Developing quantitative analysis program of blood flow velocity according to vessel diameter for neovascular age-related macular degeneration using OCTA-VISTA.

This study aimed to develop a quantitative analysis program of blood flow velocity by vessel diameter in neovascular age-related macular degeneration (nAMD) subjects using high-speed swept-source optical coherence tomography angiography. This retrospective, observational, cross-sectional study included 10 eyes of healthy volunteers and 4 eyes of patients with representative nAMD. Novel scan patterns and variable interscan time analysis were utilized to measure the flow parameter, a surrogate marker of blood flow velocity, by vessel diameter within different depths. Detected vessels at superficial and deep as well as outer retinal regions were categorized into three vessel diameters (major vessels (> 40 µm), medium vessels (20-40 µm), and capillaries (< 20 µm)). The flow parameter increased with enlarged vessel diameter in all participants at superficial and deep layer. All nAMD subjects, except for type 3 macular neovascularization (MNV), contained a structure dominated by medium vessels at outer retinal region. The mean flow parameter at outer retinal region was type 1 MNV (1.46 ms-1), type 1 + 2 MNV (0.98 ms-1), and polypoidal choroidal vasculopathy, including branching vascular networks (1.46 ms-1). This program provides the possibility to extract the blood flow information at different depths by vessel diameter types, which is considered to be useful tool for evaluating nAMD pathology and activity.

Calcified Drusen Prevent the Detection of Underlying Choriocapillaris Using Swept-Source Optical Coherence Tomography Angiography.

Purpose: In age-related macular degeneration (AMD), choriocapillaris flow deficits (CCFDs) under soft drusen can be measured using established compensation strategies. This study investigated whether CCFDs can be quantified under calcified drusen (CaD). Methods: CCFDs were measured in normal eyes (n = 30) and AMD eyes with soft drusen (n = 30) or CaD (n = 30). CCFD density masks were generated to highlight regions with higher CCFDs. Masks were also generated for soft drusen and CaD based on both structural en face OCT images and corresponding B-scans. Dice similarity coefficients were calculated between the CCFD density masks and both the soft drusen and CaD masks. A phantom experiment was conducted to simulate the impact of light scattering that arises from CaD. Results: Area measurements of CCFDs were highly correlated with those of CaD but not soft drusen, suggesting an association between CaD and underlying CCFDs. However, unlike soft drusen, the detected optical coherence tomography (OCT) signals underlying CaD did not arise from the defined CC layer but were artifacts caused by the multiple scattering property of CaD. Phantom experiments showed that the presence of highly scattering material similar to the contents of CaD caused an artifactual scattering tail that falsely generated a signal in the CC structural layer but the underlying flow could not be detected. Similarly, CaD also caused an artifactual scattering tail and prevented the penetration of light into the choroid, resulting in en face hypotransmission defects and an inability to detect blood flow within the choriocapillaris. Upon resolution of the CaD, the CC perfusion became detectable. Conclusions: The high scattering property of CaD leads to a scattering tail under these drusen that gives the illusion of a quantifiable optical coherence tomography angiography signal, but this signal does not contain the angiographic information required to assess CCFDs. For this reason, CCFDs cannot be reliably measured under CaD, and CaD must be identified and excluded from macular CCFD measurements.

Study protocol: optical coherence tomography angiography for the detection of neovascular age-related macular degeneration: a comprehensive multicentre diagnostic accuracy study in the UK-the ATHENA study.

