A concentrated machine learning-based classification system for age-related macular degeneration (AMD) diagnosis using fundus images.

The increase in eye disorders among older individuals has raised concerns, necessitating early detection through regular eye examinations. Age-related macular degeneration (AMD), a prevalent condition in individuals over 45, is a leading cause of vision impairment in the elderly. This paper presents a comprehensive computer-aided diagnosis (CAD) framework to categorize fundus images into geographic atrophy (GA), intermediate AMD, normal, and wet AMD categories. This is crucial for early detection and precise diagnosis of age-related macular degeneration (AMD), enabling timely intervention and personalized treatment strategies. We have developed a novel system that extracts both local and global appearance markers from fundus images. These markers are obtained from the entire retina and iso-regions aligned with the optical disc. Applying weighted majority voting on the best classifiers improves performance, resulting in an accuracy of 96.85%, sensitivity of 93.72%, specificity of 97.89%, precision of 93.86%, F1 of 93.72%, ROC of 95.85%, balanced accuracy of 95.81%, and weighted sum of 95.38%. This system not only achieves high accuracy but also provides a detailed assessment of the severity of each retinal region. This approach ensures that the final diagnosis aligns with the physician's understanding of AMD, aiding them in ongoing treatment and follow-up for AMD patients.

A Deep-Learning Algorithm to Predict Short-Term Progression to Geographic Atrophy on Spectral-Domain Optical Coherence Tomography.

Importance: The identification of patients at risk of progressing from intermediate age-related macular degeneration (iAMD) to geographic atrophy (GA) is essential for clinical trials aimed at preventing disease progression. DeepGAze is a fully automated and accurate convolutional neural network-based deep learning algorithm for predicting progression from iAMD to GA within 1 year from spectral-domain optical coherence tomography (SD-OCT) scans. Objective: To develop a deep-learning algorithm based on volumetric SD-OCT scans to predict the progression from iAMD to GA during the year following the scan. Design, Setting, and Participants: This retrospective cohort study included participants with iAMD at baseline and who either progressed or did not progress to GA within the subsequent 13 months. Participants were included from centers in 4 US states. Data set 1 included patients from the Age-Related Eye Disease Study 2 AREDS2 (Ancillary Spectral-Domain Optical Coherence Tomography) A2A study (July 2008 to August 2015). Data sets 2 and 3 included patients with imaging taken in routine clinical care at a tertiary referral center and associated satellites between January 2013 and January 2023. The stored imaging data were retrieved for the purpose of this study from July 1, 2022, to February 1, 2023. Data were analyzed from May 2021 to July 2023. Exposure: A position-aware convolutional neural network with proactive pseudointervention was trained and cross-validated on Bioptigen SD-OCT volumes (data set 1) and validated on 2 external data sets comprising Heidelberg Spectralis SD-OCT scans (data sets 2 and 3). Main Outcomes and Measures: Prediction of progression to GA within 13 months was evaluated with area under the receiver-operator characteristic curves (AUROC) as well as area under the precision-recall curve (AUPRC), sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. Results: The study included a total of 417 patients: 316 in data set 1 (mean [SD] age, 74 [8]; 185 [59%] female), 53 in data set 2, (mean [SD] age, 83 [8]; 32 [60%] female), and 48 in data set 3 (mean [SD] age, 81 [8]; 32 [67%] female). The AUROC for prediction of progression from iAMD to GA within 1 year was 0.94 (95% CI, 0.92-0.95; AUPRC, 0.90 [95% CI, 0.85-0.95]; sensitivity, 0.88 [95% CI, 0.84-0.92]; specificity, 0.90 [95% CI, 0.87-0.92]) for data set 1. The addition of expert-annotated SD-OCT features to the model resulted in no improvement compared to the fully autonomous model (AUROC, 0.95; 95% CI, 0.92-0.95; P = .19). On an independent validation data set (data set 2), the model predicted progression to GA with an AUROC of 0.94 (95% CI, 0.91-0.96; AUPRC, 0.92 [0.89-0.94]; sensitivity, 0.91 [95% CI, 0.74-0.98]; specificity, 0.80 [95% CI, 0.63-0.91]). At a high-specificity operating point, simulated clinical trial recruitment was enriched for patients progressing to GA within 1 year by 8.3- to 20.7-fold (data sets 2 and 3). Conclusions and Relevance: The fully automated, position-aware deep-learning algorithm assessed in this study successfully predicted progression from iAMD to GA over a clinically meaningful time frame. The ability to predict imminent GA progression could facilitate clinical trials aimed at preventing the condition and could guide clinical decision-making regarding screening frequency or treatment initiation.

