Search

To investigate the functional associations of intraretinal fluid (IRF) and subretinal fluid (SRF) volumes at baseline and after the loading dose as well as fluid change after the first injection with best-corrected visual acuity (BCVA) in patients with neovascular age-related macular degeneration (nAMD) who received an anti-VEGF treatment over 24 months.Post hoc analysis of a phase III, randomized, multicenter trial in which ranibizumab was administered monthly or in a pro re nata regimen (HARBOR).Study eyes of 1094 treatment-naïve patients with nAMD.IRF and SRF volumes were segmented automatically on monthly spectral domain OCT images. Fluid volumes and changes thereof were included as covariates into longitudinal mixed-effects models, which modeled BCVA trajectories.BCVA estimates corresponding to baseline, follow-up, and persistent IRF/SRF volumes after the loading dose; BCVA estimates of change in fluid volumes after the first injection; and marginal and conditional RAnalysis of 22 494 volumetric scans revealed that foveal IRF consistently shows a negative correlation with BCVA at baseline and subsequent visits (-3.23 and -4.32 letters/100 nL, respectively). After the first injection, BCVA increased by +2.13 letters/100 nL decrease in foveal IRF. Persistent IRF was associated with lower baseline BCVA and less improvement. Foveal SRF correlated with better BCVA at baseline and subsequent visits (+6.52 and +1.42 letters/100 nL, respectively). After the first injection, SRF decrease was associated with significant vision gain (+5.88 letters/100 nL). Foveal fluid correlated more with BCVA than parafoveal IRF/SRF.Although IRF consistently correlates with decreased function and recovery throughout therapy, SRF is associated with a more pronounced functional improvement. Moreover, SRF resolution provides increased benefit. Fluid-function correlation represents an essential base for the development of personalized treatment regimens, optimizing functional outcomes, and reducing treatment burden.

To investigate the impact of qualitatively graded and deep learning quantified imaging biomarkers on growth of geographic atrophy (GA) secondary to age-related macular degeneration.This prospective study included 1062 visits of 181 eyes of 100 patients with GA. Spectral-domain optical coherence tomography (SD-OCT) and fundus autofluorescence (FAF) images were acquired at each visit. Hyperreflective foci (HRF) were quantitatively assessed in SD-OCT volumes using a validated deep learning algorithm. FAF images were graded for FAF patterns, subretinal drusenoid deposits (SDD), GA lesion configuration and atrophy enlargement. Linear mixed models were calculated to investigate associations between all parameters and GA progression.FAF patterns were significantly associated with GA progression (p 0.001). SDD was associated with faster GA growth (p = 0.005). Eyes with higher HRF concentrations showed a trend towards faster GA progression (p = 0.072) and revealed a significant impact on GA enlargement in interaction with FAF patterns (p = 0.01). The fellow eye status had no significant effect on lesion enlargement (p 0.05). The diffuse-trickling FAF pattern exhibited significantly higher HRF concentrations than any other pattern (p 0.001).Among a wide range of investigated biomarkers, SDD and FAF patterns, particularly in interaction with HRF, significantly impact GA progression. Fully automated quantification of retinal imaging biomarkers such as HRF is both reliable and merited as HRF are indicators of retinal pigment epithelium dysmorphia, a central pathogenetic mechanism in GA. Identifying disease markers using the combination of FAF and SD-OCT is of high prognostic value and facilitates individualized patient management in a clinical setting.

To quantitatively measure hyperreflective foci (HRF) during the progression of geographic atrophy (GA) secondary to age-related macular degeneration (AMD) using deep learning (DL) and investigate the association with local and global growth of GA.Eyes with GA were prospectively included. Spectral-domain optical coherence tomography (SDOCT) and fundus autofluorescence images were acquired every 6 months. A 500-μm-wide junctional zone adjacent to the GA border was delineated and HRF were quantified using a validated DL algorithm. HRF concentrations in progressing and nonprogressing areas, as well as correlations between HRF quantifications and global and local GA progression, were assessed.A total of 491 SDOCT volumes from 87 eyes of 54 patients were assessed with a median follow-up of 28 months. Two-thirds of HRF were localized within a millimeter adjacent to the GA border. HRF concentration was positively correlated with GA progression in unifocal and multifocal GA (all P.001) and de novo GA development (P = .037). Local progression speed correlated positively with local increase of HRF (P value range.001-.004). Global progression speed, however, did not correlate with HRF concentrations (P.05). Changes in HRF over time did not have an impact on the growth in GA (P.05).Advanced artificial intelligence (AI) methods in high-resolution retinal imaging allows to identify, localize, and quantify biomarkers such as HRF. Increased HRF concentrations in the junctional zone and future macular atrophy may represent progressive migration and loss of retinal pigment epithelium. AI-based biomarker monitoring may pave the way into the era of individualized risk assessment and objective decision-making processes. NOTE: Publication of this article is sponsored by the American Ophthalmological Society.

The currently used measures of retinal function are limited by being subjective, nonlocalized, or taxing for patients. To address these limitations, we sought to develop and evaluate a deep learning (DL) method to automatically predict the functional end point (retinal sensitivity) based on structural OCT images.Retrospective, cross-sectional study.In total, 714 volumes of 289 patients were used in this study.A DL algorithm was developed to automatically predict a comprehensive retinal sensitivity map from an OCT volume. Four hundred sixty-three spectral-domain OCT volumes from 174 patients and their corresponding microperimetry examinations (Nidek MP-1) were used for development and internal validation, with a total of 15 563 retinal sensitivity measurements. The patients presented with a healthy macula, early or intermediate age-related macular degeneration, choroidal neovascularization, or geographic atrophy. In addition, an external validation was performed using 251 volumes of 115 patients, comprising 3 different patient populations: those with diabetic macular edema, retinal vein occlusion, or epiretinal membrane.We evaluated the performance of the algorithm using the mean absolute error (MAE), limits of agreement (LoA), and correlation coefficients of point-wise sensitivity (PWS) and mean sensitivity (MS).The algorithm achieved an MAE of 2.34 dB and 1.30 dB, an LoA of 5.70 and 3.07, a Pearson correlation coefficient of 0.66 and 0.84, and a Spearman correlation coefficient of 0.68 and 0.83 for PWS and MS, respectively. In the external test set, the method achieved an MAE of 2.73 dB and 1.66 dB for PWS and MS, respectively.The proposed approach allows the prediction of retinal function at each measured location directly based on an OCT scan, demonstrating how structural imaging can serve as a surrogate of visual function. Prospectively, the approach may help to complement retinal function measures, explore the association between image-based information and retinal functionality, improve disease progression monitoring, and provide objective surrogate measures for future clinical trials.

Purpose To prospectively evaluate the outcomes of different subtypes of neovascular age-related macular degeneration during intravitreal aflibercept monotherapy. Methods Forty-four eyes of 44 patients with treatment-naive polypoidal choroidal vasculopathy (PCV, n = 12), hemorrhagic choroidal neovascularization (hCNV, n = 12), pigment epithelium detachment (PED, n = 11), or retinal angiomatous proliferation (RAP, n = 9) were included and followed for 12 months. All patients received intravitreal aflibercept monotherapy. Results Mean visual acuity at baseline in PCV was 67 ± 16 Early Treatment Diabetic Retinopathy Study letters (20/50 Snellen equivalent), in hCNV 55 ± 21 (20/80), in RAP lesions 64 ± 11 (20/50), and in PED 74 ± 7 (20/32). At Month 12, visual acuity in PCV was 66 ± 16 (20/50), in hCNV 69 ± 17 (20/40), in RAP 68 ± 12 (20/50), and in PED 69 ± 18 (20/40). At the 12-month follow-up, visual acuity improved or was stable (±5 letters from baseline) in 84% of eyes (37/44 patients), with hCNV showing the greatest mean visual acuity gain. Mean central retinal thickness in patients with PCV was 523 ± 251 µm, in hCNV 497 ± 171, in RAP lesions 573 ± 132, and in PED 541 ± 158 and decreased to 310 ± 91 µm in PCV, 323 ± 75 µm in hCNV, 357 ± 173 µm in RAP lesions, and 422 ± 150 µm in PED. The mean area of atrophy increased from 2.0 ± 3.6 mm2 at baseline to 4.6 ± 8.6 mm2 at Month 12 (mean difference [95% confidence interval] -0.8 [-8.5 to 7.0], P = 0.8), with the greatest atrophy in patients with PED at Month 12. Conclusion All subtypes of neovascular age-related macular degeneration showed anatomical improvement and stabilization of visual function during intravitreal treatment.