INTRODUCTION: The diagnosis of neovascular age-related macular degeneration (nAMD), the leading cause of visual impairment in the developed world, relies on the interpretation of various imaging tests of the retina. These include invasive angiographic methods, such as Fundus Fluorescein Angiography (FFA) and, on occasion, Indocyanine-Green Angiography (ICGA). Newer, non-invasive imaging modalities, predominately Optical Coherence Tomography (OCT) and Optical Coherence Tomography Angiography (OCTA), have drastically transformed the diagnostic approach to nAMD. The aim of this study is to undertake a comprehensive diagnostic accuracy assessment of the various imaging modalities used in clinical practice for the diagnosis of nAMD (OCT, OCTA, FFA and, when a variant of nAMD called Polypoidal Choroidal Vasculopathy is suspected, ICGA) both alone and in various combinations. METHODS AND ANALYSIS: This is a non-inferiority, prospective, randomised diagnostic accuracy study of 1067 participants. Participants are patients with clinical features consistent with nAMD who present to a National Health Service secondary care ophthalmology unit in the UK. Patients will undergo OCT as per standard practice and those with suspicious features of nAMD on OCT will be approached for participation in the study. Patients who agree to take part will also undergo both OCTA and FFA (and ICGA if indicated). Interpretation of the imaging tests will be undertaken by clinicians at recruitment sites. A randomised design was selected to avoid bias from consecutive review of all imaging tests by the same clinician. The primary outcome of the study will be the difference in sensitivity and specificity between OCT+OCTA and OCT+FFA (±ICGA) for nAMD detection as interpreted by clinicians at recruitment sites. ETHICS AND DISSEMINATION: The study has been approved by the South Central-Oxford B Research Ethics Committee with reference number 21/SC/0412.Dissemination of study results will involve peer-review publications, presentations at major national and international scientific conferences. TRIAL REGISTRATION NUMBER: ISRCTN18313457.

Prediction of neovascular age-related macular degeneration recurrence using optical coherence tomography images with a deep neural network.

Neovascular age-related macular degeneration (nAMD) can result in blindness if left untreated, and patients often require repeated anti-vascular endothelial growth factor injections. Although, the treat-and-extend method is becoming popular to reduce vision loss attributed to recurrence, it may pose a risk of overtreatment. This study aimed to develop a deep learning model based on DenseNet201 to predict nAMD recurrence within 3 months after confirming dry-up 1 month following three loading injections in treatment-naïve patients. A dataset of 1076 spectral domain optical coherence tomography (OCT) images from 269 patients diagnosed with nAMD was used. The performance of the model was compared with that of 6 ophthalmologists, using 100 randomly selected samples. The DenseNet201-based model achieved 53.0% accuracy in predicting nAMD recurrence using a single pre-injection image and 60.2% accuracy after viewing all the images immediately after the 1st, 2nd, and 3rd injections. The model outperformed experienced ophthalmologists, with an average accuracy of 52.17% using a single pre-injection image and 53.3% after examining four images before and after three loading injections. In conclusion, the artificial intelligence model demonstrated a promising ability to predict nAMD recurrence using OCT images and outperformed experienced ophthalmologists. These findings suggest that deep learning models can assist in nAMD recurrence prediction, thus improving patient outcomes and optimizing treatment strategies.

Morphological changes of macular neovascularization during long-term anti-VEGF-therapy in neovascular age-related macular degeneration.

PURPOSE: To analyze the morphological changes of macular neovascularization (MNV) in exudative neovascular age-related macular degeneration under long-term intravitreal anti-vascular endothelial growth factor (VEGF) therapy in a retrospective cohort study. METHODS AND PATIENTS: We evaluated 143 nAMD eyes of 94 patients (31 male, 63 female; initial age 55-97 y, mean age 75.9 ± 7.5 y), who started anti-VEGF therapy (IVAN pro re nata (PRN) protocol) between 2009-2018 and received ongoing therapy until the last recorded visit (mean follow-up 5.3 ± 2.9 y, range 1-14 y). The mean total number of injections was 33.3 ± 19.8 with 7.0 ± 2.3 injections/year. MNV size and, if present, associated complete retinal pigment epithelium (RPE) and outer retina atrophy (cRORA) size were measured on optical coherence tomography (OCT) volume scans at the initial visit and for each year of follow-up. MNV and cRORA were identified on B-scans and their respective borders were manually transposed onto the en-face near infrared image and measured in mm2. RESULTS: MNV enlarged through follow-up, with a mean growth rate of 1.24 mm2 / year. The mean growth in MNV size was independent of initial MNV size, age, gender, MNV subtypes or number of injections per year. Nevertheless, a great interindividual variation in size and growth was observed. cRORA developed in association with increasing MNV size and its incidence increased linearly over follow-up. cRORA lesions also showed continuous growth by a rate of 1.22 mm2 / year. CONCLUSIONS: Despite frequent long-term anti-VEGF therapy, we observed ongoing MNV growth. This is consistent with the concept that the development of MNV may be a physiological biological repair mechanism to preserve RPE and photoreceptor function, provided hyperpermeability and fluid exudation are controlled. Whether recurring low VEGF levels or other factors are responsible for MNV growth remains controversial.