Geographic Atrophy Segmentation Using Multimodal Deep Learning.

Purpose: To examine deep learning (DL)-based methods for accurate segmentation of geographic atrophy (GA) lesions using fundus autofluorescence (FAF) and near-infrared (NIR) images. Methods: This retrospective analysis utilized imaging data from study eyes of patients enrolled in Proxima A and B (NCT02479386; NCT02399072) natural history studies of GA. Two multimodal DL networks (UNet and YNet) were used to automatically segment GA lesions on FAF; segmentation accuracy was compared with annotations by experienced graders. The training data set comprised 940 image pairs (FAF and NIR) from 183 patients in Proxima B; the test data set comprised 497 image pairs from 154 patients in Proxima A. Dice coefficient scores, Bland-Altman plots, and Pearson correlation coefficient (r) were used to assess performance. Results: On the test set, Dice scores for the DL network to grader comparison ranged from 0.89 to 0.92 for screening visit; Dice score between graders was 0.94. GA lesion area correlations (r) for YNet versus grader, UNet versus grader, and between graders were 0.981, 0.959, and 0.995, respectively. Longitudinal GA lesion area enlargement correlations (r) for screening to 12 months (n = 53) were lower (0.741, 0.622, and 0.890, respectively) compared with the cross-sectional results at screening. Longitudinal correlations (r) from screening to 6 months (n = 77) were even lower (0.294, 0.248, and 0.686, respectively). Conclusions: Multimodal DL networks to segment GA lesions can produce accurate results comparable with expert graders. Translational Relevance: DL-based tools may support efficient and individualized assessment of patients with GA in clinical research and practice.

Clinical validation for automated geographic atrophy monitoring on OCT under complement inhibitory treatment.

Geographic atrophy (GA) represents a late stage of age-related macular degeneration, which leads to irreversible vision loss. With the first successful therapeutic approach, namely complement inhibition, huge numbers of patients will have to be monitored regularly. Given these perspectives, a strong need for automated GA segmentation has evolved. The main purpose of this study was the clinical validation of an artificial intelligence (AI)-based algorithm to segment a topographic 2D GA area on a 3D optical coherence tomography (OCT) volume, and to evaluate its potential for AI-based monitoring of GA progression under complement-targeted treatment. 100 GA patients from routine clinical care at the Medical University of Vienna for internal validation and 113 patients from the FILLY phase 2 clinical trial for external validation were included. Mean Dice Similarity Coefficient (DSC) was 0.86 ± 0.12 and 0.91 ± 0.05 for total GA area on the internal and external validation, respectively. Mean DSC for the GA growth area at month 12 on the external test set was 0.46 ± 0.16. Importantly, the automated segmentation by the algorithm corresponded to the outcome of the original FILLY trial measured manually on fundus autofluorescence. The proposed AI approach can reliably segment GA area on OCT with high accuracy. The availability of such tools represents an important step towards AI-based monitoring of GA progression under treatment on OCT for clinical management as well as regulatory trials.

Novel Fractal-Based Sub-RPE Compartment OCT Radiomics Biomarkers Are Associated With Subfoveal Geographic Atrophy in Dry AMD.

OBJECTIVE: The purpose of this study was to quantitatively characterize the shape of the sub-retinal pigment epithelium (sub-RPE, i.e., space bounded by RPE and Bruch's membrane) compartment on SD-OCT using fractal dimension (FD) features and evaluate their impact on risk of subfoveal geographic atrophy (sfGA) progression. METHODS: This was an IRB-approved retrospective study of 137 subjects with dry age-related macular degeneration (AMD) with subfoveal GA. Based on sfGA status at year five, eyes were categorized as "Progressors" and "Non-progressors". FD analysis allows quantification of the degree of shape complexity and architectural disorder associated with a structure. To characterize the structural irregularities along the sub-RPE surface between the two groups of patients, a total of 15 shape descriptors of FD were extracted from the sub-RPE compartment of baseline OCT scans. The top four features were identified using minimum Redundancy maximum Relevance (mRmR) feature selection method and evaluated with Random Forest (RF) classifier using three-fold cross validation from the training set (N = 90). Classifier performance was subsequently validated on the independent test set (N = 47). RESULTS: Using the top four FD features, a RF classifier yielded an AUC of 0.85 on the independent test set. Mean fractal entropy (p-value = 4.8e-05) was identified as the most significant biomarker; higher values of entropy being associated with greater shape disorder and risk for sfGA progression. CONCLUSIONS: FD assessment holds promise for identifying high-risk eyes for GA progression. SIGNIFICANCE: With further validation, FD features could be potentially used for clinical trial enrichment and assessments for therapeutic response in dry AMD patients.