AimsTo investigate the impact of posterior vitreous detachment (PVD) on the efficacy of treat-and-extend (T&E) ranibizumab in neovascular age-related macular degeneration.MethodsIn a post hoc analysis of a randomised controlled clinical trial, spectral-domain optical coherence tomography images of treatment-naïve patients randomised to receive T&E (n=265) or monthly (n=264) ranibizumab for 12 months were included. Certified, masked graders diagnosed the presence or the absence of complete PVD. The main outcome measures were the mean change in best-corrected visual acuity (BCVA) and central retinal thickness (CRT) at month 12, the number of administered ranibizumab injections and the proportion of patients extended to more than 8 weeks.ResultsAt baseline, complete PVD was present in 51% and 56% of patients in the monthly and T&E arms, respectively. Mean change in BCVA at month 12 was +9.0 (PVD) vs +9.5 letters (no PVD, p=0.78) in monthly treated eyes, and +6.0 (PVD) vs +7.5 letters (no PVD, p=0.42) in T&E treated eyes. Conversely, mean change in CRT at month 12 was −174 (PVD) vs −173 µm (no PVD, p=0.98) in the monthly arm, and −175 (PVD) vs −164 µm (no PVD, p=0.58) in the T&E arm. In T&E treated patients, the median number of injections was eight vs nine (p=0.035). 71% of PVD eyes were extended successfully, compared with 55% of eyes without PVD (p=0.005).ConclusionPVD was not found to impact functional and anatomical outcomes of T&E ranibizumab therapy. However, patients without a complete PVD required more retreatments and were significantly less likely to be successfully extended.Trial registration numberNCT01948830

Retinal swelling due to the accumulation of fluid is associated with the most vision-threatening retinal diseases. Optical coherence tomography (OCT) is the current standard of care in assessing the presence and quantity of retinal fluid and image-guided treatment management. Deep learning methods have made their impact across medical imaging, and many retinal OCT analysis methods have been proposed. However, it is currently not clear how successful they are in interpreting the retinal fluid on OCT, which is due to the lack of standardized benchmarks. To address this, we organized a challenge RETOUCH in conjunction with MICCAI 2017, with eight teams participating. The challenge consisted of two tasks: fluid detection and fluid segmentation. It featured for the first time: all three retinal fluid types, with annotated images provided by two clinical centers, which were acquired with the three most common OCT device vendors from patients with two different retinal diseases. The analysis revealed that in the detection task, the performance on the automated fluid detection was within the inter-grader variability. However, in the segmentation task, fusing the automated methods produced segmentations that were superior to all individual methods, indicating the need for further improvements in the segmentation performance.

Age-related macular degeneration (AMD) is the predominant cause of vision loss in the elderly with a major impact on ageing societies and healthcare systems. A major challenge in AMD management is the difficulty to determine the disease stage, the highly variable progression speed and the risk of conversion to advanced AMD, where irreversible functional loss occurs. In this study we developed an optical coherence tomography (OCT) imaging based spatio-temporal reference frame to characterize the morphologic progression of intermediate age-related macular degeneration (AMD) and to identify distinctive patterns of conversion to the advanced stages macular neovascularization (MNV) and macular atrophy (MA). We included 10,040 OCT volumes of 518 eyes with intermediate AMD acquired according to a standardized protocol in monthly intervals over two years. Two independent masked retina specialists determined the time of conversion to MNV or MA. All scans were aligned to a common reference frame by intra-patient and inter-patient registration. Automated segmentations of retinal layers and the choroid were computed and en-face maps were transformed into the common reference frame. Population maps were constructed in the subgroups converting to MNV (n=135), MA (n=50) and in non-progressors (n=333). Topographically resolved maps of changes were computed and tested for statistical significant differences. The development over time was analysed by a joint model accounting for longitudinal and right-censoring aspect. Significantly enhanced thinning of the outer nuclear layer (ONL) and retinal pigment epithelium (RPE)–photoreceptorinner segment/outer segment (PR-IS/OS) layers within the central 3 mm and a faster thinning speed preceding conversion was documented for MA progressors. Converters to MNV presented an accelerated thinning of the choroid and appearance changes in the choroid prior to MNV onset. The large-scale automated image analysis allowed us to distinctly assess the progression of morphologic changes in intermediate AMD based on conventional OCT imaging. Distinct topographic and temporal patterns allow to prospectively determine eyes with risk of progression and thereby greatly improving early detection, prevention and development of novel therapeutic strategies.

Nanophthalmos-4 is a rare autosomal dominant disorder caused by two known variations in TMEM98. An Austrian Caucasian pedigree was identified suffering from nanophthalmos and late onset angle-closure glaucoma and premature loss of visual acuity. Whole exome sequencing identified segregation of a c.602G > C transversion in TMEM98 (p.Arg201Pro) as potentially causative. A protein homology model generated showed a TMEM98 structure comprising α4, α5/6, α7 and α8 antiparallel helix bundles and two predicted transmembrane domains in α1 and α7 that have been confirmed in vitro. Both p.Arg201Pro and the two missense variations representing proline insertions identified previously to cause nanophthalmos-4 (p.Ala193Pro and p.His196Pro) are located in the charge polarized helix α8 (p.183-p210). Stability of the C-terminal alpha helical structure of TMEM98 is therefore essential to prevent the development of human nanophthalmos-4. Precise molecular diagnosis could lead to the development of tailored therapies for patients with orphan ocular disease.

Empirical models have been an integral part of everyday clinical practice in ophthalmology since the introduction of the Sanders-Retzlaff-Kraff (SRK) formula. Recent developments in the field of statistical learning (artificial intelligence, AI) now enable an empirical approach to a wide range of ophthalmological questions with an unprecedented precision.Which criteria must be considered for the evaluation of AI-related studies in ophthalmology?Exemplary prediction of visual acuity (continuous outcome) and classification of healthy and diseased eyes (discrete outcome) using retrospectively compiled optical coherence tomography data (50 eyes of 50 patients, 50 healthy eyes of 50 subjects). The data were analyzed with nested cross-validation (for learning algorithm selection and hyperparameter optimization).Based on nested cross-validation for training, visual acuity could be predicted in the separate test data-set with a mean absolute error (MAE, 95% confidence interval, CI of 0.142 LogMAR [0.077; 0.207]). Healthy versus diseased eyes could be classified in the test data-set with an agreement of 0.92 (Cohen's kappa). The exemplary incorrect learning algorithm and variable selection resulted in an MAE for visual acuity prediction of 0.229 LogMAR [0.150; 0.309] for the test data-set. The drastic overfitting became obvious on comparison of the MAE with the null model MAE (0.235 LogMAR [0.148; 0.322]).Selection of an unsuitable measure of the goodness-of-fit, inadequate validation, or withholding of a null or reference model can obscure the actual goodness-of-fit of AI models. The illustrated pitfalls can help clinicians to identify such shortcomings.HINTERGRUND: Empirische Modelle sind seit Einführung der SRK(Sanders-Retzlaff-Kraff)-Formel im klinischen Alltag der Augenheilkunde etabliert. Rezente Entwicklungen im Bereich des statistischen Lernens („künstliche Intelligenz“ [KI]) ermöglichen jetzt ein empirisches Vorgehen für vielfältigste ophthalmologische Fragestellungen bei bislang unerreichter Präzision.Welche Kriterien müssen für die Bewertung von Arbeiten zum Thema KI in der Augenheilkunde berücksichtigt werden?Es erfolgen die beispielhafte Vorhersage des Visus (stetige Zielgröße) und Klassifikation von gesunden und kranken Augen (diskrete Zielgröße) anhand von retrospektiven optischen Kohärenztomographiebilddaten (50 Augen von 50 Patienten, 50 gesunde Augen von 50 Probanden). Die Daten wurden mit verschachtelter Kreuzvalidierung (zur Lernalgorithmusauswahl und Hyperparameteroptimierung) analysiert.Durch verschachtelte Kreuzvalidierung ließ sich der Visus im separaten Testdatensatz mit einem mittleren absoluten Fehler (MAE, [95 %-CI, Konfidenzintervall]) von 0,142 LogMAR [0,077; 0,207] vorhersagen. Kranke und gesunde Augen ließen sich im Testdatensatz mit einer Konkordanz von (Kappa nach Cohen) 0,92 klassifizieren. Die beispielhafte inkorrekte Lernalgorithmus- und Variablenauswahl resultierte in einem MAE von 0,229 LogMAR [0,150; 0,309] für den Testdatensatz. Erst durch Vergleich mit dem MAE des Nullmodells (0,235 LogMAR [0,148; 0,322]) wurde die Überanpassung offensichtlich.Die Auswahl einer ungeeigneten Kennzahl für die Anpassungsgüte, inadäquate Validierung oder Unterschlagen eines Null- oder Referenzmodells kann die tatsächliche Anpassungsgüte von KI-Modellen verschleiern. Die illustrierten Fallstricke können Klinikern und Forschern helfen, solche Unzulänglichkeiten zu erkennen.