Predicting OCT biological marker localization from weak annotations.

Recent developments in deep learning have shown success in accurately predicting the location of biological markers in Optical Coherence Tomography (OCT) volumes of patients with Age-Related Macular Degeneration (AMD) and Diabetic Retinopathy (DR). We propose a method that automatically locates biological markers to the Early Treatment Diabetic Retinopathy Study (ETDRS) rings, only requiring B-scan-level presence annotations. We trained a neural network using 22,723 OCT B-Scans of 460 eyes (433 patients) with AMD and DR, annotated with slice-level labels for Intraretinal Fluid (IRF) and Subretinal Fluid (SRF). The neural network outputs were mapped into the corresponding ETDRS rings. We incorporated the class annotations and domain knowledge into a loss function to constrain the output with biologically plausible solutions. The method was tested on a set of OCT volumes with 322 eyes (189 patients) with Diabetic Macular Edema, with slice-level SRF and IRF presence annotations for the ETDRS rings. Our method accurately predicted the presence of IRF and SRF in each ETDRS ring, outperforming previous baselines even in the most challenging scenarios. Our model was also successfully applied to en-face marker segmentation and showed consistency within C-scans, despite not incorporating volume information in the training process. We achieved a correlation coefficient of 0.946 for the prediction of the IRF area.

Pachychoroid neovasculopathy versus macular neovascularization in age-related macular degeneration with and without shallow irregular pigment epithelial detachment.

To compare the choroidal neovascular features of individuals with pachychoroid neovasculopathy (PNV) and neovascular age-related macular degeneration (nAMD) with and without shallow irregular pigment epithelial detachment (SIPED). Using optical coherence tomography angiography, the choroidal neovascular complexes of 27 patients with PNV, 34 patients with nAMD and SIPED, and 15 patients with nAMD without SIPED were analyzed with FIJI and AngioTool software. PNV compared to nAMD with SIPED had a greater vessel percentage area (P = 0.034), junction density (P = 0.045), average vessel length (P < 0.001), and fractal dimension (P < 0.001). PNV, compared to nAMD without SIPED, had a greater total vessel length (P = 0.002), total number of junctions (P < 0.001), junction density (P = 0.034), and fractal dimension (P = 0.005). nAMD with SIPED, compared to nAMD without SIPED, had greater vessel area, total number of junctions, total vessel length, and average vessel length (all P values < 0.001). Patients with nAMD plus SIPED and individuals with nAMD without SIPED have similar fractal dimension values (P = 0.703). Biomarkers of choroidal neovascular complexity, such as fractal dimension, can be used to differentiate PNV from nAMD with or without SIPED.

Wavelet scattering transform application in classification of retinal abnormalities using OCT images.

To assist ophthalmologists in diagnosing retinal abnormalities, Computer Aided Diagnosis has played a significant role. In this paper, a particular Convolutional Neural Network based on Wavelet Scattering Transform (WST) is used to detect one to four retinal abnormalities from Optical Coherence Tomography (OCT) images. Predefined wavelet filters in this network decrease the computation complexity and processing time compared to deep learning methods. We use two layers of the WST network to obtain a direct and efficient model. WST generates a sparse representation of the images which is translation-invariant and stable concerning local deformations. Next, a Principal Component Analysis classifies the extracted features. We evaluate the model using four publicly available datasets to have a comprehensive comparison with the literature. The accuracies of classifying the OCT images of the OCTID dataset into two and five classes were [Formula: see text] and [Formula: see text], respectively. We achieved an accuracy of [Formula: see text] in detecting Diabetic Macular Edema from Normal ones using the TOPCON device-based dataset. Heidelberg and Duke datasets contain DME, Age-related Macular Degeneration, and Normal classes, in which we achieved accuracy of [Formula: see text] and [Formula: see text], respectively. A comparison of our results with the state-of-the-art models shows that our model outperforms these models for some assessments or achieves nearly the best results reported so far while having a much smaller computational complexity.