Quantitative multimodal imaging of extensive macular atrophy with pseudodrusen and geographic atrophy with diffuse trickling pattern.

To compare clinical and imaging characteristics of extensive macular atrophy with pseudodrusen-like appearance (EMAP) versus diffuse-trickling geographic atrophy (DTGA) and non-diffuse-trickling geographic atrophy (nDTGA) phenotypes of age-related macular degeneration. Prospective, observational study performed in the Ophthalmology Department of IRCCS San Raffaele Hospital between January 2015 and January 2021. Patients examination included fundus autofluorescence (FAF) and optical coherence tomography at baseline and follow-up visits. We measured subfoveal choroidal thickness (SCT), Sattler/choroid ratio (SCR), choroidal vascularity index and ellipsoid zone disruption distance on OCT scans. We calculated progression rates and circularity of the atrophic lesions on FAF images. These variables were compared between the three groups and correlations with progression rates and visual acuity were assessed. Sixty-three eyes from 63 patients were included: 18 with EMAP, 18 with DTGA and 27 with nDTGA. Mean follow-up was 3.73 ± 2.12 years. EMAP and DTGA shared a faster progression, lower circularity and SCR, and higher EZ disruption distance than nDTGA, while SCT and CVI were similar between the three groups. Baseline circularity and SCR correlated with progression rates. EMAP and DTGA show similar OCT and FAF characteristics, which differ from nDTGA.

Reverse engineering for reconstructing baseline features of dry age-related macular degeneration in optical coherence tomography.

Age-related macular degeneration (AMD) is the most widespread cause of blindness and the identification of baseline AMD features or biomarkers is critical for early intervention. Optical coherence tomography (OCT) imaging produces a 3D volume consisting of cross sections of retinal tissue while fundus fluorescence (FAF) imaging produces a 2D mapping of retina. FAF has been a good standard for assessing dry AMD late-stage geographic atrophy (GA) while OCT has been used for assessing early AMD biomarkers beyond as well. However, previous approaches in large extent defined AMD features subjectively based on clinicians' observation. Deep learning-an objective artificial intelligence approach, may enable to discover 'true' salient AMD features. We develop a novel reverse engineering approach which bases on the backbone of a fully convolutional neural network to objectively identify and visualize AMD early biomarkers in OCT from baseline exams before significant atrophy occurs. Utilizing manually annotated GA regions on FAF from a follow-up visit as ground truth, we segment GA regions and reconstruct early AMD features in baseline OCT volumes. In this preliminary exploration, compared with ground truth, we achieve baseline GA segmentation accuracy of 0.95 and overlapping ratio of 0.65. The reconstructions consistently highlight that large druse and druse clusters with or without mixed hyper-reflective focus lesion on baseline OCT cause the conversion of GA after 12 months. However, hyper-reflective focus lesions and subretinal drusenoid deposit lesions alone are not seen such conversion after 12 months. Further research with larger dataset would be needed to verify these findings.

Comparison of Multimodal Imaging for the Characterization of Geographic Atrophy.

Purpose: The purpose of this study was to compare the performances of infrared (IR), fundus autofluorescence (FAF), and multicolor (MC) imaging in the characterization of geographic atrophy, with a focus on the possibility to detect incomplete retinal pigmented and outer retinal atrophy (iRORA) on en face imaging. Methods: The ground truth was established by two graders evaluating atrophy on spectral-domain optical coherence tomography (SD-OCT) images. A score for visibility of foveal sparing and margins of atrophy was attributed. Measurement of the atrophic area and the fovea-to-margin distance were performed. Accuracy of detection of foveal sparing was evaluated through comparison with B-scan images ground truth, with/without the inclusion of patients with foveal iRORA. Results: Seventy patients were included in this study. Foveal sparing and atrophy's margins subjective visibility were significantly higher rated on MC images compared to IR and FAF (P < 0.005 and P < 0.001). Agreement with OCT B-scan assessed foveal sparing revealed a significantly higher area under receiver operating characteristic curves (AUROC) for MC images at the analysis performed both with (0.876) and without (0.853) inclusion of patients with foveal iRORA (P < 0.001 and P = 0.006). Quantitative measurements revealed lower atrophy extension (P = 0.026) and fovea-to-margin distance (P = 0.019) with MC imaging. Conclusions: MC imaging performed better at foveal sparing assessment, especially in the setting of foveal iRORA. MC also resulted in higher visibility of atrophy's margins, lower atrophy extension measurements, and lower distance from the fovea to atrophy's margins compared to both FAF and IR. Translational Relevance: MC rated significantly higher in foveal sparing and atrophy detection, higher visibility of atrophy's margins, lower atrophy extension measurements, and lower distance from the fovea to atrophy's margins, compared to FAF and IR.