Artificial intelligence has recently made a disruptive impact in medical imaging by successfully automatizing expert-level diagnostic tasks. However, replicating human-made decisions may inherently be biased by the fallible and dogmatic nature of human experts, in addition to requiring prohibitive amounts of training data. In this paper, we introduce an unsupervised deep learning architecture particularly designed for OCT representations for unbiased, purely data-driven biomarker discovery. We developed artificial intelligence technology that provides biomarker candidates without any restricting input or domain knowledge beyond raw images. Analyzing 54,900 retinal optical coherence tomography (OCT) volume scans of 1094 patients with age-related macular degeneration, we generated a vocabulary of 20 local and global markers capturing characteristic retinal patterns. The resulting markers were validated by linking them with clinical outcomes (visual acuity, lesion activity and retinal morphology) using correlation and machine learning regression. The newly identified features correlated well with specific biomarkers traditionally used in clinical practice (r up to 0.73), and outperformed them in correlating with visual acuity ($$\hbox {R}^{2} = 0.46$$R2=0.46 compared to $$\hbox {R}^{2} = 0.29$$R2=0.29 for conventional markers), despite representing an enormous compression of OCT imaging data (67 million voxels to 20 features). In addition, our method also discovered hitherto unknown, clinically relevant biomarker candidates. The presented deep learning approach identified known as well as novel medical imaging biomarkers without any prior domain knowledge. Similar approaches may be worthwhile across other medical imaging fields.

Major advances in diagnostic technologies are offering unprecedented insight into the condition of the retina and beyond ocular disease. Digital images providing millions of morphological datasets can fast and non-invasively be analyzed in a comprehensive manner using artificial intelligence (AI). Methods based on machine learning (ML) and particularly deep learning (DL) are able to identify, localize and quantify pathological features in almost every macular and retinal disease. Convolutional neural networks thereby mimic the path of the human brain for object recognition through learning of pathological features from training sets, supervised ML, or even extrapolation from patterns recognized independently, unsupervised ML. The methods of AI-based retinal analyses are diverse and differ widely in their applicability, interpretability and reliability in different datasets and diseases. Fully automated AI-based systems have recently been approved for screening of diabetic retinopathy (DR). The overall potential of ML/DL includes screening, diagnostic grading as well as guidance of therapy with automated detection of disease activity, recurrences, quantification of therapeutic effects and identification of relevant targets for novel therapeutic approaches. Prediction and prognostic conclusions further expand the potential benefit of AI in retina which will enable personalized health care as well as large scale management and will empower the ophthalmologist to provide high quality diagnosis/therapy and successfully deal with the complexity of 21st century ophthalmology.

Prediction of treatment responses from available data is key to optimizing personalized treatment. Retinal diseases are treated over long periods and patients' response patterns differ substantially, ranging from a complete response to a recurrence of the disease and need for re-treatment at different intervals. Linking observable variables in high-dimensional observations to outcome is challenging. In this paper, we present and evaluate two different data-driven machine learning approaches operating in a high-dimensional feature space: sparse logistic regression and random forests-based extra trees (ET). Both identify spatio-temporal signatures based on retinal thickness features measured in longitudinal spectral-domain optical coherence tomography (OCT) imaging data and predict individual patient outcome using these quantitative characteristics. We demonstrate on a data set of monthly SD-OCT scans of 155 patients with central retinal vein occlusion (CRVO) and 92 patients with branch retinal vein occlusion (BRVO) followed over one year that we can predict from initial three observations if the treated disease will recur within the covered interval. ET predicts the outcome on fivefold cross-validation with an area under the receiver operating characteristic curve (AuC) of 0.83 for BRVO and 0.76 for CRVO. Logistic regression achieved an AuC of 0.78 and 0.79, respectively. At the same time, the methods identified stable predictive signatures in the longitudinal imaging data that are the basis for accurate prediction. Furthermore, our results show that taking spatio-temporal features into account improves accuracy compared with features extracted at a single time-point. Our results demonstrate the feasibility of mining longitudinal data for predictive signatures, and building predictive models based on observed data.

Background/aimsThe objective is to identify imaging biomarkers in optical coherence tomography predicting functional/anatomical outcomes in diabetic macular oedema (DMO).MethodsThe presented study is a post hoc analysis of the RESTORE/RESTORE-extension studies. Best-corrected visual acuity (BCVA) was analysed using general estimating equation models using treatment group/morphological features as predictor variables. In addition, linear multiple regression models analysed BCVA gain up to 12 and 36 months with BCVA/morphological baseline characteristics as independent predictor variables. The correlations between central retinal thickness (CRT)/BCVA were calculated as Spearman’s/Pearson’s correlation coefficients.ResultsA weak negative linear correlation between CRT/BCVA was observed in all study arms at baseline (r=−0.34, p380 µm (64.84±10.63 vs 61.66±9.92 letters; p=0.0071, respectively), which was maintained until the end of month 12 (70.5±12.33 vs 67.0±14.09 letters; p=0.0252, respectively). With laser, there was a trend for patients with subretinal fluid (SRF) at baseline to lose BCVA letters at month 12 (−5.38±16.54 vs 2.49±9.72 letters; p=0.1038), whereas ranibizumab patients trended towards higher BCVA gains (10.28±7.14 vs 6.76±7.67; p=0.0563), compared with those without SRF. With combined therapy, all patients had similar BCVA gains regardless of SRF (p=0.3768).ConclusionWith ranibizumab treatment, the height of IRC spaces at baseline was a better predictor of functional/anatomical improvement than CRT alone. There was also a trend for SRF to show a positive impact on ranibizumab therapy response and a negative impact on laser therapy response.

PurposeThe purpose of the present study is to develop fast automated quantification of retinal fluid in optical coherence tomography (OCT) image sets.MethodsWe developed an image analysis pipeline tailored towards OCT images that consists of five steps for binary retinal fluid segmentation. The method is based on feature extraction, pre-segmention, dimension reduction procedures, and supervised learning tools.ResultsFluid identification using our pipeline was tested on two separate patient groups: one associated to neovascular age-related macular degeneration, the other showing diabetic macular edema. For training and evaluation purposes, retinal fluid was annotated manually in each cross-section by human expert graders of the Vienna Reading Center. Compared with the manual annotations, our pipeline yields good quantification, visually and in numbers.ConclusionsBy demonstrating good automated retinal fluid quantification, our pipeline appears useful to expert graders within their current grading processes. Owing to dimension reduction, the actual learning part is fast and requires only few training samples. Hence, it is well-suited for integration into actual manufacturer's devices, further improving segmentation by its use in daily clinical life.

Diabetic macular edema (DME) and retina vein occlusion (RVO) are macular diseases in which central photoreceptors are affected due to pathological accumulation of fluid. Optical coherence tomography allows to visually assess and evaluate photoreceptor integrity, whose alteration has been observed as an important biomarker of both diseases. However, the manual quantification of this layered structure is challenging, tedious and time-consuming. In this paper we introduce a deep learning approach for automatically segmenting and characterising photoreceptor alteration. The photoreceptor layer is segmented using an ensemble of four different convolutional neural networks. En-face representations of the layer thickness are produced to characterize the photoreceptors. The pixel-wise standard deviation of the score maps produced by the individual models is also taken to indicate areas of photoreceptor abnormality or ambiguous results. Experimental results showed that our ensemble is able to produce results in pair with a human expert, outperforming each of its constitutive models. No statistically significant differences were observed between mean thickness estimates obtained from automated and manually generated annotations. Therefore, our model is able to reliable quantify photoreceptors, which can be used to improve prognosis and managment of macular diseases.

Procedures with artificial intelligence (AI), such as deep neural networks, show promising results in automatic analysis of ophthalmological imaging data.This article discusses to what extent the application of AI algorithms can contribute to quality assurance in the field of ophthalmology.Relevant aspects from the literature are discussed.Systems based on artificial deep neural networks achieve remarkable results in the diagnostics of eye diseases, such as diabetic retinopathy and are very helpful, for example by segmenting optical coherence tomographic (OCT) images and detecting lesion components with high fidelity. To train these algorithms large data sets are required. The quality and availability of such data sets determine the continuous improvement of the algorithms. The comparison between the AI algorithms and physicians for image interpretation has also enabled insights into the diagnostic concordance between physicians. Current challenges include the development of methods for modelling decision uncertainty and improved interpretability of automated diagnostic decisions.Systems based on AI can support decision making for physicians and thereby contribute to a more efficient quality assurance.

To evaluate multimodal imaging findings of solitary idiopathic choroiditis (SIC; also known as unifocal helioid choroiditis) to clarify its origin, anatomic location, and natural course.Multicenter retrospective observational case series.Sixty-three patients with SIC in 1 eye.Demographic and clinical data were collected. Multimodal imaging included color fundus photography, OCT (including swept-source OCT), OCT angiography (OCTA), fundus autofluorescence, fluorescein and indocyanine green angiography, and B-scan ultrasonography.Standardized grading of imaging features.Mean age at presentation was 56 ± 15 years (range, 12-83 years). Mean follow-up duration in 39 patients was 39 ± 55 months (range, 1 month-25 years). The lesions measured a mean of 2.4 × 2.1 mm in basal diameter, were located inferior (64%) or nasal to the optic disc, and appeared yellow (53%). No systemic associations were found. The lesions all appeared as an elevated subretinal mass, with OCT demonstrating all lesions to be confined to the sclera, not the choroid. On OCT, the deep lesion margin was visible in 12 eyes with a mean lesion thickness of 0.6 mm. Overlying choroidal thinning or absence was seen in 95% (mean choroidal thickness, 28 ± 35 μm). Mild subretinal fluid was observed overlying the lesions in 9 patients (14%). Retinal pigment epithelial disruption and overlying retinal thinning was observed in 56% and 57%, respectively. OCT angiography was performed in 13 eyes and demonstrated associated choroidal and lesional flow voids. Four lesions (6%) were identified at the macula, leading to visual loss in 1 patient. One lesion demonstrated growth and another lesion showed spontaneous resolution.In this largest series to date, multimodal imaging of SIC demonstrated a scleral location in all patients. The yellow and white clinical appearance may be related to scleral unmasking resulting from atrophy of overlying tissues. Additional associated features included documentation of deep margin on swept-source OCT, trace subretinal fluid in a few patients, and OCTA evidence of lesional flow voids. Because of the scleral location of this lesion in every patient, a new name, focal scleral nodule, is proposed.

Purpose To quantify morphologic photoreceptor integrity during anti-vascular endothelial growth factor (anti-VEGF) therapy of neovascular age-related macular degeneration and correlate these findings with disease morphology and function. Methods This presents a post hoc analysis on spectral-domain optical coherence tomography data of 185 patients, acquired at baseline, Month 3, and Month 12 in a multicenter, prospective trial. Loss of the ellipsoid zone (EZ) was manually quantified in all optical coherence tomography volumes. Intraretinal cystoid fluid, subretinal fluid (SRF), and pigment epithelial detachments were automatically segmented in the full volumes using validated deep learning methods. Spatiotemporal correlation of fluid markers with EZ integrity as well as bivariate analysis between EZ integrity and best-corrected visual acuity was performed. Results At baseline, EZ integrity was predominantly impaired in the fovea, showing progressive recovery during anti-vascular endothelial growth factor therapy. Topographic analysis at baseline revealed EZ integrity to be more likely intact in areas with SRF and vice versa. Moreover, we observed a correlation between EZ integrity and resolution of SRF. Foveal EZ integrity correlated with best-corrected visual acuity at all timepoints. Conclusion Improvement of EZ integrity during anti-VEGF therapy of neovascular age-related macular degeneration occurred predominantly in the fovea. Photoreceptor integrity correlated with best-corrected visual acuity. Ellipsoid zone integrity was preserved in areas of SRF and showed deterioration upon SRF resolution.

Diagnosis and treatment guidance are aided by detecting relevant biomarkers in medical images. Although supervised deep learning can perform accurate segmentation of pathological areas, it is limited by requiring a-priori definitions of these regions, large-scale annotations, and a representative patient cohort in the training set. In contrast, anomaly detection is not limited to specific definitions of pathologies and allows for training on healthy samples without annotation. Anomalous regions can then serve as candidates for biomarker discovery. Knowledge about normal anatomical structure brings implicit information for detecting anomalies. We propose to take advantage of this property using bayesian deep learning, based on the assumption that epistemic uncertainties will correlate with anatomical deviations from a normal training set. A Bayesian U-Net is trained on a well-defined healthy environment using weak labels of healthy anatomy produced by existing methods. At test time, we capture epistemic uncertainty estimates of our model using Monte Carlo dropout. A novel post-processing technique is then applied to exploit these estimates and transfer their layered appearance to smooth blob-shaped segmentations of the anomalies. We experimentally validated this approach in retinal optical coherence tomography (OCT) images, using weak labels of retinal layers. Our method achieved a Dice index of 0.789 in an independent anomaly test set of age-related macular degeneration (AMD) cases. The resulting segmentations allowed very high accuracy for separating healthy and diseased cases with late wet AMD, dry geographic atrophy (GA), diabetic macular edema (DME) and retinal vein occlusion (RVO). Finally, we qualitatively observed that our approach can also detect other deviations in normal scans such as cut edge artifacts., Accepted for publication in IEEE Transactions on Medical Imaging, 2019

Purpose To characterize retinal morphology differences among different types of choroidal neovascularization and visual function changes at the onset of exudative age-related macular degeneration. Methods In a post hoc analysis of a prospective clinical study, 1,097 fellow eyes from subjects with choroidal neovascularization in the study eye enrolled in the HARBOR trial were evaluated. The onset of exudation was diagnosed on monthly optical coherence tomography by two masked graders. At conversion as well as 1 month earlier, pigment epithelial detachment, intraretinal cystoid fluid, subretinal fluid, subretinal hyperreflective material, as well as ellipsoid zone and external limiting membrane loss were quantitatively analyzed. Hyperreflective foci, retinal pigment epithelial defects, haze and vitreoretinal interface status were evaluated qualitatively. Main outcome measures included visual acuity and rates of morphologic features at conversion and 1 month earlier. Results New-onset exudation was detected in 92 eyes. One month before conversion, hyperreflective foci, pigment epithelial detachment, retinal pigment epithelial defects, and haze were present in the majority of eyes. At the onset of exudation, the volumes of intraretinal cystoid fluid, subretinal fluid, subretinal hyperreflective material and pigment epithelial detachment, and the areas of external limiting membrane and ellipsoid zone loss significantly increased. The mean vision loss was -2.2 letters. Pathognomonic patterns of the different choroidal neovascularization types were already apparent 1 month before conversion. Conclusion Characteristic choroidal neovascularization-associated morphological changes are preceding disease conversion, while vision loss at the onset of exudation is minimal. Individual lesion types are related to specific changes in optical coherence tomography morphology already before the time of conversion. Our findings may help to elucidate the pathophysiology of neovascular age-related macular degeneration and support the diagnosis of imminent disease conversion.

Optical coherence tomography (OCT) has become an established diagnostic technology in the clinical management of age-related macular degeneration (AMD). OCT is being used for primary diagnosis, evaluation of therapeutic efficacy, and long-term monitoring. Computer-based advances in image analysis provide complementary imaging tools such as OCT angiography, further novel automated analysis methods as well as feature detection and prediction of prognosis in disease and therapy by machine learning. In early AMD, pathognomonic features such as drusen, pseudodrusen, and abnormalities of the retinal pigment epithelium (RPE) can be imaged in a qualitative and quantitative way to identify early signs of disease activity and define the risk of progression. In advanced AMD, disease activity can be monitored clearly by qualitative and quantified analyses of fluid pooling, such as intraretinal cystoid fluid, subretinal fluid, and pigment epithelial detachment (PED). Moreover, machine learning methods detect a large spectrum of new biomarkers. Evaluation of treatment efficacy and definition of optimal therapeutic regimens are an important aim in managing neovascular AMD. In atrophic AMD hallmarked by geographic atrophy (GA), advanced spectral domain (SD)-OCT imaging largely replaces conventional fundus autofluorescence (FAF) as it adds insight into the condition of the neurosensory layers and associated alterations at the level of the RPE and choroid. Exploration of imaging features by computerized methods has just begun but has already opened relevant and reliable horizons for the optimal use of OCT imaging for individualized and population-based management of AMD—the leading retinal epidemic of modern times.

Importance The morphologic changes and their pathognomonic distribution in progressing age-related macular degeneration (AMD) are not well understood. Objectives To characterize the pathognomonic distribution and time course of morphologic patterns in AMD and to quantify changes distinctive for progression to macular neovascularization (MNV) and macular atrophy (MA). Design, Setting, and Participants This cohort study included optical coherence tomography (OCT) volumes from study participants with early or intermediate AMD in the fellow eye in the HARBOR (A Study of Ranibizumab Administered Monthly or on an As-needed Basis in Patients With Subfoveal Neovascular Age-Related Macular Degeneration) trial. Patients underwent imaging monthly for 2 years (July 1, 2009, to August 31, 2012) following a standardized protocol. Data analysis was performed from June 1, 2018, to January 21, 2020. Main Outcomes and Measures To obtain topographic correspondence between patients and over time, all scans were mapped into a joint reference frame. The time of progression to MNV and MA was established, and drusen volumes and hyperreflective foci (HRF) volumes were automatically segmented in 3 dimensions using validated artificial intelligence algorithms. Topographically resolved population means of these markers were constructed by averaging quantified drusen and HRF maps in the patient subgroups. Results Of 1097 patients enrolled in HARBOR, 518 (mean [SD] age, 78.1 [8.2] years; 309 [59.7%] female) had early or intermediate AMD in the fellow eye at baseline. During the 24-month follow-up period, 135 (26%) eyes developed MNV, 50 eyes (10%) developed MA, and 333 (64%) eyes did not progress to advanced AMD. Drusen and HRF had distinct topographic patterns. Mean drusen thickness at the fovea was 29.6 μm (95% CI, 20.2-39.0 μm) for eyes progressing to MNV, 17.2 μm (95% CI, 9.8-24.6 μm) for eyes progressing to MA, and 17.1 μm (95% CI, 12.5-21.7 μm) for eyes without disease progression. At 0.5-mm eccentricity, mean drusen thickness was 25.8 μm (95% CI, 19.1-32.5 μm) for eyes progressing to MNV, 21.7 μm (95% CI, 14.6-28.8 μm) for eyes progressing to MA, and 14.4 μm (95% CI, 11.2-17.6 μm) for eyes without disease progression. The mean HRF thickness at the foveal center was 0.072 μm (95% CI, 0-0.152 μm) for eyes progressing to MNV, 0.059 μm (95% CI, 0-0.126 μm) for eyes progressing to MA, and 0.044 μm (95% CI, 0.007-0.081) for eyes without disease progression. At 0.5-mm eccentricity, the largest mean HRF thickness was seen in eyes progressing to MA (0.227 μm; 95% CI, 0.104-0.349 μm) followed by eyes progressing to MNV (0.161 μm; 95% CI, 0.101-0.221 μm) and eyes without disease progression (0.085 μm; 95% CI, 0.058-0.112 μm). Conclusions and Relevance In this study, drusen and HRF represented imaging biomarkers of disease progression in AMD, demonstrating distinct topographic patterns over time that differed between eyes progressing to MNV, eyes progressing to MA, or eyes without disease progression. Automated localization and precise quantification of these factors may help to develop reliable methods of predicting future disease progression.

Purpose While millions of individuals show early age-related macular degeneration (AMD) signs, yet have excellent vision, the risk of progression to advanced AMD with legal blindness is highly variable. We suggest means of artificial intelligence to individually predict AMD progression. Methods In eyes with intermediate AMD, progression to the neovascular type with choroidal neovascularization (CNV) or the dry type with geographic atrophy (GA) was diagnosed based on standardized monthly optical coherence tomography (OCT) images by independent graders. We obtained automated volumetric segmentation of outer neurosensory layers and retinal pigment epithelium, drusen, and hyperreflective foci by spectral domain-OCT image analysis. Using imaging, demographic, and genetic input features, we developed and validated a machine learning-based predictive model assessing the risk of conversion to advanced AMD. Results Of a total of 495 eyes, 159 eyes (32%) had converted to advanced AMD within 2 years, 114 eyes progressed to CNV, and 45 to GA. Our predictive model differentiated converting versus nonconverting eyes with a performance of 0.68 and 0.80 for CNV and GA, respectively. The most critical quantitative features for progression were outer retinal thickness, hyperreflective foci, and drusen area. The features for conversion showed pathognomonic patterns that were distinctly different for the neovascular and the atrophic pathways. Predictive hallmarks for CNV were mostly drusen-centric, while GA markers were associated with neurosensory retina and age. Conclusions Artificial intelligence with automated analysis of imaging biomarkers allows personalized prediction of AMD progression. Moreover, pathways of progression may be specific in respect to the neovascular/atrophic type.

Obtaining expert labels in clinical imaging is difficult since exhaustive annotation is time-consuming. Furthermore, not all possibly relevant markers may be known and sufficiently well described a priori to even guide annotation. While supervised learning yields good results if expert labeled training data is available, the visual variability, and thus the vocabulary of findings, we can detect and exploit, is limited to the annotated lesions. Here, we present fast AnoGAN (f-AnoGAN), a generative adversarial network (GAN) based unsupervised learning approach capable of identifying anomalous images and image segments, that can serve as imaging biomarker candidates. We build a generative model of healthy training data, and propose and evaluate a fast mapping technique of new data to the GAN's latent space. The mapping is based on a trained encoder, and anomalies are detected via a combined anomaly score based on the building blocks of the trained model - comprising a discriminator feature residual error and an image reconstruction error. In the experiments on optical coherence tomography data, we compare the proposed method with alternative approaches, and provide comprehensive empirical evidence that f-AnoGAN outperforms alternative approaches and yields high anomaly detection accuracy. In addition, a visual Turing test with two retina experts showed that the generated images are indistinguishable from real normal retinal OCT images. The f-AnoGAN code is available at https://github.com/tSchlegl/f-AnoGAN.

The purpose of the study was to create a standardised protocol for choroidal thickness measurements and to determine whether choroidal thickness measurements made on images obtained by spectral domain optical coherence tomography (SD-OCT) and swept source (SS-) OCT from patients with healthy retina are interchangeable when performed manually or with an automatic algorithm.36 grid cell measurements for choroidal thickness for each volumetric scan were obtained, which were measured for SD-OCT and SS-OCT with two methods on 18 eyes of healthy volunteers. Manual segmentation by experienced retinal graders from the Vienna Reading Center and automated segmentation on6300 images of the choroid from both devices were statistically compared.Model-based comparison between SD-OCT/SS-OCT showed a systematic difference in choroidal thickness of 16.26±0.725 μm (p0.001) for manual segmentation and 21.55±0.725 μm (p0.001) for automated segmentation. Comparison of automated with manual segmentations revealed small differences in thickness of -0.68±0.513 μm (p=0.1833). The correlation coefficients for SD-OCT and SS-OCT measures within eyes were 0.975 for manual segmentation and 0.955 for automatic segmentation.Choroidal thickness measurements of SD-OCT and SS-OCT indicate that these two devices are interchangeable with a trend of choroidal thickness measurements being slightly thicker on SD-OCT with limited clinical relevance. Use of an automated algorithm to segment choroidal thickness was validated in healthy volunteers.

Neovascular age-related macular degeneration (AMD) has undergone substantial break-throughs in diagnostic as well as therapeutic respect, with optical coherence tomography (OCT) allowing to identify disease morphology in great detail, and intravitreal anti-vascular endothelial growth factor therapy providing unprecedented benefit. However, these two paths have yet not been combined in an optimal way, real-world outcomes are inferior to expectations, and disease management is largely inefficient in the real-world setting. This dilemma can be solved by identification of valid biomarkers relevant for visual function, disease activity and prognosis, which can provide solid guidance for therapeutic management on an individual level as well as on the population base. Qualitative and quantitative morphological features obtained by advanced OCT provide novel insight into exudative and degenerative stages of neovascular AMD. However, conclusions from structure/function correlations evolve differently from previous paradigms. While central retinal thickness was used as biomarker for guiding retreatment management in clinical trials and practice, fluid localization in different compartments offers superior prognostic value: Intraretinal cystoid fluid has a negative impact on visual acuity and is considered as degenerative when persisting through the initial therapeutic interval. Subretinal fluid is associated with superior visual benefit and a lower rate of progression towards geographic atrophy. Detachment of the retinal pigment epithelium was identified as most pathognomonic biomarker, often irresponsive to therapy and responsible for visual decline during a pro-re-nata regimen. Alterations of neurosensory tissue are usually associated with irreversible loss of functional elements and a negative prognosis. Novel OCT technologies offer crucial insight into corresponding changes at the level of the photoreceptor--retinal pigment epithelial--choriocapillary unit, identifying the biological limits of therapeutic interventions. To optimally benefit from high-resolution multi-modal imaging, an integrated analysis of all functional and structural features is required involving reliable automated algorithms and computational data analyses. Using innovative analysis methods, retinal biomarkers can be used to provide efficient personalized therapy for the individual patient, predictive disease- and population-based models for large-scale management and identifying promising targets for the development of novel therapeutic strategies.

Purpose Intravitreal antiangiogenic therapy is the major therapeutic breakthrough in neovascular age-related macular degeneration (AMD). Optical coherence tomography (OCT) is the leading diagnostic tool, but solid criteria for optimal therapeutic outcomes are lacking. A comprehensive analysis of structure/function correlations using Food and Drug Administration- and European Medicines Agency–approved substances and fixed and flexible regimens was performed. Design Post hoc analysis of a prospective, randomized multicenter clinical trial including 189 study sites. Participants A total of 1240 patients with active neovascular AMD. Methods Participants received intravitreal ranibizumab or aflibercept. A fixed regimen was used for 48 weeks followed by a flexible regimen until week 96. At monthly intervals, best-corrected visual acuity (BCVA) was measured and retinal morphology was assessed by standardized OCT, including intraretinal cysts (IRCs), subretinal fluid (SRF), and pigment epithelial detachment (PED), presenting with a width ≥400 μm or a height of ≥200 μm. Results were correlated for each regimen, feature, and time. Main Outcome Measures The BCVA outcomes in relation to retinal pathomorphology based on noninferiority for all treatment arms. Results In neovascular AMD, only IRC at baseline and persistent through week 12 had a negative impact on BCVA. With therapeutic intervention, exudative features such as IRC and SRF resolved rapidly in 74% of eyes, whereas PED responded only slowly with 38%. Independent of the type of regimen, fixed or flexible, retinal morphology correlated tightly with visual function. Intraretinal cysts consistently showed the lowest BCVA gains with either regimen or substance. With the switch from a fixed to a flexible pro re nata (PRN) regimen, progressive visual loss occurred exclusively in the group with primary PED presenting as the hallmark of neovascular activity and was induced by secondary formation of IRC in the neurosensory retina. Conclusions The efficacy of antiangiogenic therapy in neovascular AMD is strongly determined by morphologic features. The subretinal pigment epithelium lesion underlying PED appears to be the primary indicator for progressive disease activity, whereas secondary cystoid degeneration is the most relevant imaging marker for visual function. Clinically, PED emerged as trigger for consecutive vision loss in PRN treatment.

IMPORTANCE: Anti–vascular endothelial growth factor treatment is the first-line therapy in the treatment of center-involving diabetic macular edema. Data on capillary perfusion changes under repeated treatment in a possibly compromised vascular network are limited. OBJECTIVE: To evaluate the association of repeated ranibizumab injections on macular perfusion in patients with diabetic macular edema. DESIGN, SETTING, AND PARTICIPANTS: This study analyzed prospectively collected data from the 12-month RESTORE core study and the 24-month open label RESTORE extension study, which assessed the efficacy and safety of ranibizumab in patients with visual impairment due to diabetic macular edema. Of 345 patients with center-involving diabetic macular edema who had enrolled in the 12-month RESTORE core study, 240 entered the 24-month RESTORE extension study. Of these, 83 (34.6%) received ranibizumab, 83 (34.6%) received ranibizumab and laser combination therapy, and 74 (30.8%) received laser monotherapy in the first year of the study; 208 completed the 24-month extension study. Fluorescence angiography images were taken from each participant twice each year graded by Vienna Reading Center on severity of capillary loss in the parafoveal area, regularity of the foveal avascular zone outline, and measurement of the size of the foveal avascular zone, following a standardized protocol. Data analysis took place from July 2014 through December 2017. MAIN OUTCOMES AND MEASURES: Change in 3 fluorescence angiography perfusion parameters over the course of treatment. RESULTS: Mean (SD) patient age was 62.6 (8.8) years; 124 of 208 (59.2%) were male and 197 of 208 (94.6%) were white. The number of patients with definite altered foveal avascular zone regularity at baseline was 103 of 240 patients (42.9%); another 118 patients (49.2%) had questionably altered regularity at baseline. Definitive capillary loss was found in 65 of 240 patients (27.1%) at baseline. Mean (SD) foveal avascular zone size at baseline was 0.261 (0.232) mm(2) in ranibizumab monotherapy, 0.231 (0.219) mm(2) in ranibizumab and macular laser combination therapy, and 0.201 (0.13) mm(2) in laser monotherapy. No treatment arm experienced significant increase in foveal avascular zone size at any time in the study period. At month 36, ranibizumab monotherapy resulted in a mean increase of 0.073 mm(2) (95% CI, 0.005-0.142 mm(2)) and combination therapy resulted in a mean increase of 0.117 mm(2) (95% CI, 0.045-0.188 mm2), but no changes were statistically significant. No changes occurred in foveal avascular zone regularity in any treatment group, and no differences were found in capillary loss around the fovea in the 3 treatment groups; neither element could be correlated with visual acuity or central retinal thickness. CONCLUSIONS AND RELEVANCE: Repeated ranibizumab treatment was not associated with impaired macular perfusion in our study cohort. Because our data do not suggest a harmful effect of anti–vascular endothelial growth factor therapy on capillary integrity, patients with severe microangiopathy and advanced capillary dropout should not be denied these treatments.

To compare signal penetration depth and deep structure-visualization of swept source (SS) and spectral domain (SD)-optical coherence tomography (OCT) with and without enhanced depth imaging (EDI) and B-scan averaging modes.Volume scans were obtained from 20 eyes of healthy volunteers by DRI OCT-1, Spectralis using EDI and B-scan averaging, and Cirrus HD-OCT. The signal penetration depth was measured as the distance between the retinal pigment epithelium and the deepest visible anatomical structure at the foveal center. Visibility and contrast of the choroidoscleral junction and of vascular details within the choroid were assessed across the entire volume using an ordinal scoring scale. Outcome measures were compared using paired t-test and rank-sum test.The mean signal penetration depth was 498±114 μm for Spectralis, 491±85 μm for DRI OCT-1, and 123±65 μm for Cirrus; P=0.9708 Spectralis vs DRI OCT-1, P0.0001 Spectralis vs Cirrus, and P0.0001 DRI OCT-1 vs Cirrus. Mean ranks for visibility and contrast of the choroidoscleral junction were 3.83 for Spectralis, 3.98 for DRI OCT-1, and 2.00 for Cirrus; and 3.45 for Spectralis, 2.93 for DRI OCT-1, and 1.58 for Cirrus. Mean ranks for visibility and contrast of vascular details were 3.73 (Spectralis), 3.70 (DRI OCT-1), and 2.23 (Cirrus); and 3.53 (Spectralis), 2.05 (DRI OCT-1), and 1.98 (Cirrus).Signal penetration depths are similar for SS-OCT and SD-OCT using EDI and frame averaging, and statistically significantly lower without EDI/averaging. Both SD-OCT using EDI/frame averaging and SS-OCT offer excellent visualization capabilities for volumetric imaging of the choroidoscleral interface.

Comparison of optical coherence tomography (OCT) segmentation performance regarding technical accuracy and clinical relevance.29 eyes were imaged prospectively with Spectralis (Sp), Cirrus (Ci), 3D-OCT 2000 (3D) and RS-3000 (RS) OCTs. Raw data were evaluated in validated custom software. A 1 mm diameter subfield, centred on the fovea, was investigated to compare identical regions for each case. Segmentation errors were corrected on each B-scan enclosed in this subfield. Proportions of wrongly segmented A-scans were noted for inner and outer retinal boundaries. Centre point thickness (CPT) and central macular thickness (CMT) were compared before and after correction.Segmentation errors occurred in 77% and affected on average 29% of A-scans, resulting in mean differences of 24/13 µm (CPT/CMT). The incidence of segmentation errors was 48% (Sp), 79% (Ci), 86% (3D) and 93% (RS), p0.001. Mean proportions of A-scans with wrong outer retinal boundary were 30% (Sp), 9% (Ci), 23% (3D) and 10% (RS), p=0.006; proportions for the inner retinal boundary were 11% (Sp), 12% (Ci), 6% (3D) and 21% (RS), p=0.034. Mean deviations in CPT/CMT were 41/28 µm (Sp), 17/11 µm (Ci), 30/13 µm (3D) and 18/8 µm (RS), p=0.409/0.477.By comparison of identical regions, substantial differences were detected between the tested OCT devices regarding technical accuracy and clinical impact. Spectralis showed lowest error incidence but highest error impact.

Background/aimsTo characterise neuroretinal atrophy in retinal vein occlusion (RVO).MethodsWe included patients with central/branch RVO (CRVO=196, BRVO=107) who received ranibizumab according to a standardised protocol for 6 months. Retinal atrophy was defined as the presence of an area of retinal thickness (RT) Results23 patients with CRVO and 11 patients with BRVO demonstrated retinal atrophy, presenting as sharply demarcated retinal thinning confined to a macular quadrant. The mean RT in the atrophic quadrant at month 6 was 249±26 µm (CRVO) and 244±29 µm (BRVO). Individual layer analysis revealed pronounced thinning in the IPL to ONL compartment. Change in BCVA at 6 months was similar between the groups (BRVO, +15 vs +18 letters; CRVO, +14 vs +18 letters).ConclusionsIn this exploratory analysis, we describe the characteristics of neuroretinal atrophy in RVO eyes with resolved macular oedema after ranibizumab therapy. Our analysis shows significant, predominantly retinal thinning in the IPL to ONL compartment in focal macular areas in 11% of patients with RVO. Eyes with retinal atrophy did not show poorer BCVA outcomes.

Purpose We develop a longitudinal statistical model describing best-corrected visual acuity (BCVA) changes in anti-VEGF therapy in relation to imaging data, and predict the future BCVA outcome for individual patients by combining population-wide trends and initial subject-specific time points. Methods Automatic segmentation algorithms were used to measure intraretinal (IRF) and subretinal (SRF) fluid volume on monthly spectral-domain optical coherence tomography scans of eyes with central retinal vein occlusion (CRVO) receiving standardized anti-VEGF treatment. The trajectory of BCVA over time was modeled as a multivariable repeated-measure mixed-effects regression model including fluid volumes as covariates. Subject-specific BCVA trajectories and final treatment outcomes were predicted using a population-wide model and individual observations from early follow-up. Results A total of 193 eyes (one per patient, 12-month follow-up, 2420 visits) were analyzed. The population-wide mixed model revealed that the impact of fluid on BCVA is highest for IRF in the central millimeter around the fovea, with -31.17 letters/mm3 (95% confidence interval [CI], -39.70 to -23.32), followed by SRF in the central millimeter, with -17.50 letters/mm3 (-31.17 to -4.60) and by IRF in the parafovea, with -2.87 letters/mm3 (-4.71 to -0.44). The influence of SRF in the parafoveal area was -1.24 letters/mm3 (-3.37-1.05). The conditional R2 of the model, including subject-specific deviations, was 0.887. The marginal R2 considering the population-wide trend and fluid changes was 0.109. BCVA at 1 year could be predicted for an individual patient after three visits with a mean absolute error of six letters and a predicted R2 of 0.658 using imaging information. Conclusions The mixed-effects model revealed that retinal fluid volumes and population-wide trend only explains a small proportion of the variation in BCVA. Individual BCVA outcomes after 1 year could be predicted from initial BCVA and fluid measurements combined with the population-wide model. Accounting for fluid in the predictive model increased prediction accuracy.

Vitreomacular adhesion (VMA) represents a prognostic biomarker in the management of exudative macular disease using anti-vascular endothelial growth factor (VEGF) agents. However, manual evaluation of VMA in 3D optical coherence tomography (OCT) is laborious and data on its impact on therapy of retinal vein occlusion (RVO) are limited. The aim of this study was to (1) develop a fully automated segmentation algorithm for the posterior vitreous boundary and (2) to study the effect of VMA on anti-VEGF therapy for RVO. A combined machine learning/graph cut segmentation algorithm for the posterior vitreous boundary was designed and evaluated. 391 patients with central/branch RVO under standardized ranibizumab treatment for 6/12 months were included in a systematic post-hoc analysis. VMA (70%) was automatically differentiated from non-VMA (30%) using the developed method combined with unsupervised clustering. In this proof-of-principle study, eyes with VMA showed larger BCVA gains than non-VMA eyes (BRVO: 15 ± 12 vs. 11 ± 11 letters, p = 0.02; CRVO: 18 ± 14 vs. 9 ± 13 letters, p

Purpose Development and validation of a fully automated method to detect and quantify macular fluid in conventional OCT images. Design Development of a diagnostic modality. Participants The clinical dataset for fluid detection consisted of 1200 OCT volumes of patients with neovascular age-related macular degeneration (AMD, n = 400), diabetic macular edema (DME, n = 400), or retinal vein occlusion (RVO, n = 400) acquired with Zeiss Cirrus (Carl Zeiss Meditec, Dublin, CA) (n = 600) or Heidelberg Spectralis (Heidelberg Engineering, Heidelberg, Germany) (n = 600) OCT devices. Methods A method based on deep learning to automatically detect and quantify intraretinal cystoid fluid (IRC) and subretinal fluid (SRF) was developed. The performance of the algorithm in accurately identifying fluid localization and extent was evaluated against a manual consensus reading of 2 masked reading center graders. Main Outcome Measures Performance of a fully automated method to accurately detect, differentiate, and quantify intraretinal and SRF using area under the receiver operating characteristics curves, precision, and recall. Results The newly designed, fully automated diagnostic method based on deep learning achieved optimal accuracy for the detection and quantification of IRC for all 3 macular pathologies with a mean accuracy (AUC) of 0.94 (range, 0.91–0.97), a mean precision of 0.91, and a mean recall of 0.84. The detection and measurement of SRF were also highly accurate with an AUC of 0.92 (range, 0.86–0.98), a mean precision of 0.61, and a mean recall of 0.81, with superior performance in neovascular AMD and RVO compared with DME, which was represented rarely in the population studied. High linear correlation was confirmed between automated and manual fluid localization and quantification, yielding an average Pearson's correlation coefficient of 0.90 for IRC and of 0.96 for SRF. Conclusions Deep learning in retinal image analysis achieves excellent accuracy for the differential detection of retinal fluid types across the most prevalent exudative macular diseases and OCT devices. Furthermore, quantification of fluid achieves a high level of concordance with manual expert assessment. Fully automated analysis of retinal OCT images from clinical routine provides a promising horizon in improving accuracy and reliability of retinal diagnosis for research and clinical practice in ophthalmology.

Purpose To investigate the influence of the vitreomacular interface (VMI) on the functional and anatomic efficacy of ranibizumab therapy in patients with neovascular age-related macular degeneration (AMD). Design Subanalysis of a prospective, 12-month, multicenter, phase IIIb trial. Participants A total of 353 treatment-naive patients with subfoveal choroidal neovascularization (CNV) receiving quarterly or monthly ranibizumab therapy. Methods On monthly optical coherence tomography (OCT) scan sets, the VMI configuration was graded by a certified reading center into one of the following conditions: continuous posterior vitreoretinal attachment (PVA), vitreomacular adhesion (VMA), partial vitreous detachment without vitreomacular contact, or complete posterior vitreous detachment (PVD). Best-corrected visual acuity (BCVA) and central retinal thickness (CRT) measurements were performed at monthly intervals. Analysis included patients with a minimum of 10 OCT examinations, including baseline and month 12 (n = 251). After integration of the VMI configuration over 12 months, patients were divided into one of the following categories: PVD (n = 162), release of vitreomacular contact (RELEASE; n = 48), VMA (n = 37), or PVA (n = 4). General estimation equation analyses were applied to test for noninferiority of quarterly versus monthly treatment. Main Outcome Measures The BCVA and CRT changes at month 12. Results Mean BCVA changes in letters were +4.7 (PVD), +3.2 (RELEASE), and −0.2 (VMA) in the quarterly regimen and +4.9 (PVD), +12.7 (RELEASE), and +7.5 (VMA) in the monthly regimen. No difference in therapeutic efficiency between monthly and quarterly intervention was found in eyes with PVD, and quarterly treatment was noninferior to monthly treatment ( P = 0.001). However, monthly treatment was superior to quarterly treatment in the RELEASE ( P = 0.008) and VMA ( P = 0.043) groups. Mean CRT changes were −98 and −96 μm (PVD), −117 and −136 μm (RELEASE), and −93 and −87 μm (VMA) in the monthly and quarterly regimens, respectively, without statistically significant differences. Conclusions The configuration of the VMI seems to have an important effect on visual outcomes and need for retreatment. In patients with PVD, a lower treatment frequency may be feasible, whereas patients with RELEASE or VMA may benefit from intensive retreatment. These findings may serve as a basis for individualized treatment decisions in anti-angiogenic therapy of neovascular AMD and perhaps other indications. Financial Disclosure(s) Proprietary or commercial disclosure may be found after the references.

Purpose To evaluate the potential of machine learning to predict best-corrected visual acuity (BCVA) outcomes from structural and functional assessments during the initiation phase in patients receiving standardized ranibizumab therapy for neovascular age-related macular degeneration (AMD). Design Post hoc analysis of a randomized, prospective clinical trial. Participants Data of 614 evaluable patients receiving intravitreal ranibizumab monthly or pro re nata according to protocol-specified criteria in the HARBOR trial. Methods Monthly spectral-domain (SD) OCT volume scans were processed by validated, fully automated computational image analysis. This system performs spatially resolved 3-dimensional segmentation of retinal layers, intraretinal cystoid fluid (IRF), subretinal fluid (SRF), and pigment epithelial detachments (PED). The extracted quantitative OCT biomarkers and BCVA measurements at baseline and months 1, 2, and 3 were used to predict BCVA at 12 months using random forest machine learning. This approach was also used to correlate OCT morphology to BCVA at baseline (structure–function correlation). Main Outcome Measures Accuracy (R 2 ) of the prediction models; ranking of input variables. Results Computational image analysis enabled fully automated quantitative characterization of neovascular lesions in a large-scale clinical SD-OCT data set. At baseline, OCT features and BCVA were correlated with R 2 = 0.21. The most relevant biomarker for BCVA was the horizontal extension of IRF in the foveal region, whereas SRF and PED ranked low. In predicting functional outcomes, the model's accuracy increased in a linear fashion with each month. If only the baseline visit was considered, the accuracy was R 2 = 0.34. At month 3, final visual acuity outcomes could be predicted with an accuracy of R 2 = 0.70. The strongest predictive factor for functional outcomes at 1 year was consistently the individual BCVA level during the initiation phase. Conclusions In this large-scale study based on a wide spectrum of morphologic and functional features, baseline BCVA correlated modestly with baseline SD-OCT, whereas functional outcomes were determined by BCVA levels during the initiation phase with a minor influence of fluid-related features. This finding suggests a re-evaluation of current diagnostic imaging features and a search for novel imaging approaches, where machine learning is a promising approach.

To assess the efficacy and safety profile of an individualized, stabilization criteria-driven regimen of ranibizumab 0.5 mg in patients with visual impairment due to macular edema secondary to central retinal vein occlusion (CRVO).A 24-month, prospective, open-label, single-arm, multicenter study.A total of 357 patients.Patients received monthly ranibizumab 0.5 mg injections (minimum, 3 injections) until stable visual acuity (VA) was maintained for 3 consecutive months. Thereafter, ranibizumab 0.5 mg injections were administered if monitoring indicated a loss of VA due to disease activity. The primary outcome results have been published previously.Mean change from baseline at months 1 through 24 in best-corrected VA (BCVA) in the overall population and in subgroups categorized according to baseline BCVA, CRVO duration, or presence of macular ischemia.The baseline mean BCVA was 53.0 letters and baseline mean CRVO duration was 8.9 months (median, 2.4 months). The mean (standard deviation) gain in BCVA from baseline with ranibizumab 0.5 mg at month 24 was 12.1 (18.60) letters (P0.0001). Best-corrected VA gains at month 24 were similar in patients with or without baseline macular ischemia (mean change, 11.1 and 12.9 letters, respectively). The mean BCVA gain at month 24 was higher in patients with CRVO duration3 months (13.2 letters) compared with that in those with CRVO duration9 months (10.5 letters). Patients with lower baseline BCVA had larger mean BCVA gains at month 24 (≤39 letters; 18.5 letters) than those with higher baseline BCVA (40-59/≥60 letters; 13.9/7.2 letters), although the absolute BCVA values at month 24 were higher in patients with higher baseline BCVA. The mean (standard deviation) and median number of ranibizumab injections up to month 23 were 13.1 (6.39) and 15.0 injections, respectively. No new ocular or nonocular safety events were reported.An individualized, stabilization criteria-driven dosing regimen of ranibizumab 0.5 mg led to sustained BCVA gains for up to 24 months in patients with CRVO. The presence of macular ischemia at baseline did not influence VA gains. Shorter duration of CRVO at baseline was associated with better VA gains. Safety findings were consistent with those reported in previous ranibizumab studies in patients with CRVO.

To evaluate the long-term (24-month) efficacy and safety of ranibizumab 0.5 mg administered pro re nata (PRN) with or without laser using an individualized visual acuity (VA) stabilization criteria in patients with visual impairment due to macular edema secondary to branch retinal vein occlusion (BRVO).Phase IIIb, open-label, randomized, active-controlled, 3-arm, multicenter study.A total of 455 patients.Patients were randomized (2:2:1) to ranibizumab 0.5 mg (n = 183), ranibizumab 0.5 mg with laser (n = 180), or laser (with optional ranibizumab 0.5 mg after month 6; n = 92). After initial 3 monthly injections, patients in the ranibizumab with or without laser arms received VA stabilization criteria-driven PRN treatment. Patients assigned to the laser arm received laser at the investigator's discretion.Mean (and mean average) change in best-corrected visual acuity (BCVA) and central subfield thickness (CSFT) from baseline to month 24, and safety over 24 months.A total of 380 patients (83.5%) completed the study. Ranibizumab with or without laser led to superior BCVA outcomes versus laser (monotherapy and combined with ranibizumab from month 6; 17.3/15.5 vs. 11.6 letters; P0.0001). