLRBVS underscores its significant potential in advancing medical image processing, particularly in the realm of retinal blood vessel segmentation. © 2024. The Author(s).

In ophthalmic diagnostics, achieving precise segmentation of retinal blood vessels is a critical yet challenging task, primarily due to the complex nature of retinal images. The intricacies of these images often hinder the accuracy and efficiency of segmentation processes. To overcome these challenges, we introduce the cognitive DL retinal blood vessel segmentation (CoDLRBVS), a novel hybrid model that synergistically combines the deep learning capabilities of the U-Net architecture with a suite of advanced image processing techniques. This model uniquely integrates a preprocessing phase using a matched filter (MF) for feature enhancement and a post-processing phase employing morphological techniques (MT) for refining the segmentation output. Also, the model incorporates multi-scale line detection and scale space methods to enhance its segmentation capabilities. Hence, CoDLRBVS leverages the strengths of these combined approaches within the cognitive computing framework, endowing the system with human-like adaptability and reasoning. This strategic integration enables the model to emphasize blood vessels, accurately segment effectively, and proficiently detect vessels of varying sizes. CoDLRBVS achieves a notable mean accuracy of 96.7%, precision of 96.9%, sensitivity of 99.3%, and specificity of 80.4% across all of the studied datasets, including DRIVE, STARE, HRF, retinal blood vessel and Chase-DB1. CoDLRBVS has been compared with different models, and the resulting metrics surpass the compared models and establish a new benchmark in retinal vessel segmentation. The success of CoDLRBVS underscores its significant potential in advancing medical image processing, particularly in the realm of retinal blood vessel segmentation. © 2024. The Author(s).

Biological age, distinct from an individual's chronological age, has been studied extensively through predictive aging clocks. However, these clocks have limited accuracy in short time-scales. Here we trained deep learning models on fundus images from the EyePACS dataset to predict individuals' chronological age. Our retinal aging clocking, 'eyeAge', predicted chronological age more accurately than other aging clocks (mean absolute error of 2.86 and 3.30 years on quality-filtered data from EyePACS and UK Biobank, respectively). Additionally, eyeAge was independent of blood marker-based measures of biological age, maintaining an all-cause mortality hazard ratio of 1.026 even when adjusted for phenotypic age. The individual-specific nature of eyeAge was reinforced via multiple GWAS hits in the UK Biobank cohort. The top GWAS locus was further validated via knockdown of the fly homolog, Alk, which slowed age-related decline in vision in flies. This study demonstrates the potential utility of a retinal aging clock for studying aging and age-related diseases and quantitatively measuring aging on very short time-scales, opening avenues for quick and actionable evaluation of gero-protective therapeutics.

In this study, we explored inter-ocular asymmetry (between the two eyes of the same patient) using optical coherence tomography angiography (OCTA) in patients with diabetes mellitus (DM) at different retinopathy stages. A total of 258 patients were divided into four groups: no DM, DM without diabetic retinopathy (DR), non-proliferative DR (NPDR), and proliferative DR (PDR). Superficial and deep vessel density (SVD, DVD), superficial and deep perfusion density (SPD, DPD), foveal avascular zone (FAZ) area, perimeter and circularity were calculated, and asymmetry index (AI) was used to evaluate the asymmetry of two eyes of the same subject. AIs of SPD, SVD, FAZ area and FAZ perimeter in the PDR group were larger than all other 3 groups (all p < 0.05). The AIs of DPD, DVD, FAZ area and FAZ perimeter in males were larger than in females (p = 0.015, p = 0.023, p = 0.006 and p = 0.017). Hemoglobin A1c (HbA1c) was positively correlated with AI of FAZ perimeter (p = 0.02) and circularity (p = 0.022). In conclusion, PDR patients' eyes were significantly asymmetric in both vascular density and FAZ metrics. Male sex and HbA1c are risk factors that influenced symmetry. This study highlights that right-left asymmetry should be taken into account in DR-related studies, particularly those analyzing microvascular changes with OCTA.

Retinal hemorrhages in pediatric patients can be a diagnostic challenge for ophthalmologists. These hemorrhages can occur due to various underlying etiologies, including abusive head trauma, accidental trauma, and medical conditions. Accurate identification of the etiology is crucial for appropriate management and legal considerations. In recent years, deep learning techniques have shown promise in assisting healthcare professionals in making more accurate and timely diagnosis of a variety of disorders. We explore the potential of deep learning approaches for differentiating etiologies of pediatric retinal hemorrhages. Our study, which spanned multiple centers, analyzed 898 images, resulting in a final dataset of 597 retinal hemorrhage fundus photos categorized into medical (49.9%) and trauma (50.1%) etiologies. Deep learning models, specifically those based on ResNet and transformer architectures, were applied; FastViT-SA12, a hybrid transformer model, achieved the highest accuracy (90.55%) and area under the receiver operating characteristic curve (AUC) of 90.55%, while ResNet18 secured the highest sensitivity value (96.77%) on an independent test dataset. The study highlighted areas for optimization in artificial intelligence (AI) models specifically for pediatric retinal hemorrhages. While AI proves valuable in diagnosing these hemorrhages, the expertise of medical professionals remains irreplaceable. Collaborative efforts between AI specialists and pediatric ophthalmologists are crucial to fully harness AIs potential in diagnosing etiologies of pediatric retinal hemorrhages.

BACKGROUND: The atrophic late stage of age-related macular degeneration (AMD) is termed geographic atrophy (GA), and affects visual acuity (VA) as well as quality of life (QoL). Previous studies have found that best-corrected VA (BCVA), the standard vision assessment often underrepresents functional deficits. Therefore, the purpose of this study was to evaluate the correlation between atrophic lesion size, VA and QoL measured with the National Eye Institute Visual Function Questionnaire (VFQ-39) in a Danish population. Moreover, we wanted to evaluate the correlation between comorbidities, behavioural factors, and QoL.METHODS: This was prospective clinical study of 51 patients with GA in one or both eyes, of these 45 patients had bilateral GA. Patients were consecutively included between April 2021 and February 2022. All patients filled in the VFQ-39 questionnaire except the subscales "ocular pain" and "peripheral vision." Lesion size was measured from fundus autoflourescense images, and BCVA was assessed by the Early Treatment Diabetic Retinopathy Study (ETDRS) protocol.RESULTS: We found an overall low score in each VFQ-39 subscale scores reflected by GA. Lesion size and VA were both significantly associated with all VFQ-39 subscale scores except for "general health." VA showed a larger effect on QoL than lesion size. Chronic obstructive pulmonary disease (COPD) was associated with a lower score in the subscale score "general health" but none of the other subscale scores were affected. Cardiovascular disease (CVD) was associated with a lower BCVA as well as in QoL reflected in the subscale scores "poor general vision," "near activities," and "dependency" of VFQ-39.CONCLUSION: Both atrophic lesion size and visual acuity affects QoL in Danish patients with GA, who reports an overall poor QoL. CVD seems to have a negative effect on disease, as well as in VFQ-39 in several subscales, whereas COPD did not affect disease severity or vision-related

BACKGROUND: The atrophic late stage of age-related macular degeneration (AMD) is termed geographic atrophy (GA), and affects visual acuity (VA) as well as quality of life (QoL). Previous studies have found that best-corrected VA (BCVA), the standard vision assessment often underrepresents functional deficits. Therefore, the purpose of this study was to evaluate the correlation between atrophic lesion size, VA and QoL measured with the National Eye Institute Visual Function Questionnaire (VFQ-39) in a Danish population. Moreover, we wanted to evaluate the correlation between comorbidities, behavioural factors, and QoL.METHODS: This was prospective clinical study of 51 patients with GA in one or both eyes, of these 45 patients had bilateral GA. Patients were consecutively included between April 2021 and February 2022. All patients filled in the VFQ-39 questionnaire except the subscales "ocular pain" and "peripheral vision." Lesion size was measured from fundus autoflourescense images, and BCVA was assessed by the Early Treatment Diabetic Retinopathy Study (ETDRS) protocol.RESULTS: We found an overall low score in each VFQ-39 subscale scores reflected by GA. Lesion size and VA were both significantly associated with all VFQ-39 subscale scores except for "general health." VA showed a larger effect on QoL than lesion size. Chronic obstructive pulmonary disease (COPD) was associated with a lower score in the subscale score "general health" but none of the other subscale scores were affected. Cardiovascular disease (CVD) was associated with a lower BCVA as well as in QoL reflected in the subscale scores "poor general vision," "near activities," and "dependency" of VFQ-39.CONCLUSION: Both atrophic lesion size and visual acuity affects QoL in Danish patients with GA, who reports an overall poor QoL. CVD seems to have a negative effect on disease, as well as in VFQ-39 in several subscales, whereas COPD did not affect disease severity or vision-related

BackgroundProper evaluation of ocular fundi is an integral part of neurological examination. Unfortunately, neurology residents are increasingly uncomfortable performing fundoscopy and interpreting findings because of diminishing skills and lack of experience. This became more prominent during the COVID-19 pandemic as fundoscopy requires proximity to the patient. With the recent dramatic improvement of smartphone cameras, fundus photography using the PanOptic Ophthalmoscope (Welch Allyn, Skaneateles Falls, NY) with a smartphone adapter offered an alternative to direct fundoscopic examination. We present the first experience with our own design of a universal smartphone adapter.MethodsThis is a single-center case series, consecutive for a single user and certain presenting neurological symptoms, which is aimed to evaluate the feasibility and practicality of a new, universal PanOptic smartphone adapter. Presenting symptoms included headache, ocular symptoms, seizure, or encephalopathy. We used 3D modeling and printing techniques to create the adapter. We also developed a methodology of capturing stereoscopic images of the optic disc using a single smartphone camera, but the method was not systematically evaluated in this paper.ResultsHere we present our initial experience of fundus video/photography in patients, who presented with encephalopathy, headache, seizure, vision loss, and other ocular symptoms. Fundoscopic abnormalities were discovered in 11 out of 100 patients. Some were incidental findings and were unrelated to the presentation. In one case, fundoscopy played a critical role in establishing the correct diagnosis.ConclusionsOur custom-designed smartphone adapter allowed obtaining high-quality video and photo recordings using PanOptic Ophthalmoscope. The acquisition of high-quality photos enables a high-yield diagnostic tool and allows revisiting the image in the patient's chart. Improvement of smartphone cameras opens vast horizons for stereo-fundoscopy and

BackgroundProper evaluation of ocular fundi is an integral part of neurological examination. Unfortunately, neurology residents are increasingly uncomfortable performing fundoscopy and interpreting findings because of diminishing skills and lack of experience. This became more prominent during the COVID-19 pandemic as fundoscopy requires proximity to the patient. With the recent dramatic improvement of smartphone cameras, fundus photography using the PanOptic Ophthalmoscope (Welch Allyn, Skaneateles Falls, NY) with a smartphone adapter offered an alternative to direct fundoscopic examination. We present the first experience with our own design of a universal smartphone adapter.MethodsThis is a single-center case series, consecutive for a single user and certain presenting neurological symptoms, which is aimed to evaluate the feasibility and practicality of a new, universal PanOptic smartphone adapter. Presenting symptoms included headache, ocular symptoms, seizure, or encephalopathy. We used 3D modeling and printing techniques to create the adapter. We also developed a methodology of capturing stereoscopic images of the optic disc using a single smartphone camera, but the method was not systematically evaluated in this paper.ResultsHere we present our initial experience of fundus video/photography in patients, who presented with encephalopathy, headache, seizure, vision loss, and other ocular symptoms. Fundoscopic abnormalities were discovered in 11 out of 100 patients. Some were incidental findings and were unrelated to the presentation. In one case, fundoscopy played a critical role in establishing the correct diagnosis.ConclusionsOur custom-designed smartphone adapter allowed obtaining high-quality video and photo recordings using PanOptic Ophthalmoscope. The acquisition of high-quality photos enables a high-yield diagnostic tool and allows revisiting the image in the patient's chart. Improvement of smartphone cameras opens vast horizons for stereo-fundoscopy and

ObjectiveDiseases such as age-related macular degeneration (AMD) are classified based on human rubrics that are prone to bias. Supervised neural networks trained using human-generated labels require labor-intensive annotations and are restricted to specific trained tasks. Here, we trained a self-supervised deep learning network using unlabeled fundus images, enabling data-driven feature classification of AMD severity and discovery of ocular phenotypes.DesignDevelopment of a self-supervised training pipeline to evaluate fundus photographs from the Age-Related Eye Disease Study (AREDS).ParticipantsOne hundred thousand eight hundred forty-eight human-graded fundus images from 4757 AREDS participants between 55 and 80 years of age.MethodsWe trained a deep neural network with self-supervised Non-Parametric Instance Discrimination (NPID) using AREDS fundus images without labels then evaluated its performance in grading AMD severity using 2-step, 4-step, and 9-step classification schemes using a supervised classifier. We compared balanced and unbalanced accuracies of NPID against supervised-trained networks and ophthalmologists, explored network behavior using hierarchical learning of image subsets and spherical k-means clustering of feature vectors, then searched for ocular features that can be identified without labels.Main outcome measuresAccuracy and kappa statistics.ResultsNPID demonstrated versatility across different AMD classification schemes without re-training and achieved balanced accuracies comparable with those of supervised-trained networks or human ophthalmologists in classifying advanced AMD (82% vs. 81-92% or 89%), referable AMD (87% vs. 90-92% or 96%), or on the 4-step AMD severity scale (65% vs. 63-75% or 67%), despite never directly using these labels during self-supervised feature learning. Drusen area drove network predictions on the 4-step scale, while depigmentation and geographic atrophy (GA) areas correlated with advanced AMD classes. Self-supervi

El análisis del fondo del ojo es crucial para prevenir enfermedades retinianas y coroideas ya que la mayoría no causan síntomas en etapas tempranas. Tratarlas cuando aparecen los primeros indicios es fundamental para evitar pérdidas de visión irreversibles. Con este propósito, el color de las estructuras de fondo de ojo de pacientes sanos y enfermos se evaluó a partir de imágenes adquiridas con una cámara de fondo de ojo multiespectral (400nm-1300nm) con elevada resolución espectral y espacial. En ojos sanos aparecieron diferencias de color CIEDE2000 considerables entre arterias y venas por su diferente oxigenación; las fibras nerviosas y la fóvea aparecieron más contrastadas respecto al fondo, produciendo diferencias de color relevantes. En ojos afectados por degeneración macular asociada a la edad, se pudieron identificar mejor que en retinografías en color las características drusas y alteraciones del disco óptico en pacientes con glaucoma mostraron valores CIEDE2000 elevados respecto a pacientes sanos., Este proyecto ha recibido financiación del programa de investigación e innovación Horizon 2020 de la Unión Europea bajo el acuerdo de subvención Marie kłodowska-Curie No. 801342 (Tecniospring INDUSTRY), así como de la Agència per a la Competitivitat de l’Empresa de la Generalitat de Catalunya (ACCIÓ). Además, esta publicación es parte del proyecto de I+D+i PID2020- 112527RB-I00, financiada por MCIN/AEI/10.13039/501100011033.

We examined the spectral reflectance of fundus structures in the visible and near-infrared (400–1300 nm) range for contributing to the medical diagnosis of fundus diseases. Spectral images of healthy eye fundus and other ocular diseases were acquired using a novel multispectral fundus camera. Reflectance metrics were computed based on contrast to analyze the spectral features. Significant differences were observed among the structures in healthy and diseased eye fundus. Specifically, near-infrared analysis allows imaging of deeper layers, such as the choroid, which, to date, has not been retrieved using traditional color fundus cameras. Pathological structures, which were hardly observable in color fundus images owing to metamerism, were also revealed by the developed multispectral fundus camera., "Funding: H2020Marie Skłodowska-Curie Actions (801342, Tecniospring INDUSTRY); Agència per a la Competitivitat de l’Empresa (801342, Tecniospring INDUSTRY); Generalitat de Catalunya (801342, Tecniospring INDUSTRY); Ministerio de Ciencia e Innovación- Agencia Estatal de Investigación (MCIN/AEI/10.13039/501100011033, PID2020112527RB-I00)", Peer Reviewed, Postprint (published version)

El análisis del fondo del ojo es crucial para prevenir enfermedades retinianas y coroideas ya que la mayoría no causan síntomas en etapas tempranas. Tratarlas cuando aparecen los primeros indicios es fundamental para evitar pérdidas de visión irreversibles. Con este propósito, el color de las estructuras de fondo de ojo de pacientes sanos y enfermos se evaluó a partir de imágenes adquiridas con una cámara de fondo de ojo multiespectral (400nm-1300nm) con elevada resolución espectral y espacial. En ojos sanos aparecieron diferencias de color CIEDE2000 considerables entre arterias y venas por su diferente oxigenación; las fibras nerviosas y la fóvea aparecieron más contrastadas respecto al fondo, produciendo diferencias de color relevantes. En ojos afectados por degeneración macular asociada a la edad, se pudieron identificar mejor que en retinografías en color las características drusas y alteraciones del disco óptico en pacientes con glaucoma mostraron valores CIEDE2000 elevados respecto a pacientes sanos, Este proyecto ha recibido financiación del programa de investigación e innovación Horizon 2020 de la Unión Europea bajo el acuerdo de subvención Marie Skłodowska-Curie No. 801342 (Tecniospring INDUSTRY), así como de la Agència per a la Competitivitat de l’Empresa de la Generalitat de Catalunya (ACCIÓ). Además, esta publicación es parte del proyecto de I+D+i PID2020-112527RB-I00, financiada por MCIN/AEI/10.13039/501100011033., Peer Reviewed, Postprint (published version)

PurposeTo develop geometric perfusion deficits (GPD), an optical coherence tomography angiography (OCTA) biomarker based on oxygen diffusion, and to evaluate its utility in a pilot study of healthy subjects and patients with diabetic retinopathy (DR).DesignRetrospective cross-sectional study.MethodsCommercial spectral-domain optical coherence tomography angiography (OCTA) instruments were used to acquire repeated 3 × 3-mm2 and 6 × 6-mm2 motion-corrected macular OCTA volumes. En face OCTA images corresponding to the superficial capillary plexus (SCP), deep capillary plexus (DCP), and full retinal projections were obtained using automatic segmentation. For each projection, the GPD percentage and the vessel density percentage, the control metric, were computed, and their values were compared between the normal and DR eyes. The repeated OCTA acquisitions were used to assess the test-retest repeatability of the GPD and vessel density percentages.ResultsRepeated OCTA scans of 15 normal eyes and 12 DR eyes were obtained. For all en face projections, GPD percentages were significantly higher in DR eyes than in normal eyes; vessel density percentages were significantly lower in all but 1 projection (DCP). Large GPD areas were used to identify focal perfusion deficits. Test-retest analysis showed that the GPD percentage had superior repeatability than the vessel density percentage in most cases. A strong negative correlation between the GPD percentage and the vessel density percentage was also found.ConclusionsGeometric perfusion deficits, an OCTA biomarker based on oxygen diffusion, provides a quantitative metric of macular microvascular remodeling with a strong physiological underpinning. The GPD percentage may serve as a useful biomarker for detecting and monitoring DR.

PurposeTo develop geometric perfusion deficits (GPD), an optical coherence tomography angiography (OCTA) biomarker based on oxygen diffusion, and to evaluate its utility in a pilot study of healthy subjects and patients with diabetic retinopathy (DR).DesignRetrospective cross-sectional study.MethodsCommercial spectral-domain optical coherence tomography angiography (OCTA) instruments were used to acquire repeated 3 × 3-mm2 and 6 × 6-mm2 motion-corrected macular OCTA volumes. En face OCTA images corresponding to the superficial capillary plexus (SCP), deep capillary plexus (DCP), and full retinal projections were obtained using automatic segmentation. For each projection, the GPD percentage and the vessel density percentage, the control metric, were computed, and their values were compared between the normal and DR eyes. The repeated OCTA acquisitions were used to assess the test-retest repeatability of the GPD and vessel density percentages.ResultsRepeated OCTA scans of 15 normal eyes and 12 DR eyes were obtained. For all en face projections, GPD percentages were significantly higher in DR eyes than in normal eyes; vessel density percentages were significantly lower in all but 1 projection (DCP). Large GPD areas were used to identify focal perfusion deficits. Test-retest analysis showed that the GPD percentage had superior repeatability than the vessel density percentage in most cases. A strong negative correlation between the GPD percentage and the vessel density percentage was also found.ConclusionsGeometric perfusion deficits, an OCTA biomarker based on oxygen diffusion, provides a quantitative metric of macular microvascular remodeling with a strong physiological underpinning. The GPD percentage may serve as a useful biomarker for detecting and monitoring DR.

The analysis of the eye fundus is critical to prevent retinal and choroidal diseases since most of them cause no symptoms at early stages. Treating them when the very first signs appear is crucial to avoid vision losses. To this end, the color of eye fundus structures of healthy and diseased patients was assessed from images acquired with a novel multispectral fundus camera (400 nm – 1300 nm) with high spectral and spatial resolution. Characteristic color traits were found: in healthy eyes, large CIEDE2000 color differences were reported between arteries and veins due to different blood oxygenation; the contrast of nerve fibers/fovea was enhanced, giving rise to relevant color differences; in eyes with age related macular degeneration, lesions such as drusen could be better distinguished than with traditional color retinography; alterations of the optic disk in patients with glaucoma were also assessed, showing remarkable CIEDE2000 values when compared to healthy patients, This project has received funding from the European Union's Horizon 2020 research and innovation programme under Marie Skłodowska-Curie grant agreement No. 801342 (Tecniospring INDUSTRY) and the Government of Catalonia's Agency for Business Competitiveness (ACCIÓ). This research was also supported by the Ministerio de Economía, Industria y Competitividad (MINECO), the Agencia Estatal de Investigación (AEI) and the European Regional Development Fund (FEDER) under the grant DPI2017-89414-R., Peer Reviewed, Postprint (published version)

Background/objectivesWe aim to describe the clinical and imaging characteristics of young children with dark without pressure- a well-defined area of dark retina in the mid periphery associated with attenuated hyperreflective outer segment bands on spectral domain OCT.Subjects/methodsRetrospective medical chart review. We reviewed the charts and retinal images of children with dark without pressure (DWP) seen in the paediatric eye clinic that had undergone retinal imaging.ResultsWe identified 6 children (3 males, 3 females) ranging in age from 3 to 13 years with DWP lesions. These were an incidental finding in children referred for other disorders. The lesions spared the macula and were typically in the mid-peripheral retina. In those children who could cooperate with detailed imaging, the DWP lesions were hypoautofluorecent on fundus autofluorescence imaging and OCT imaging showed reduced reflectance of the outer segments and EZ line as has been described in older children and adults. In two of the six subjects a DWP lesion appeared in an area of previously normal fundus; in no subjects did the lesion resolve over the period of follow up.ConclusionsThe clinical and imaging characteristics of DWP lesions are similar in young children to those described in adults. The site of dysfunction appears to be at the level of the outer retina, but the precise cause is unknown. Recognition of this benign disorder will prevent patient anxiety and unnecessary further investigation.

PurposeTo determine the prevalence of different types of artifacts seen in OCT angiography (OCTA) images of healthy and glaucoma eyes and evaluate the characteristics associated with poor-quality images.DesignRetrospective study.ParticipantsA total of 649 eyes of 368 healthy, glaucoma suspect, and glaucoma patients.MethodsAngiovue (Optovue Inc) high-density (HD) and non-HD optic nerve head and macula OCTA images of participants were evaluated by 4 expert reviewers for the presence of different artifacts, including eye movement, defocus, shadow, decentration, segmentation error, blink, and Z offset in the superficial vascular layer. Each OCTA scan was designated to have good or poor quality based on the presence of artifacts. The association of demographic and ocular characteristics with the likelihood of obtaining poor-quality OCTA images was evaluated.Main outcome measuresThe prevalence of OCTA artifacts and the factors associated with increased likelihood of capturing poor-quality OCTA images.ResultsA total of 5263 OCTA images were evaluated. Overall, 33.9% of the OCTA images had poor quality. The majority of images with acceptable quality scores (QS ≥ 4) had no artifacts (76.6%). Other images had 1 (13.6%) or 2 or more artifacts (9.8%). Older age (P < 0.001), male gender (P = 0.045), worse visual field mean deviation (P < 0.001), absence of eye tracking (P < 0.001), and macular scan area (P < 0.001) were associated with a higher likelihood of obtaining poor-quality images. In images with acceptable QS, the commercially available quality measures including QS and signal strength index had the area under the receiver operating characteristic curves of 0.65 (95% confidence interval [CI], 0.62-0.69) and 0.70 (95% CI, 0.68-0.73) to detect good-quality images, respectively.ConclusionsOCTA artifacts associated with poor-quality images are frequent, and their prevalence is affected by ocular and patient characteristics. One should not rely solely on the

PurposeTo describe longitudinal multimodal imaging findings of nonexudative choroidal neovascularization in CTRP5 late-onset retinal degeneration.MethodsFour patients with CTRP5-positive late-onset retinal degeneration underwent repeated ophthalmoscopic examination and multimodal imaging. All four patients (two siblings and their cousins, from a pedigree described previously) had the heterozygous S163R mutation.ResultsAll four patients demonstrated large subretinal lesions in the mid-peripheral retina of both eyes. The lesions were characterized by confluent hypercyanescence with hypocyanescent borders on indocyanine green angiography, faintly visible branching vascular networks with absent/minimal leakage on fluorescein angiography, Type 1 neovascularization on optical coherence tomography angiography, and absent retinal fluid, consistent with nonexudative choroidal neovascularization. The neovascular membranes enlarged substantially over time and the birth of new membranes was observed, but all lesions remained nonexudative/minimally exudative. Without treatment, all involved retinal areas remained free of atrophy and subretinal fibrosis.ConclusionWe report the existence of massive advancing nonexudative Type 1 choroidal neovascularization in CTRP5 late-onset retinal degeneration. These findings have implications for age-related macular degeneration. They provide a monogenic model system for studying the mechanisms underlying the distinct events of choroidal neovascularization development, enlargement, progression to exudation, and atrophy in age-related macular degeneration. They suggest that choroidal hypoperfusion precedes neovascularization and that nonexudative neovascularization may protect against atrophy.

PurposeTo assess the influence of lesion morphology and location on geographic atrophy (GA) growth rate.MethodsWe manually delineated GA on color fundus photographs of 237 eyes in the Age-Related Eye Disease Study. We calculated local border expansion rate (BER) as the linear distance that a point on the GA border traveled over 1 year based on a Euclidean distance map. Eye-specific BER was defined as the mean local BER of all points on the GA border in an eye. The percentage area affected by GA was defined as the GA area divided by the total retinal area in the region.ResultsGA enlarged 1.51 ± 1.96 mm2 in area and 0.13 ± 0.11 mm in distance over 1 year. The GA area growth rate (mm2/y) was associated with the baseline GA area (P < 0.001), perimeter (P < 0.001), lesion number (P < 0.001), and circularity index (P < 0.001); in contrast, eye-specific BER (mm/y) was not significantly associated with any of these factors. As the retinal eccentricity increased from 0 to 3.5 mm, the local BER increased from 0.10 to 0.24 mm/y (P < 0.001); in contrast, the percentage of area affected by GA decreased from 49.3% to 2.3%.ConclusionsUsing distance-based measurements allows GA progression evaluation without significant confounding effects from baseline GA morphology. Local GA progression rates increased as a function of retinal eccentricity within the macula which is opposite of the trend for GA distribution, suggesting that GA initiation and enlargement may be mediated by different biological processes.

Las manifestaciones oculares de la enfermedad reumática son frecuentes y variables estando asociadas a elementos diagnósticos, actividad de la enfermedad o reacciones terapéuticas adversas. Este estudio se realizó con el fin de describir los hallazgos fundoscópicos en 33 pacientes con enfermedad reumática y sin antecedentes de hipertensión, diabetes o lesión ocular bilateral que acudieron a los Servicios de Medicina Interna del Hospital Central Universitario Dr. Antonio María Pineda durante el lapso noviembre-diciembre del 2018. El 72,1% de los pacientes se encontraban en el grupo etario de 41 a 60 años, predominando el sexo femenino (83,7%); las patologías más frecuentes fueron la artritis reumatoide (58,1%) y el lupus (18,6%). El 33,3% de los casos presentaron hallazgos anómalos dados por cruces arteriovenosos (54,5%), disminución del diámetro vascular (36,4%) y hemorragia retiniana (9,1%). Con relación a los cruces patológicos, éstos fueron observados en el 16,6% de los pacientes con artritis reumatoide, en todos los pacientes con artritis psoriásica y en el 33,3% de los pacientes con espondiloartritis. La disminución del diámetro vascular se observó en el 11,1% de los pacientes con artritis reumatoide, en el 33,3% de los pacientes con espondiloartritis y en el 25% de los pacientes con síndrome de Overlap. La hemorragia retiniana fue observada sólo en los pacientes con LES (16,6%).

PURPOSE: To evaluate the incidence of neovascularization (NV) secondary to central serous chorioretinopathy (CSC)-a condition belonging to the spectrum of pachychoroid disorders by means of optical coherence tomography angiography. METHODS: One hundred and seventy five eyes with CSC were evaluated in this retrospective observational study. The eyes with acute or chronic CSC with no NV were included in Group 1, and those with NV were evaluated in Group 2. Only eyes that had undergone structural optical coherence tomography and optical coherence tomography angiography were included. Age, best-corrected visual acuity, and subfoveal choroidal thickness were evaluated in all eyes. In Group 2, the type and morphology of NV and the occurrence of exudation were considered. RESULTS: Of a total of 175 eyes with CSC, 86 had the acute form and 89 the chronic. Approximately 140 belonged to Group 1 (80%) and 35 to Group 2 (20%). Approximately 39.2% of all patient with chronic CSC developed NV. Mean age in Groups 1 and 2 was 53.3 years (±10.9) and 66.6 years (±10.2), respectively. Mean best-corrected visual acuity in Groups 1 and 2 was 45.7 (±11.7) and 30.9 (±17.9) early treatment diabetic retinopathy study letters, respectively. Mean CCT in Group 1 and 2 was 417.5 µm (±123) and 344.2 µm (±165.9), respectively. In Group 2, all patients had Type 1 NV (100%); 29 eyes (83%) had filamentous feature, and 6 eyes (17%) had irregular shape. Silent nonexudative NV was observed in 7 eyes (20%), all belonging to Group 2. CONCLUSION: The use of optical coherence tomography angiography in everyday clinical practice allows for the accurate analysis of the chorioretinal vascular setting, with the identification of new vessels that could remain misdiagnosed.

PURPOSE: To investigate the vascular remodeling of optical coherence tomography angiography in full-thickness macular hole surgery. METHODS: This retrospective, observational case series included 33 eyes of 33 patients with a full-thickness macular hole who underwent epiretinal membrane removal. Data were collected on best-corrected visual acuity, structural B-scan optical coherence tomography, and optical coherence tomography angiography preoperatively and at 1, 3, and 6 months postoperatively. Optical coherence tomography angiography was used to assess the retinal vascular density (VD) of the superficial vascular plexus and deep vascular plexus (DVP). Vascular density was assessed using the Early Treatment Diabetic Retinopathy Study grid for the whole, inner, and outer grids. RESULTS: A statistically significant correlation was found between the VD (whole, inner, and outer grids) of the superficial vascular plexus alone from baseline to 1-month postoperatively (P < 0.001). Similarly, VD was correlated from baseline to 1, 3, and 6 months postoperatively to the whole (P < 0.0005, F = 23.22), inner (P < 0.0005, F = 28.23), and outer Early Treatment Diabetic Retinopathy Study grids of DVP (P = 0.033, F = 3.49). The best-corrected visual acuity and VD were significantly correlated with the superficial vascular plexus and DVP at baseline and 6 months (P < 0.05, all correlations). The most significant correlation was observed at 6 months between best-corrected visual acuity and DVP in the whole, inner, and outer Early Treatment Diabetic Retinopathy Study grids (P < 0.001). CONCLUSION: Superficial vascular plexus and DVP are affected by full-thickness macular holes, with the most significant effects being on the DVP. Full-thickness macular hole surgery leads to an improvement in the best-corrected visual acuity in many eyes and the restoration of the VD, especially of the DVP.

Spectral reflectance of the eye fundus was evaluated in adult healthy patients through a fast visible and near-infrared multispectral fundus camera. Spectral signatures were analyzed for different ocular structures of the retina and the choroid., This project has received funding from the European Union's Horizon 2020 research and innovation programme under Marie Skłodowska-Curie grant agreement No. 801342 (Tecniospring INDUSTRY) and the Government of Catalonia's Agency for Business Competitiveness (ACCIÓ). This research was also supported by the Ministerio de Economía, Industria y Competitividad (MINECO), the Agencia Estatal de Investigación (AEI) and the European Regional Development Fund (FEDER) under the grant DPI2017-89414-R., Peer Reviewed, Postprint (published version)

Spectral reflectance of the eye fundus was evaluated in adult healthy patients through a fast visible and near-infrared multispectral fundus camera. Spectral signatures were analyzed for different ocular structures of the retina and the choroid., This project has received funding from the European Union's Horizon 2020 research and innovation programme under Marie Skłodowska-Curie grant agreement No. 801342 (Tecniospring INDUSTRY) and the Government of Catalonia's Agency for Business Competitiveness (ACCIÓ). This research was also supported by the Ministerio de Economía, Industria y Competitividad (MINECO), the Agencia Estatal de Investigación (AEI) and the European Regional Development Fund (FEDER) under the grant DPI2017-89414-R., Peer Reviewed, Postprint (author's final draft)

PurposeTo characterize the change rate of ganglion cell complex (GCC) thickness and macular vessel density in healthy, preperimetric glaucoma and primary open-angle glaucoma (POAG) eyes.DesignProspective, longitudinal study.ParticipantsOne hundred thirty-nine eyes (23 healthy eyes, 36 preperimetric glaucoma eyes, and 80 POAG eyes) of 94 patients who had at least 3 visits were included from the Diagnostic Innovations in Glaucoma Study. The mean follow-up was 2.0 years for healthy eyes, 2.6 years for preperimetric glaucoma eyes, and 2.6 years for POAG eyes.MethodsOCT angiography (OCTA)-based vessel density and OCT-based structural thickness of the same 3×3-mm2 GCC scan slab were evaluated. The dynamic range-based normalized rates of vessel density and thickness change were calculated and compared within each diagnostic group. The association between the rates of thickness and vessel density change and potential factors were evaluated.Main outcome measuresThe rates of GCC thinning and macular vessel density loss.ResultsSignificant rates of GCC thinning and macular vessel density decrease were detectable in all diagnostic groups (all P < 0.05). In healthy eyes and preperimetric glaucoma eyes, the normalized rates of GCC thinning and macular vessel density decrease were comparable (all P > 0.1). In contrast, the normalized rate (mean, 95% confidence interval) of macular vessel density decrease in the POAG eyes (-7.12 [-8.36, -5.88]%/year) was significantly faster than GCC thinning (-2.13 [-3.35, -0.90]%/year; P < 0.001). In the POAG group, more than two thirds of the eyes showed faster macular vessel density decrease than GCC thinning; faster macular vessel density decrease rate was associated significantly with worse glaucoma severity (P = 0.037). The association between GCC thinning rate and glaucoma severity was not significant (P = 0.586). Intraocular pressure during follow-up significantly affected the rate of GCC thinning in all groups (all P &lt

PurposeTo determine clear-cut distinctions between tractional and exudative intraretinal cystoid spaces subtypes.DesignRetrospective, multicenter, observational case series.MethodsA cohort of patients diagnosed with intraretinal cystoid spaces and imaged with optical coherence tomography (OCT), fluorescein angiography (FA), blue fundus autofluorescence (BFAF), en face OCT, and OCT angiography (OCT-A) was included in the study. All images were qualitatively and quantitatively evaluated.ResultsIn this study were included 72 eyes of 69 patients. Exudative intraretinal cystoid spaces (36/72 eyes, 50%) displayed a "petaloid" morphology as seen with en face OCT, FA, and BFAF. Tractional intraretinal cystoid spaces (24/72 eyes, 33.3%), displayed a radial "spoke-wheel" en face OCT pattern. There was no leakage with FA and BFAF did not reveal specific patterns. Eyes with full-thickness macular hole (FTMH, 12/72 eyes, 16.7%) displayed a "sunflower" en face OCT appearance. FTMH showed OCT, OCT-A, and BFAF features of both exudative and tractional cystoid spaces, but without any FA leakage. Inner nuclear layer (INL) thickness was significantly lower in tractional cystoid spaces (P < .001). There were a greater number of INL cystoid spaces in both the exudative and FTMH subgroups (P = .001). The surface area of INL cystoid spaces was significantly lower in the tractional subgroup (P < .001). There was a significant reduction of the microvascular density in eyes with exudative vs tractional (P = .002) and FTMH (P < .001) subgroups.ConclusionsExudative and tractional intraretinal cystoid spaces displayed characteristic multimodal imaging features and they may represent 2 different pathologic conditions with equally different clinical implications.

PurposeTo characterize the change rate of ganglion cell complex (GCC) thickness and macular vessel density in healthy, preperimetric glaucoma and primary open-angle glaucoma (POAG) eyes.DesignProspective, longitudinal study.ParticipantsOne hundred thirty-nine eyes (23 healthy eyes, 36 preperimetric glaucoma eyes, and 80 POAG eyes) of 94 patients who had at least 3 visits were included from the Diagnostic Innovations in Glaucoma Study. The mean follow-up was 2.0 years for healthy eyes, 2.6 years for preperimetric glaucoma eyes, and 2.6 years for POAG eyes.MethodsOCT angiography (OCTA)-based vessel density and OCT-based structural thickness of the same 3×3-mm2 GCC scan slab were evaluated. The dynamic range-based normalized rates of vessel density and thickness change were calculated and compared within each diagnostic group. The association between the rates of thickness and vessel density change and potential factors were evaluated.Main outcome measuresThe rates of GCC thinning and macular vessel density loss.ResultsSignificant rates of GCC thinning and macular vessel density decrease were detectable in all diagnostic groups (all P < 0.05). In healthy eyes and preperimetric glaucoma eyes, the normalized rates of GCC thinning and macular vessel density decrease were comparable (all P > 0.1). In contrast, the normalized rate (mean, 95% confidence interval) of macular vessel density decrease in the POAG eyes (-7.12 [-8.36, -5.88]%/year) was significantly faster than GCC thinning (-2.13 [-3.35, -0.90]%/year; P < 0.001). In the POAG group, more than two thirds of the eyes showed faster macular vessel density decrease than GCC thinning; faster macular vessel density decrease rate was associated significantly with worse glaucoma severity (P = 0.037). The association between GCC thinning rate and glaucoma severity was not significant (P = 0.586). Intraocular pressure during follow-up significantly affected the rate of GCC thinning in all groups (all P &lt

PurposeTo compare performance of independently developed deep learning algorithms for detecting glaucoma from fundus photographs and to evaluate strategies for incorporating new data into models.MethodsTwo fundus photograph datasets from the Diagnostic Innovations in Glaucoma Study/African Descent and Glaucoma Evaluation Study and Matsue Red Cross Hospital were used to independently develop deep learning algorithms for detection of glaucoma at the University of California, San Diego, and the University of Tokyo. We compared three versions of the University of California, San Diego, and University of Tokyo models: original (no retraining), sequential (retraining only on new data), and combined (training on combined data). Independent datasets were used to test the algorithms.ResultsThe original University of California, San Diego and University of Tokyo models performed similarly (area under the receiver operating characteristic curve = 0.96 and 0.97, respectively) for detection of glaucoma in the Matsue Red Cross Hospital dataset, but not the Diagnostic Innovations in Glaucoma Study/African Descent and Glaucoma Evaluation Study data (0.79 and 0.92; P < .001), respectively. Model performance was higher when classifying moderate-to-severe compared with mild disease (area under the receiver operating characteristic curve = 0.98 and 0.91; P < .001), respectively. Models trained with the combined strategy generally had better performance across all datasets than the original strategy.ConclusionsDeep learning glaucoma detection can achieve high accuracy across diverse datasets with appropriate training strategies. Because model performance was influenced by the severity of disease, labeling, training strategies, and population characteristics, reporting accuracy stratified by relevant covariates is important for cross study comparisons.Translational relevanceHigh sensitivity and specificity of deep learning algorithms for moderate-to-severe glaucoma across diverse p

PurposeTo employ quantitative fundus autofluorescence (qAF) imaging in rhesus macaques to noninvasively assess retinal pigment epithelial (RPE) lipofuscin in nonhuman primates (NHPs) as a model of aging and age-related macular degeneration (AMD).MethodsThe qAF imaging was performed on eyes of 26 rhesus macaques (mean age 18.8 ± 8.2 years, range 4-27 years) with normal-appearing fundus or with age-related soft drusen using a confocal scanning laser ophthalmoscope with 488 nm excitation and an internal fluorescence reference. Eyes with soft drusen also underwent spectral-domain optical coherence tomography imaging to measure drusen volume and height of individual drusen lesions. The qAF levels were measured from the perifoveal annular ring (quantitative autofluorescence 8 [qAF8]) using the Delori grid, as well as focally over individual drusen lesions in this region. The association between qAF levels and age, sex, and drusen presence and volume were determined using multivariable regression analysis.ResultsMean qAF levels increased with age (P < 0.001) and were higher in females (P = 0.047). Eyes with soft drusen exhibited reduced mean qAF compared with age-matched normal eyes (P = 0.003), with greater drusen volume showing a trend toward decreased qAF levels. However, qAF levels are focally increased over most individual drusen (P < 0.001), with larger drusen appearing more hyperautofluorescent (R2 = 0.391, P < 0.001).ConclusionsIn rhesus macaques, qAF levels are increased with age and female sex, but decreased in eyes with soft drusen, similar to human AMD. However, drusen lesions appear hyperautofluorescent unlike those in humans, suggesting similarities and differences in RPE lipofuscin between humans and NHPs that may provide insight into drusen biogenesis and AMD pathogenesis.

PurposeTo describe the pathology of AMD in eyes with geographic atrophy (GA) using confocal scanning laser ophthalmoscopy (SLO) blue light autofluorescence (BAF), and near-infrared (IR) AF and to correlate it with the histology and immunohistochemistry analysis at the margins of the GA lesion.MethodsEnucleated, fixed eyes from seventeen donors with GA were imaged and analyzed by BAF-SLO, IRAF-SLO, and by fundus macroscopy (FM). Tissue from the margins of the GA lesions was cut and processed for resin embedding and histology or cryosectioning and fluorescence in the green and far-red channels, and immunohistochemistry to assess markers of inflammation. Isolated DNA from donors was genotyped for single nucleotide polymorphisms (SNPs) previously shown to be risk factors for the development and progression of AMD.ResultsAround the leading edge of the GA lesions we observed hypertrophic RPE cells with cytoplasm filled with granules fluorescent both in the far-red and green-red channels; abundant microglia and macrophage; deposition of complement factor H (CFH) in Bruch's membrane (BM) and increased membrane attack complex (MAC) on RPE cells.ConclusionsFluorescence imaging of cryosections of RPE cells around the leading edge of the GA lesions suggest that IRAF-SLO visualizes mostly melanin-related compounds. In addition, medium-size GA atrophy displayed the most significant changes in inflammation markers.

PurposeTo determine clear-cut distinctions between tractional and exudative intraretinal cystoid spaces subtypes.DesignRetrospective, multicenter, observational case series.MethodsA cohort of patients diagnosed with intraretinal cystoid spaces and imaged with optical coherence tomography (OCT), fluorescein angiography (FA), blue fundus autofluorescence (BFAF), en face OCT, and OCT angiography (OCT-A) was included in the study. All images were qualitatively and quantitatively evaluated.ResultsIn this study were included 72 eyes of 69 patients. Exudative intraretinal cystoid spaces (36/72 eyes, 50%) displayed a "petaloid" morphology as seen with en face OCT, FA, and BFAF. Tractional intraretinal cystoid spaces (24/72 eyes, 33.3%), displayed a radial "spoke-wheel" en face OCT pattern. There was no leakage with FA and BFAF did not reveal specific patterns. Eyes with full-thickness macular hole (FTMH, 12/72 eyes, 16.7%) displayed a "sunflower" en face OCT appearance. FTMH showed OCT, OCT-A, and BFAF features of both exudative and tractional cystoid spaces, but without any FA leakage. Inner nuclear layer (INL) thickness was significantly lower in tractional cystoid spaces (P < .001). There were a greater number of INL cystoid spaces in both the exudative and FTMH subgroups (P = .001). The surface area of INL cystoid spaces was significantly lower in the tractional subgroup (P < .001). There was a significant reduction of the microvascular density in eyes with exudative vs tractional (P = .002) and FTMH (P < .001) subgroups.ConclusionsExudative and tractional intraretinal cystoid spaces displayed characteristic multimodal imaging features and they may represent 2 different pathologic conditions with equally different clinical implications.

PurposeThe purpose of this study was to evaluate the ability to align two types of retinal images taken on different platforms; color fundus (CF) photographs and infrared scanning laser ophthalmoscope (IR SLO) images using mathematical warping and artificial intelligence (AI).MethodsWe collected 109 matched pairs of CF and IR SLO images. An AI algorithm utilizing two separate networks was developed. A style transfer network (STN) was used to segment vessel structures. A registration network was used to align the segmented images to each. Neither network used a ground truth dataset. A conventional image warping algorithm was used as a control. Software displayed image pairs as a 5 × 5 checkerboard grid composed of alternating subimages. This technique permitted vessel alignment determination by human observers and 5 masked graders evaluated alignment by the AI and conventional warping in 25 fields for each image.ResultsOur new AI method was superior to conventional warping at generating vessel alignment as judged by masked human graders (P < 0.0001). The average number of good/excellent matches increased from 90.5% to 94.4% with AI method.ConclusionsAI permitted a more accurate overlay of CF and IR SLO images than conventional mathematical warping. This is a first step toward developing an AI that could allow overlay of all types of fundus images by utilizing vascular landmarks.Translational relevanceThe ability to align and overlay imaging data from multiple instruments and manufacturers will permit better analysis of this complex data helping understand disease and predict treatment.

PurposeTo map and measure the depths of corneal neovascularization (NV) using 3-dimensional optical coherence tomography angiography (OCTA) at 2 different wavelengths.MethodsCorneal NV of varying severity, distribution, and underlying etiology was examined. Average NV depth and vessel density were measured using 840-nm spectral-domain OCTA and 1050-nm swept-source OCTA. The OCTA results were compared with clinical slit-lamp estimation of NV depth.ResultsTwelve eyes with corneal NV from 12 patients were imaged with OCTA. Clinically "superficial," "midstromal," and "deep" cases had an average vessel depth of 23%, 39%, and 66% on 1050-nm OCTA, respectively. Average vessel depth on OCTA followed a statistically significant ordinal trend according to the clinical classification of vessel depth (Jonckheere-Terpstra test, P < 0.001). In 8 cases where both 840-nm OCTA and 1050-nm OCTA were acquired, there was excellent agreement in the mean vessel depth between the 2 systems (concordance correlation coefficient = 0.94, P < 0.001). The average vessel density measured by 840-nm OCTA was higher (average 1.6-fold) than that measured by 1050-nm OCTA.ConclusionsCorneal OCTA was able to map corneal NV in 3 dimensions and measure vessel depth and density. The depth of corneal NV varied between different pathologies in a manner consistent with previous pathologic studies. The measured vessel density appeared to be affected by the interscan time, which affects blood flow velocity sensitivity, and the wavelength, which affects the ability to penetrate through opacity. These findings suggest possible clinical applications of OCTA for the diagnosis of corneal pathology and quantitative monitoring of therapeutic response in patients with corneal NV.

PurposeTo determine whether quantitative OCT angiography (OCTA) parameters can be used to distinguish among eyes at various stages of diabetic retinopathy (DR) in an urban safety-net hospital population.DesignProspective cross-sectional study.ParticipantsThree hundred twenty-nine eyes from 329 patients were included in this study: 90 nondiabetic patients, 170 diabetic patients without DR, 57 diabetic patients with mild to moderate nonproliferative DR (NPDR), and 12 diabetic patients with severe NPDR to proliferative DR.MethodsPatients underwent OCTA imaging and ultra-widefield fundus photography at Zuckerberg San Francisco General Hospital and Trauma Center between April and October 2018. For participants with diabetes, imaging was classified according to DR severity by a telemedicine reading center. Eight OCTA parameters were analyzed. Perfusion density and vessel length density (VD) were examined from both the superficial capillary plexus (SCP) and deep capillary plexus. The other 4 parameters were examined only from the SCP. Total extrafoveal avascular area (tEAA) was based on the area of absent capillary vessels. Foveal avascular zone (FAZ)-related metrics consisted of FAZ area, FAZ circularity index, and FAZ acircularity index.Main outcome measuresArea under the receiver operating characteristic curve (AUC) for OCTA parameters to distinguish among groups according to DR severity.ResultsAll OCTA parameters demonstrated a significant relationship with DR severity (P < 0.05). No significant difference was found when comparing nondiabetic participants versus diabetic participants without retinopathy. The FAZ area was the only metric that demonstrated a significant difference between genders: mean of 0.29±0.12 mm2 in men and 0.34±0.13 mm2 in women (P < 0.001). Receiver operating characteristic curve analyses showed that tEAA had the highest AUC when comparing various stages of the disease.ConclusionsIn this urban, public hospital population, quantification of

PurposeTo compare gradient-boosting classifier (GBC) analysis of optical coherence tomography angiography (OCTA)-measured vessel density (VD) and OCT-measured tissue thickness to standard OCTA VD and OCT thickness parameters for classifying healthy eyes and eyes with early to moderate glaucoma.DesignComparison of diagnostic tools.MethodsA total of 180 healthy eyes and 193 glaucomatous eyes with OCTA and OCT imaging of the macula and optic nerve head (ONH) were studied. Four GBCs were evaluated that combined 1) all macula VD and thickness measurements (Macula GBC), 2) all ONH VD and thickness measurements (ONH GBC), 3) all VD measurements from the macula and ONH (vessel density GBC), and 4) all thickness measurements from the macula and ONH (thickness GBC). ROC curve (AUROC) analyses compared the diagnostic accuracy of GBCs to that of standard instrument-provided parameters. A fifth GBC that combined all parameters (full GBC) also was investigated.ResultsGBCs had better diagnostic accuracy than standard OCTA and OCT parameters with AUROCs ranging from 0.90 to 0.93 and 0.64 to 0.91, respectively. The full GBC (AUROC = 0.93) performed significantly better than the ONH GBC (AUROC = 0.91; P = .036) and the vessel density GBC (AUROC = 0.90; P = .010). All other GBCs performed similarly. The mean relative influence of each parameter included in the full GBC identified a combination of macular thickness and ONH VD measurements as the greatest contributors.ConclusionsGBCs that combine OCTA and OCT macula and ONH measurements can improve diagnostic accuracy for glaucoma detection compared to most but not all instrument provided parameters.

The main objective of this study was to evaluate the retinas of severely or critically ill COVID-19 patients during their hospital stay, at varying time points after symptoms onset. This was a case series observed during May 2020 in two referral centers for COVID-19 treatment in Rio de Janeiro, Brazil. 47 eyes from 25 hospitalized patients with severe or critical confirmed illness were evaluated. A handheld retinal camera was used to acquire bilateral fundus images at several time points after symptoms onset. Electronic health records were retrospectively analyzed and clinical data collected. Severe and critical diseases were noticed in 52% (13/25) and 48% (12/25) of enrolled patients, respectively. Retinal changes were present in 12% (3/25) of patients: a 35 year-old male demonstrated bilateral nerve fiber layer infarcts and microhemorrhages in the papillomacular bundle, but required mechanical ventilation and developed severe anemia and systemic hypotension, acute kidney injury and neurologic symptoms during the course of the disease (critical illness); a 56 year-old male, who required full enoxaparin anticoagulation due to particularly elevated D-dimer (>5.0 mcg/mL), demonstrated unilateral and isolated flame-shaped hemorrhages; and a 49 year-old hypertensive male showed bilateral and discrete retinal dot and blot microhemorrhages. The other 22 patients evaluated did not demonstrate convincing retinal changes upon examination. There was no correlation between disease severity and admission serum levels of CRP, D-dimer and ferritin. This was the first study to show that vascular retinal changes may be present in not insignificant numbers of severe or critical COVID-19 inpatients. These retinal changes, only seen after morbid developments, were likely secondary to clinical intercurrences or comorbidities instead of a direct damage by SARS-CoV-2, and may be important and easily accessible outcome measures of therapeutic interventions and sentinels of neurologic a

Purpose: To assess the performance of deep learning algorithms for different tasks in retinal fundus images: (1) detection of retinal fundus images versus optical coherence tomography (OCT) or other images, (2) evaluation of good quality retinal fundus images, (3) distinction between right eye (OD) and left eye (OS) retinal fundus images,(4) detection of age-related macular degeneration (AMD) and (5) detection of referable glaucomatous optic neuropathy (GON). Patients and Methods: Five algorithms were designed. Retrospective study from a database of 306,302 images, Optretina’s tagged dataset. Three different ophthalmologists, all retinal specialists, classified all images. The dataset was split per patient in a training (80%) and testing (20%) splits. Three different CNN architectures were employed, two of which were custom designed to minimize the number of parameters with minimal impact on its accuracy. Main outcome measure was area under the curve (AUC) with accuracy, sensitivity and specificity. Results: Determination of retinal fundus image had AUC of 0.979 with an accuracy of 96% (sensitivity 97.7%, specificity 92.4%). Determination of good quality retinal fundus image had AUC of 0.947, accuracy 91.8% (sensitivity 96.9%, specificity 81.8%). Algorithm for OD/OS had AUC 0.989, accuracy 97.4%. AMD had AUC of 0.936, accuracy 86.3% (sensitivity 90.2% specificity 82.5%), GON had AUC of 0.863, accuracy 80.2% (sensitivity 76.8%, specificity 83.8%). Conclusion: Deep learning algorithms can differentiate a retinal fundus image from other images. Algorithms can evaluate the quality of an image, discriminate between right or left eye and detect the presence of AMD and GON with a high level of accuracy, sensitivity and specificity., Peer Reviewed, Postprint (published version)

PURPOSE: To characterize the change rate of ganglion cell complex (GCC) thickness and macular vessel density in healthy, preperimetric glaucoma and primary open-angle glaucoma (POAG) eyes. DESIGN: Prospective, longitudinal study. PARTICIPANTS: One hundred thirty-nine eyes (23 healthy eyes, 36 preperimetric glaucoma eyes, and 80 POAG eyes) of 94 patients who had at least 3 visits were included from the Diagnostic Innovations in Glaucoma Study. The mean follow-up was 2.0 years for healthy eyes, 2.6 years for preperimetric glaucoma eyes, and 2.6 years for POAG eyes. METHODS: OCT angiography (OCTA)-based vessel density and OCT-based structural thickness of the same 3×3-mm2 GCC scan slab were evaluated. The dynamic range-based normalized rates of vessel density and thickness change were calculated and compared within each diagnostic group. The association between the rates of thickness and vessel density change and potential factors were evaluated. MAIN OUTCOME MEASURES: The rates of GCC thinning and macular vessel density loss. RESULTS: Significant rates of GCC thinning and macular vessel density decrease were detectable in all diagnostic groups (all P < 0.05). In healthy eyes and preperimetric glaucoma eyes, the normalized rates of GCC thinning and macular vessel density decrease were comparable (all P > 0.1). In contrast, the normalized rate (mean, 95% confidence interval) of macular vessel density decrease in the POAG eyes (-7.12 [-8.36, -5.88]%/year) was significantly faster than GCC thinning (-2.13 [-3.35, -0.90]%/year; P < 0.001). In the POAG group, more than two thirds of the eyes showed faster macular vessel density decrease than GCC thinning; faster macular vessel density decrease rate was associated significantly with worse glaucoma severity (P = 0.037). The association between GCC thinning rate and glaucoma severity was not significant (P = 0.586). Intraocular pressure during follow-up significantly affected the rate of GCC thinning in all

Mutations in KCNJ13 are associated with two retinal disorders; Leber congenital amaurosis (LCA) and snowflake vitreoretinal degeneration (SVD). We describe a novel fibrovascular proliferation in the retina of two affected members of a KCNJ13-related LCA family with a homozygous c.458C > T, p.(Thr153Ile) missense mutation. Optical coherence tomography retinal imaging of the kcnj13 mutant zebrafish (obelixtd15 c.502T > C, p.[Phe168Leu]) revealed a late onset retinal degeneration at 12 months, with retinal thinning and associated retinovascular changes, including increased vessel calibre and vitreous deposits. Both human and zebrafish variants are missense and located within the conserved transmembrane M2 protein domain, suggesting that disruption of this region may contribute to retinovascular changes as an additional feature to the previously described LCA phenotype. Close monitoring of other patients with similar mutations may be required to minimise the ensuing retinal damage.

Postprint (author's final draft)

Choroidal melanocytes (HCMs) are melanin-producing cells in the vascular uvea of the human eye (iris, ciliary body and choroid). These cranial neural crest-derived cells migrate to populate a mesodermal microenvironment, and display cellular functions and extracellular interactions that are biologically distinct to skin melanocytes. HCMs (and melanins) are important in normal human eye physiology with roles including photoprotection, regulation of oxidative damage and immune responses. To extend knowledge of cytoplasmic melanins and melanosomes in label-free HCMs, a non-invasive 'fit-free' approach, combining 2-photon excitation fluorescence lifetimes and emission spectral imaging with phasor plot segmentation was applied. Intracellular melanin-mapped FLIM phasors showed a linear distribution indicating that HCM melanins are a ratio of two fluorophores, eumelanin and pheomelanin. A quantitative histogram of HCM melanins was generated by identifying the image pixel fraction contributed by phasor clusters mapped to varying eumelanin/pheomelanin ratio. Eumelanin-enriched dark HCM regions mapped to phasors with shorter lifetimes and longer spectral emission (580-625 nm) and pheomelanin-enriched lighter pigmented HCM regions mapped to phasors with longer lifetimes and shorter spectral emission (550-585 nm). Overall, we demonstrated that these methods can identify and quantitatively profile the heterogeneous eumelanins/pheomelanins within in situ HCMs, and visualize melanosome spatial distributions, not previously reported for these cells.

ImportanceA deep learning system (DLS) that could automatically detect glaucomatous optic neuropathy (GON) with high sensitivity and specificity could expedite screening for GON.ObjectiveTo establish a DLS for detection of GON using retinal fundus images and glaucoma diagnosis with convoluted neural networks (GD-CNN) that has the ability to be generalized across populations.Design, setting, and participantsIn this cross-sectional study, a DLS for the classification of GON was developed for automated classification of GON using retinal fundus images obtained from the Chinese Glaucoma Study Alliance, the Handan Eye Study, and online databases. The researchers selected 241 032 images were selected as the training data set. The images were entered into the databases on June 9, 2009, obtained on July 11, 2018, and analyses were performed on December 15, 2018. The generalization of the DLS was tested in several validation data sets, which allowed assessment of the DLS in a clinical setting without exclusions, testing against variable image quality based on fundus photographs obtained from websites, evaluation in a population-based study that reflects a natural distribution of patients with glaucoma within the cohort and an additive data set that has a diverse ethnic distribution. An online learning system was established to transfer the trained and validated DLS to generalize the results with fundus images from new sources. To better understand the DLS decision-making process, a prediction visualization test was performed that identified regions of the fundus images utilized by the DLS for diagnosis.ExposuresUse of a deep learning system.Main outcomes and measuresArea under the receiver operating characteristics curve (AUC), sensitivity and specificity for DLS with reference to professional graders.ResultsFrom a total of 274 413 fundus images initially obtained from CGSA, 269 601 images passed initial image quality review and were graded for GON. A total of 241 032 images

Mutations in KCNJ13 are associated with two retinal disorders; Leber congenital amaurosis (LCA) and snowflake vitreoretinal degeneration (SVD). We describe a novel fibrovascular proliferation in the retina of two affected members of a KCNJ13-related LCA family with a homozygous c.458C > T, p.(Thr153Ile) missense mutation. Optical coherence tomography retinal imaging of the kcnj13 mutant zebrafish (obelixtd15 c.502T > C, p.[Phe168Leu]) revealed a late onset retinal degeneration at 12 months, with retinal thinning and associated retinovascular changes, including increased vessel calibre and vitreous deposits. Both human and zebrafish variants are missense and located within the conserved transmembrane M2 protein domain, suggesting that disruption of this region may contribute to retinovascular changes as an additional feature to the previously described LCA phenotype. Close monitoring of other patients with similar mutations may be required to minimise the ensuing retinal damage.

This report describes the first documented case of choriovitreal neovascularization (CVNV) occurring after infectious chorioretinitis, with OCT-angiography (OCT-A) demonstrating flow through the vitreal portion of the lesion, and response to anti-VEGF therapy.