Prediction of visual function from automatically quantified optical coherence tomography biomarkers in patients with geographic atrophy using machine learning.

Geographic atrophy (GA) is a vision-threatening manifestation of age-related macular degeneration (AMD), one of the leading causes of blindness globally. Objective, rapid, reliable, and scalable quantification of GA from optical coherence tomography (OCT) retinal scans is necessary for disease monitoring, prognostic research, and clinical endpoints for therapy development. Such automatically quantified biomarkers on OCT are likely to further elucidate structure-function correlation in GA and thus the pathophysiological mechanisms of disease development and progression. In this work, we aimed to predict visual function with machine-learning applied to automatically acquired quantitative imaging biomarkers in GA. A post-hoc analysis of data from a clinical trial and routine clinical care was conducted. A deep-learning automated segmentation model was applied on OCT scans from 476 eyes (325 patients) with GA. A separate machine learning prediction model (Random Forest) used the resultant quantitative OCT (qOCT) biomarkers to predict cross-sectional visual acuity under standard (VA) and low luminance (LLVA). The primary outcome was regression coefficient (r2) and mean absolute error (MAE) for cross-sectional VA and LLVA in Early Treatment Diabetic Retinopathy Study (ETDRS) letters. OCT parameters were predictive of VA (r2 0.40 MAE 11.7 ETDRS letters) and LLVA (r2 0.25 MAE 12.1). Normalised random forest feature importance, as a measure of the predictive value of the three constituent features of GA; retinal pigment epithelium (RPE)-loss, photoreceptor degeneration (PDR), hypertransmission and their locations, was reported both on voxel-level heatmaps and ETDRS-grid subfields. The foveal region (46.5%) and RPE-loss (31.1%) had greatest predictive importance for VA. For LLVA, however, non-foveal regions (74.5%) and PDR (38.9%) were most important. In conclusion, automated qOCT biomarkers demonstrate predictive significance for VA and LLVA in GA. LLVA is itself predictive of GA progression, implying that the predictive qOCT biomarkers provided by our model are also prognostic.

Evolving Patterns of Hyperfluorescent Fundus Autofluorescence Accompany Retinal Atrophy in the Rat and Mimic Atrophic Age-Related Macular Degeneration.

PURPOSE: Complex two-dimensional (2D) patterns of hyperfluorescent short-wave fundus autofluorescence (FAF) at the border of geographic atrophy (GA) can predict its expansion in patients with late non-exudative "dry" AMD. However, preclinical models do not phenocopy this important feature of disease. We sought to describe the spatiotemporal changes in hyperfluorescent FAF patterns that occur following acute oxidative stress, potentially in association with GA expansion. METHODS: Sprague Dawley rats (n = 54) received systemic sodium iodate (25-45 mg/kg, n = 90 eyes) or saline (n = 18 eyes) and underwent serial full fundus imaging by confocal scanning laser ophthalmoscopy, including blue FAF and delayed near-infrared analysis. Composite images of the fundus were assembled, and the 2D patterns were described qualitatively and quantitatively. A subset of eyes underwent tissue analysis, and four underwent optical coherence tomography (OCT) imaging. RESULTS: Reproducibly changing, complex patterns of hyperfluorescent FAF emerge at the borders of toxin-induced damage; however, in the absence of GA expansion, they percolate inward within the region of retinal pigment epithelium loss, evolving, maturing, and senescing in situ over time. Unexpectedly, the late FAF patterns most closely resemble the diffuse tricking form of clinical disease. A five-stage classification system is presented. CONCLUSIONS: Longitudinal, full-fundus imaging of outer retinal atrophy in the rat eye identifies evolving, complex patterns of hyperfluorescent FAF that phenocopy aspects of disease. TRANSLATIONAL RELEVANCE: This work provides a novel tool to assess hyperfluorescent FAF in association with progressive retinal atrophy, a therapeutic target in late AMD.

DEVELOPMENT AND VALIDATION OF AN EXPLAINABLE ARTIFICIAL INTELLIGENCE FRAMEWORK FOR MACULAR DISEASE DIAGNOSIS BASED ON OPTICAL COHERENCE TOMOGRAPHY IMAGES.

PURPOSE: To develop and validate an artificial intelligence framework for identifying multiple retinal lesions at image level and performing an explainable macular disease diagnosis at eye level in optical coherence tomography images. METHODS: A total of 26,815 optical coherence tomography images were collected from 865 eyes, and 9 retinal lesions and 3 macular diseases were labeled by ophthalmologists, including diabetic macular edema and dry/wet age-related macular degeneration. We applied deep learning to classify retinal lesions at image level and random forests to achieve an explainable disease diagnosis at eye level. The performance of the integrated two-stage framework was evaluated and compared with human experts. RESULTS: On testing data set of 2,480 optical coherence tomography images from 80 eyes, the deep learning model achieved an average area under curve of 0.978 (95% confidence interval, 0.971-0.983) for lesion classification. In addition, random forests performed accurate disease diagnosis with a 0% error rate, which achieved the same accuracy as one of the human experts and was better than the other three experts. It also revealed that the detection of specific lesions in the center of macular region had more contribution to macular disease diagnosis. CONCLUSION: The integrated method achieved high accuracy and interpretability in retinal lesion classification and macular disease diagnosis in optical coherence tomography images and could have the potential to facilitate the clinical diagnosis.

CHOROIDAL VASCULARITY INDEX IS ASSOCIATED WITH GEOGRAPHIC ATROPHY PROGRESSION.

PURPOSE: To investigate the correlation between choroidal vascularity index and the enlargement of geographic atrophy (GA) lesion secondary to age-related macular degeneration during the 2-year follow-up. METHODS: In this longitudinal observational study, 26 eyes (26 patients, mean age 75.7 ± 8.8 years) affected by GA were included. Choroidal vascularity index was calculated in the subfoveal 3000-µm area. The main outcome measure included correlation analysis between baseline choroidal vascularity index and the rate of GA enlargement. RESULTS: During the 2-year follow-up, the mean GA area increased from 6.99 ± 5.28 mm2 to 10.69 ± 6.61 mm2(P < 0.001), accounting for a growth rate of 0.35 ± 0.20 and 0.31 ± 0.17 mm/year after the square root transformation in the first and second year of follow-up, respectively. Stromal choroidal area significantly decreased during the 2-year follow-up (P = 0.002). Interestingly, there was a significant correlation between the baseline choroidal vascularity index and the rate of GA enlargement (r=-0.432, P = 0.027) and between stromal choroidal area and the rate of GA enlargement (r = 0.422, P = 0.032). No other significant relationship was disclosed among choroidal parameters with the rate of GA enlargement. CONCLUSION: Choroidal vascularity index impairment is strictly related to the rate of GA enlargement during the 1-year and 2-year follow-up in patients affected by GA. For this reason, choroidal vascularity index could be considered a predictor of GA progression in the clinical setting, and it could be considered as a new potential biomarker in the efficacy evaluation of new GA interventions.

Fully-automated atrophy segmentation in dry age-related macular degeneration in optical coherence tomography.

Age-related macular degeneration (AMD) is a progressive retinal disease, causing vision loss. A more detailed characterization of its atrophic form became possible thanks to the introduction of Optical Coherence Tomography (OCT). However, manual atrophy quantification in 3D retinal scans is a tedious task and prevents taking full advantage of the accurate retina depiction. In this study we developed a fully automated algorithm segmenting Retinal Pigment Epithelial and Outer Retinal Atrophy (RORA) in dry AMD on macular OCT. 62 SD-OCT scans from eyes with atrophic AMD (57 patients) were collected and split into train and test sets. The training set was used to develop a Convolutional Neural Network (CNN). The performance of the algorithm was established by cross validation and comparison to the test set with ground-truth annotated by two graders. Additionally, the effect of using retinal layer segmentation during training was investigated. The algorithm achieved mean Dice scores of 0.881 and 0.844, sensitivity of 0.850 and 0.915 and precision of 0.928 and 0.799 in comparison with Expert 1 and Expert 2, respectively. Using retinal layer segmentation improved the model performance. The proposed model identified RORA with performance matching human experts. It has a potential to rapidly identify atrophy with high consistency.

Intraocular iron injection induces oxidative stress followed by elements of geographic atrophy and sympathetic ophthalmia.

Iron has been implicated in the pathogenesis of age-related retinal diseases, including age-related macular degeneration (AMD). Previous work showed that intravitreal (IVT) injection of iron induces acute photoreceptor death, lipid peroxidation, and autofluorescence (AF). Herein, we extend this work, finding surprising chronic features of the model: geographic atrophy and sympathetic ophthalmia. We provide new mechanistic insights derived from focal AF in the photoreceptors, quantification of bisretinoids, and localization of carboxyethyl pyrrole, an oxidized adduct of docosahexaenoic acid associated with AMD. In mice given IVT ferric ammonium citrate (FAC), RPE died in patches that slowly expanded at their borders, like human geographic atrophy. There was green AF in the photoreceptor ellipsoid, a mitochondria-rich region, 4 h after injection, followed later by gold AF in rod outer segments, RPE and subretinal myeloid cells. The green AF signature is consistent with flavin adenine dinucleotide, while measured increases in the bisretinoid all-trans-retinal dimer are consistent with the gold AF. FAC induced formation carboxyethyl pyrrole accumulation first in photoreceptors, then in RPE and myeloid cells. Quantitative PCR on neural retina and RPE indicated antioxidant upregulation and inflammation. Unexpectedly, reminiscent of sympathetic ophthalmia, autofluorescent myeloid cells containing abundant iron infiltrated the saline-injected fellow eyes only if the contralateral eye had received IVT FAC. These findings provide mechanistic insights into the potential toxicity caused by AMD-associated retinal iron accumulation. The mouse model will be useful for testing antioxidants, iron chelators, ferroptosis inhibitors, anti-inflammatory medications, and choroidal neovascularization inhibitors.

Clinically relevant deep learning for detection and quantification of geographic atrophy from optical coherence tomography: a model development and external validation study.

BACKGROUND: Geographic atrophy is a major vision-threatening manifestation of age-related macular degeneration, one of the leading causes of blindness globally. Geographic atrophy has no proven treatment or method for easy detection. Rapid, reliable, and objective detection and quantification of geographic atrophy from optical coherence tomography (OCT) retinal scans is necessary for disease monitoring, prognostic research, and to serve as clinical endpoints for therapy development. To this end, we aimed to develop and validate a fully automated method to detect and quantify geographic atrophy from OCT. METHODS: We did a deep-learning model development and external validation study on OCT retinal scans at Moorfields Eye Hospital Reading Centre and Clinical AI Hub (London, UK). A modified U-Net architecture was used to develop four distinct deep-learning models for segmentation of geographic atrophy and its constituent retinal features from OCT scans acquired with Heidelberg Spectralis. A manually segmented clinical dataset for model development comprised 5049 B-scans from 984 OCT volumes selected randomly from 399 eyes of 200 patients with geographic atrophy secondary to age-related macular degeneration, enrolled in a prospective, multicentre, phase 2 clinical trial for the treatment of geographic atrophy (FILLY study). Performance was externally validated on an independently recruited dataset from patients receiving routine care at Moorfields Eye Hospital (London, UK). The primary outcome was segmentation and classification agreement between deep-learning model geographic atrophy prediction and consensus of two independent expert graders on the external validation dataset. FINDINGS: The external validation cohort included 884 B-scans from 192 OCT volumes taken from 192 eyes of 110 patients as part of real-life clinical care at Moorfields Eye Hospital between Jan 1, 2016, and Dec, 31, 2019 (mean age 78·3 years [SD 11·1], 58 [53%] women). The resultant geographic atrophy deep-learning model produced predictions similar to consensus human specialist grading on the external validation dataset (median Dice similarity coefficient [DSC] 0·96 [IQR 0·10]; intraclass correlation coefficient [ICC] 0·93) and outperformed agreement between human graders (DSC 0·80 [0·28]; ICC 0·79). Similarly, the three independent feature-specific deep-learning models could accurately segment each of the three constituent features of geographic atrophy: retinal pigment epithelium loss (median DSC 0·95 [IQR 0·15]), overlying photoreceptor degeneration (0·96 [0·12]), and hypertransmission (0·97 [0·07]) in the external validation dataset versus consensus grading. INTERPRETATION: We present a fully developed and validated deep-learning composite model for segmentation of geographic atrophy and its subtypes that achieves performance at a similar level to manual specialist assessment. Fully automated analysis of retinal OCT from routine clinical practice could provide a promising horizon for diagnosis and prognosis in both research and real-life patient care, following further clinical validation FUNDING: Apellis Pharmaceuticals.

Normalization of series of fundus images to monitor the geographic atrophy growth in dry age-related macular degeneration.

BACKGROUND AND OBJECTIVE: Age-related macular degeneration (ARMD) is a degenerative disease that affects the retina, and the leading cause of visual loss. In its dry form, the pathology is characterized by the progressive, centrifugal expansion of retinal lesions, called geographic atrophy (GA). In infrared eye fundus images, the GA appears as localized bright areas and its growth can be observed in series of images acquired at regular time intervals. However, illumination distortions between the images make impossible the direct comparison of intensities in order to study the GA progress. Here, we propose a new method to compensate for illumination distortion between images. METHODS: We process all images of the series so that any two images have comparable gray levels. Our approach relies on an illumination/reflectance model. We first estimate the pixel-wise illumination ratio between any two images of the series, in a recursive way; then we correct each image against all the others, based on those estimates. The algorithm is applied on a sliding temporal window to cope with large changes in reflectance. We also propose morphological processing to suppress illumination artefacts. RESULTS: The corrected illumination function is homogeneous in the series, enabling the direct comparison of grey-levels intensities in each pixel, and so the detection of the GA growth between any two images. To demonstrate that, we present numerous experiments performed on a dataset of 18 series (328 images), manually segmented by an ophthalmologist. First, we show that the normalization preprocessing dramatically increases the contrast of the GA growth areas. Secondly, we apply segmentation algorithms derived from Otsu's thresholding to detect automatically the GA total growth and the GA progress between consecutive images. We demonstrate qualitatively and quantitatively that these algorithms, although fully automatic, unsupervised and basic, already lead to interesting segmentation results when applied to the normalized images. Colored maps representing the GA evolution can be derived from the segmentations. CONCLUSION: To our knowledge, the proposed method is the first one which corrects automatically and jointly the illumination inhomogeneity in a series of fundus images, regardless of the number of images, the size, shape and progression of lesion areas. This algorithm greatly facilitates the visual interpretation by the medical expert. It opens up the possibility of treating automatically each series as a whole (not just in pairs of images) to model the GA growth.

Quantification of Key Retinal Features in Early and Late Age-Related Macular Degeneration Using Deep Learning.

PURPOSE: We sought to develop and validate a deep learning model for segmentation of 13 features associated with neovascular and atrophic age-related macular degeneration (AMD). DESIGN: Development and validation of a deep-learning model for feature segmentation. METHODS: Data for model development were obtained from 307 optical coherence tomography volumes. Eight experienced graders manually delineated all abnormalities in 2712 B-scans. A deep neural network was trained with these data to perform voxel-level segmentation of the 13 most common abnormalities (features). For evaluation, 112 B-scans from 112 patients with a diagnosis of neovascular AMD were annotated by 4 independent observers. The main outcome measures were Dice score, intraclass correlation coefficient, and free-response receiver operating characteristic curve. RESULTS: On 11 of 13 features, the model obtained a mean Dice score of 0.63 ± 0.15, compared with 0.61 ± 0.17 for the observers. The mean intraclass correlation coefficient for the model was 0.66 ± 0.22, compared with 0.62 ± 0.21 for the observers. Two features were not evaluated quantitatively because of a lack of data. Free-response receiver operating characteristic analysis demonstrated that the model scored similar or higher sensitivity per false positives compared with the observers. CONCLUSIONS: The quality of the automatic segmentation matches that of experienced graders for most features, exceeding human performance for some features. The quantified parameters provided by the model can be used in the current clinical routine and open possibilities for further research into treatment response outside clinical trials.

[Are There Different Phenotypes in Geographic Atrophy of AMD? - Pilot Study on Differentiation Using Multimodal Imaging]. / Gibt es unterschiedliche Phänotypen bei der geografischen Atrophie der AMD? ­ Pilotstudie zur Differenzierung mittels multimodaler Bildgebung.

BACKGROUND: Geographic atrophy (GA) in patients with age-related macular degeneration (AMD) involves a loss of photoreceptors (PR), retinal pigment epithelium (RPE) and choriocapillaris (CC). For treatment decisions, it is crucial to discern which of these layers the damage originates, subsequently spreading to the others. It has long been thought that the RPE, with its accumulation of lipofuscin, is the site of primary damage in the development of GA. However, histological studies have shown that in some patients, the PR are affected first, followed by secondary damage to the RPE and CC, and in others regression of the CC is the first manifestation. The aim of this study was to use multimodal imaging to determine the extent of the damage at the levels of the PR, RPE and CC, to characterise the individual phenotypic variations of GA and to investigate the corresponding functional impairment. PATIENTS AND METHODS: Twenty eyes of 20 patients (mean age 78 years; 14 female, 6 male) with the clinical diagnosis of GA were examined by means of fundus autofluorescence (FAF) to evaluate the damage to the RPE, en face SD-OCT at the level of the PR to characterise the area of cell loss in this layer and OCT angiography (OCT-A, AngioVue, Optovue; 50 µm CC-segmentation with localization below the RPE) to assess regression of the CC. The affected area of each layer was measured. Best-corrected visual acuity (BCVA) test and fundus correlated automated 10° microperimetry (MAIA Microperimetry, CENTERVUE; 4-2 strategy, 68 stimuli) were performed in all patients. The results of these examinations were evaluated and correlated. RESULTS: All eyes showed a different extent of the areas of atrophy in the PR, RPE and CC. The layer with the largest area of atrophy was the RPE in 13 eyes (65%), the PR in 3 eyes (15%) and the CC in 4 eyes (20%). While the visual loss depended entirely on the presence of foveal sparing, microperimetry revealed a correlation between the extent of detectable functional deficit and the largest atrophic area. CONCLUSIONS: Multimodal imaging with FAF, en face OCT, OCT-A and a correlation with microperimetry enables a clinical phenotypic differentiation in GA as well as a more precise characterisation of the associated functional impairment. This confirms clinically the histologically demonstrated diversity of the damaged structure (PR, RPE or CC) in patients with GA. However, the variations identified in this pilot study must be confirmed in Reading Center-based larger cohorts. The approach described here may lead to differentiated consideration of the anatomical and functional aspects of the disease and turn out to be helpful in patient selection as well as in identifying and monitoring future therapeutic approaches.

Multimodal Imaging for the Assessment of Geographic Atrophy in Patients with "Foveal" and "No-Foveal" Sparing.

INTRODUCTION: The aim of the study was to evaluate the applicability of optical coherence tomography (OCT) angiography (OCTA) for measuring geographic atrophy (GA) areas in age-related macular degeneration (AMD) patients with "foveal" and "no-foveal" sparing disease and compare it to other imaging modalities. METHODS: A multimodal imaging protocol was applied, using infrared (IR) imaging, fundus autofluorescence (FAF), OCTA, and en-face OCT in 35 eyes of 23 AMD patients with GA. Patients were classified into 2 groups, with and without foveal sparing disease. GA area measurements for all imaging modalities were compared for each group separately. RESULTS: The measured GA area was estimated to be 6.68 ± 3.18 mm2 using IR; 6.99 ± 3.09 mm2 using FAF; 6.56 ± 3.11 mm2 using OCTA, and 6.65 ± 3.14 mm2 using en-face OCT. There was no statistically significant difference in the GA area between different modalities (p = 0.977). When separate analysis was conducted for patients with "foveal" and "no-foveal" sparing disease, although GA measurements in FAF imaging displayed higher numerical values than the other modalities, especially in patients with foveal sparing, no statistically significant difference in the GA area was found between the different imaging modalities in either group (p = 0.816 for foveal sparing; p = 0.992 for no-foveal sparing group). CONCLUSIONS: OCTA can be reliably used in the assessment of GA in AMD patients with and without foveal sparing disease. For both groups, measurements are comparable to IR, en-face OCT, and FAF, despite the fact that the latter recorded larger area of GA, mainly in the foveal sparing cases.

Correlations Between Choriocapillaris and Choroidal Measurements and the Growth of Geographic Atrophy Using Swept Source OCT Imaging.

PURPOSE: Correlations among enlargement rates (ERs) of geographic atrophy (GA) and choriocapillaris (CC) flow deficits (FDs), mean choroidal thickness (MCT), and choroidal vascularity index (CVI) were investigated using swept source-optical coherence tomography (SS-OCT) in age-related macular degeneration (AMD). DESIGN: A retrospective review of prospective, observational case series. METHODS: Eyes with GA from AMD were imaged with SS-OCT using 6 × 6-mm scan pattern. GA lesions were identified and measured using customized en face structural images, and annual square root ERs of GA were calculated. At baseline, choriocapillaris FDs from different regions outside the GA were measured, and MCT and CVI from the entire scan area were measured. All measurements were performed using previously published and validated algorithms. RESULTS: A total of 38 eyes from 27 patients were included. The CC FDs within each region around GA lesions were highly correlated with ERs of GA (all P < .005). CVI inside the GA region was correlated with the ERs (P = .03), whereas other choroidal measurements had no significant correlation with the ERs of GA (P > .06). CONCLUSIONS: Statistically significant correlations were found between the ERs of GA and CC percentage of FD (FD%) from the entire scan region outside the GA and not just the region immediately adjacent to the GA. These results suggest that abnormal CC perfusion throughout the macula contributes to disease progression in eyes with GA. CVI inside the GA region could also be a potential indicator for the growth of GA.