Radiomics-Based Prediction of Anti-VEGF Treatment Response in Neovascular Age-Related Macular Degeneration With Pigment Epithelial Detachment.

Purpose: Machine learning models based on radiomic feature extraction from clinical imaging data provide effective and interpretable means for clinical decision making. This pilot study evaluated whether radiomics features in baseline optical coherence tomography (OCT) images of eyes with pigment epithelial detachment (PED) associated with neovascular age-related macular degeneration (nAMD) can predict treatment response to as-needed anti-vascular endothelial growth factor (VEGF) therapy. Methods: Thirty-nine eyes of patients with PED undergoing anti-VEGF therapy were included. All eyes underwent a loading dose followed by as-needed therapy. OCT images at baseline, month 3, and month 6 were analyzed. Images were manually separated into non-responding, recurring, and responding eyes based on the presence or absence of subretinal fluid at month 6. PED radiomics features were then extracted from each image and images were classified as responding or recurring using a machine learning classifier applied to the radiomics features. Results: Linear discriminant analysis classification of baseline features as responsive versus recurring resulted in classification performance of 64.0% (95% confidence interval [CI] = 0.63-0.65), area under the curve (AUC = 0.78, 95% CI = 0.72-0.82), sensitivity 0.79 (95% CI = 0.63-0.87), and specificity 0.58 (95% CI = 0.50-0.67). Further analysis of features in recurring eyes identified a significant shift toward non-responding mean feature values over 6 months. Conclusions: Our results demonstrate the use of radiomics features as predictors for treatment response to as-needed anti-VEGF therapy. Our study demonstrates the potential for radiomics feature in clinical decision support for personalizing anti-VEGF therapy. Translational Relevance: The ability to use PED texture features to predict treatment response facilitates personalized clinical decision making.

Accurate drusen segmentation in optical coherence tomography via order-constrained regression of retinal layer heights.

Drusen are an important biomarker for age-related macular degeneration (AMD). Their accurate segmentation based on optical coherence tomography (OCT) is therefore relevant to the detection, staging, and treatment of disease. Since manual OCT segmentation is resource-consuming and has low reproducibility, automatic techniques are required. In this work, we introduce a novel deep learning based architecture that directly predicts the position of layers in OCT and guarantees their correct order, achieving state-of-the-art results for retinal layer segmentation. In particular, the average absolute distance between our model's prediction and the ground truth layer segmentation in an AMD dataset is 0.63, 0.85, and 0.44 pixel for Bruch's membrane (BM), retinal pigment epithelium (RPE) and ellipsoid zone (EZ), respectively. Based on layer positions, we further quantify drusen load with excellent accuracy, achieving 0.994 and 0.988 Pearson correlation between drusen volumes estimated by our method and two human readers, and increasing the Dice score to 0.71 ± 0.16 (from 0.60 ± 0.23) and 0.62 ± 0.23 (from 0.53 ± 0.25), respectively, compared to a previous state-of-the-art method. Given its reproducible, accurate, and scalable results, our method can be used for the large-scale analysis of OCT data.

Evaluation of non-exudative tomographic signs in cases of exudative age-related macular degeneration.

OBJECTIVE: To analyse the prevalence of non-exudative tomographic signs (onion sign, pseudoswelling, external retinal tubulation, pseudocysts, subretinal clefts and macular atrophy) in patients with neovascular age-related macular degeneration. MATERIAL AND METHODS: A total of 174 eyes of patients with neovascular age-related macular degeneration who had not received previous treatment were included in the study. Visual acuity, neovascularization activity, and the appearance or not of the different signs under study were assessed at times 0 (initial visit), 4 months, one year, year and a half, and at 2 and 3 years of follow-up. The following were also evaluated: age, sex, affected eye and type of neovascularization (1, 2, 3, polypoid or mixed). The analysis were performed using the statistical software R (version 3.3.2) and the glmmADMB package (version 0.8.3.3). RESULTS: The presence of pseudocysts and external retinal tubulation increases throughout the follow-up. The onion sign begins with an ascending frequency up to 12 months, then decreases at 18 months and increases again at 24 months. Regarding pseudowelling, it maintains an increase until 18 months to finally decrease. Subretinal clefts is the rarest sign, presenting in 1.1% on the first visit. Finally, macular atrophy, present in 12.6% of the eyes initially, is found in 25% after 2 years. CONCLUSION: Pseudocysts, external retinal tubulation and macular atrophy were the most prevalent signs, while subretinal clefts were the most infrequent.

Correlation between contrast sensitivity and morphological features obtained by OCT in patients with age-related macular degeneration treated with a loading dose of vascular endothelial growth factor inhibitors.

OBJECTIVE: To determine the correlation between contrast sensitivity and morphological characteristics obtained by Optical Coherence Tomography (OCT) in patients with Age-Related Macular Degeneration treated with a loading dose of vascular endothelial growth factor inhibitors (anti-VEGF). DESIGN: This is an ambispective (prospective + retrospective) observational, cross-sectional, and analytical study. PARTICIPANTS: All patients over 55 years of age with Age-Related Macular Degeneration who attended the Retina service of the Ophthalmology department and met the inclusion criteria between March-May 2022. METHODS: Data collection was carried out by reviewing the records of patients diagnosed with Age-Related Macular Degeneration of the neovascular variety treated with the loading dose of anti-VEGF. OCT studies obtained by Optovue® iVue80 prior to the application of intravitreal injections of patients who met the inclusion criteria and were currently in the first month after the last dose of anti-VEGF were analyzed. A total of 33 subjects were included, of which 30 continued follow-ups. The subjects underwent a new ophthalmological evaluation and new retinal measurements of the affected eye. Normality tests (Shapiro‒Wilk) were performed where a nonparametric data distribution was demonstrated. RESULTS: A linear regression analysis was performed comparing the logarithmic values of both visual acuity and contrast sensitivity, obtaining a significant relationship between both values after the application of treatment (P = <.0001***). Likewise, correlation was demonstrated between the decrease in contrast sensitivity values and all the characteristics evaluated in the patients' OCT. CONCLUSIONS: Antiangiogenesis strategies can lead to better results in global visual function, positively impacting contrast sensitivity.

Pigment epithelial detachment composition indices (PEDCI) in neovascular age-related macular degeneration.

We provide an automated analysis of the pigment epithelial detachments (PEDs) in neovascular age-related macular degeneration (nAMD) and estimate areas of serous, neovascular, and fibrous tissues within PEDs. A retrospective analysis of high-definition spectral-domain OCT B-scans from 43 eyes of 37 patients with nAMD with presence of fibrovascular PED was done. PEDs were manually segmented and then filtered using 2D kernels to classify pixels within the PED as serous, neovascular, or fibrous. A set of PED composition indices were calculated on a per-image basis using relative PED area of serous (PEDCI-S), neovascular (PEDCI-N), and fibrous (PEDCI-F) tissue. Accuracy of segmentation and classification within the PED were graded in masked fashion. Mean overall intra-observer repeatability and inter-observer reproducibility were 0.86 ± 0.07 and 0.86 ± 0.03 respectively using intraclass correlations. The mean graded scores were 96.99 ± 8.18, 92.12 ± 7.97, 91.48 ± 8.93, and 92.29 ± 8.97 for segmentation, serous, neovascular, and fibrous respectively. Mean (range) PEDCI-S, PEDCI-N, and PEDCI-F were 0.253 (0-0.952), 0.554 (0-1), and 0.193 (0-0.693). A kernel-based image processing approach demonstrates potential for approximating PED composition. Evaluating follow up changes during nAMD treatment with respect to PEDCI would be useful for further clinical applications.

Six-year incidence of age-related macular degeneration and correlation to OCT-derived drusen volume measurements in a Chinese population.

AIMS: To report the 6-year incidence of optical coherence tomography (OCT)-derived age-related changes in drusen volume and related systemic and ocular associations. METHODS: Chinese adults aged 40 years and older were assessed at baseline and 6 years with colour fundus photography (CFP) and spectral domain (SD) OCT. CFPs were graded for age-related macular degeneration (AMD) features and drusen volume was generated using commercially available automated software. RESULTS: A total of 4172 eyes of 2580 participants (mean age 58.12±9.03 years; 51.12% women) had baseline and 6-year follow-up CFP for grading, of these, 2130 eyes of 1305 participants had gradable SD-OCT images, available for analysis. Based on CFP grading, 136 (3.39%) participants developed incident early AMD and 10 (0.25%) late AMD. Concurrently, retinal pigment epithelial-Bruch's membrane (RPE-BrC) volumes decreased, remained stable and increased in 6.8%, 78.5% and 14.7%, respectively, over 6 years. In eyes where RPE-BrC volumes were >0 mm3 at baseline, this was associated with two-fold higher prevalence rate of any AMD at baseline (p<0.001). Multivariable analysis showed that when compared with eyes where RPE-BrC volume was unchanged, volume decrease was significantly associated with older age (OR=1.30; p<0.001), smoking (OR=2.21; p=0.001) and chronic kidney disease (OR=3.4, p=0.008), while increase was associated with older age (OR=1.36; p<0.001) and hypertension (OR=1.43; p=0.016). CONCLUSION: AMD incidence detected at 6 years on CFP and correlated OCT-derived drusen volume measurement change is low. Older age and some systemic risk factors are associated with drusen volume change, and our data provide new insights into relationship between systemic risk factors and outer retinal morphology in Asian eyes.

[Prognosis criteria of optical coherence tomography angiography for the long-term efficacy of anti-VEGF therapy of neovascular age-related macular degeneration]. / Prognosticheskie kriterii opticheskoi kogerentnoi tomografii-angiografii v otsenke dolgosrochnoi effektivnosti anti-VEGF-terapii neovaskulyarnoi vozrastnoi makulyarnoi degeneratsii setchatki.

PURPOSE: The study aimed to determine the most significant optical coherence tomography angiography (OCTA) parameters in terms of predicting anti-VEGF therapy effectiveness during long-term (3-year) follow-up of patients with neovascular age-related macular degeneration (nAMD). MATERIAL AND METHODS: The study included 122 patients (122 eyes) with mean age of 73.4±6.6 years who were diagnosed with nAMD. Subgroup analysis included 50 patients (50 eyes) with detailed OCT angiography examination of macular neovascularization (MNV) characteristics and their changes in the course of the follow-up, which lasted 144 weeks. All patients were treated by angiogenesis inhibitor (aflibercept 2 mg), most of them - according to Treat-and-Extend protocol. RESULTS: Treatment response (either 'good' or 'partial') was achieved in all patients, and the proportion of the response types was similar in both types 1 and 2 MNV. Key OCTA parameters associated with the number of injections, as well as morphological and functional response (best-corrected visual acuity, retinal neuroepithelium and pigment epithelium detachment), were vascular network area and MNV area assessed at baseline and three months after treatment initiation. Both of these parameters were closely related in both MNV types during the follow-up. Key parameter with maximum number of clinically significant correlations ('very high' strength, p<0.05) in eyes with 'good' response was MNV area, in eyes with 'partial' response - vascular density and greatest vascular caliber. CONCLUSIONS: Vascular network area and MNV area assessed at baseline and after three loading doses were determined as the most significant OCTA characteristics for predicting the number of injections and treatment response based on functional and morphological parameters. MNV area was found to be the most clinically significant marker in 'good' response, vascular density and greatest vascular caliber - in 'partial' response.

Morphology of en face Haller vessel and macular neovascularization at baseline and 3 months as predictive factors in age-related macular degeneration.

The clinical implication of en face imaging of Haller vessels and macular neovascularization (MNV) in neovascular age-related macular degeneration (nAMD) is not well established. The purpose of this study is to investigate whether the early-phase morphology of en face Haller vessel and MNV is related to the injection frequency and visual outcome in treatment-naïve nAMD. En face images of Haller vessel and MNV were acquired from 52 eyes at baseline, after three loading doses and at 12 months later using optical coherence tomography (OCT) and OCT angiography. Vessel area, diameter, length, intersection number, fractal dimension, and lacunarity were calculated. Patients were classified according to the injection frequency (< 5 as the infrequent group) and visual gain (≥ 0.3 logMAR) over 12 months. The infrequent group was associated with a longer Haller vessel length after loading doses (OR 3.05, P = 0.01), while visual gain was associated with a smaller maximal MNV diameter after loading doses (OR 0.22, P = 0.03). A predictive model for frequent injection based on the Haller vessel length demonstrated an AUC of 0.71. In conclusion, the en face Haller vessel and MNV morphology after loading doses can be used as biomarkers for the injection frequency and visual gain during the first year in treatment-naïve nAMD patients.

Prediction of treatment outcome in neovascular age-related macular degeneration using a novel convolutional neural network.

While prognosis and risk of progression are crucial in developing precise therapeutic strategy in neovascular age-related macular degeneration (nAMD), limited predictive tools are available. We proposed a novel deep convolutional neural network that enables feature extraction through image and non-image data integration to seize imperative information and achieve highly accurate outcome prediction. The Heterogeneous Data Fusion Net (HDF-Net) was designed to predict visual acuity (VA) outcome (improvement ≥ 2 line or not) at 12th months after anti-VEGF treatment. A set of pre-treatment optical coherence tomography (OCT) image and non-image demographic features were employed as input data and the corresponding 12th-month post-treatment VA as the target data to train, validate, and test the HDF-Net. This newly designed HDF-Net demonstrated an AUC of 0.989 (95% CI 0.970-0.999), accuracy of 0.936 [95% confidence interval (CI) 0.889-0.964], sensitivity of 0.933 (95% CI 0.841-0.974), and specificity of 0.938 (95% CI 0.877-0.969). By simulating the clinical decision process with mixed pre-treatment information from raw OCT images and numeric data, HDF-Net demonstrated promising performance in predicting individualized treatment outcome. The results highlight the potential of deep learning to simultaneously process a broad range of clinical data to weigh and leverage the complete information of the patient. This novel approach is an important step toward real-world personalized therapeutic strategy for typical nAMD.

Visualizing lipid behind the retina in aging and age-related macular degeneration, via indocyanine green angiography (ASHS-LIA).

Age-related macular degeneration (AMD) causes legal blindness in older adults worldwide. Soft drusen are the most extensively documented intraocular risk factor for progression to advanced AMD. A long-standing paradox in AMD pathophysiology has been the vulnerability of Asian populations to polypoidal choroidal vasculopathy (PCV) in the presence of relatively few drusen. Age-related scattered hypofluorescent spots on late phase indocyanine green angiography (ASHS-LIA) was recently proposed as precursors of PCV. Herein, we offer a resolution to the paradox by reviewing evidence that ASHS-LIA indicates the diffuse form of lipoprotein-related lipids accumulating in Bruch's membrane (BrM) throughout adulthood. Deposition of these lipids leads to soft drusen and basal linear deposit (BLinD), a thin layer of soft drusen material in AMD; Pre-BLinD is the precursor. This evidence includes: 1. Both ASHS-LIA and pre-BLinD/BLinD accumulate in older adults and start under the macula; 2. ASHS-LIA shares hypofluorescence with soft drusen, known to be physically continuous with pre-BLinD/BLinD. 3. Model system studies illuminated a mechanism for indocyanine green uptake by retinal pigment epithelium. 4. Neither ASHS-LIA nor pre-BLinD/ BLinD are visible by multimodal imaging anchored on current optical coherence tomography, as confirmed with direct clinicopathologic correlation. To contextualize ASHS-LIA, we also summarize angiographic characteristics of different drusen subtypes in AMD. As possible precursors for PCV, lipid accumulation in forms beyond soft drusen may contribute to the pathogenesis of this prevalent disease in Asia. ASHS-LIA also might help identify patients at risk for progression, of value to clinical trials for therapies targeting early or intermediate AMD.

Classifying neovascular age-related macular degeneration with a deep convolutional neural network based on optical coherence tomography images.

Neovascular age-related macular degeneration (nAMD) is among the main causes of visual impairment worldwide. We built a deep learning model to distinguish the subtypes of nAMD using spectral domain optical coherence tomography (SD-OCT) images. Data from SD-OCT images of nAMD (polypoidal choroidal vasculopathy, retinal angiomatous proliferation, and typical nAMD) and normal healthy patients were analyzed using a convolutional neural network (CNN). The model was trained and validated based on 4749 SD-OCT images from 347 patients and 50 healthy controls. To adopt an accurate and robust image classification architecture, we evaluated three well-known CNN structures (VGG-16, VGG-19, and ResNet) and two customized classification layers (fully connected layer with dropout vs. global average pooling). Following the test set performance, the model with the highest classification accuracy was used. Transfer learning and data augmentation were applied to improve the robustness and accuracy of the model. Our proposed model showed an accuracy of 87.4% on the test data (920 images), scoring higher than ten ophthalmologists, for the same data. Additionally, the part that our model judged to be important in classification was confirmed through Grad-CAM images, and consequently, it has a similar judgment criteria to that of ophthalmologists. Thus, we believe that our model can be used as an auxiliary tool in clinical practice.

From Data to Deployment: The Collaborative Community on Ophthalmic Imaging Roadmap for Artificial Intelligence in Age-Related Macular Degeneration.

OBJECTIVE: Health care systems worldwide are challenged to provide adequate care for the 200 million individuals with age-related macular degeneration (AMD). Artificial intelligence (AI) has the potential to make a significant, positive impact on the diagnosis and management of patients with AMD; however, the development of effective AI devices for clinical care faces numerous considerations and challenges, a fact evidenced by a current absence of Food and Drug Administration (FDA)-approved AI devices for AMD. PURPOSE: To delineate the state of AI for AMD, including current data, standards, achievements, and challenges. METHODS: Members of the Collaborative Community on Ophthalmic Imaging Working Group for AI in AMD attended an inaugural meeting on September 7, 2020, to discuss the topic. Subsequently, they undertook a comprehensive review of the medical literature relevant to the topic. Members engaged in meetings and discussion through December 2021 to synthesize the information and arrive at a consensus. RESULTS: Existing infrastructure for robust AI development for AMD includes several large, labeled data sets of color fundus photography and OCT images; however, image data often do not contain the metadata necessary for the development of reliable, valid, and generalizable models. Data sharing for AMD model development is made difficult by restrictions on data privacy and security, although potential solutions are under investigation. Computing resources may be adequate for current applications, but knowledge of machine learning development may be scarce in many clinical ophthalmology settings. Despite these challenges, researchers have produced promising AI models for AMD for screening, diagnosis, prediction, and monitoring. Future goals include defining benchmarks to facilitate regulatory authorization and subsequent clinical setting generalization. CONCLUSIONS: Delivering an FDA-authorized, AI-based device for clinical care in AMD involves numerous considerations, including the identification of an appropriate clinical application; acquisition and development of a large, high-quality data set; development of the AI architecture; training and validation of the model; and functional interactions between the model output and clinical end user. The research efforts undertaken to date represent starting points for the medical devices that eventually will benefit providers, health care systems, and patients.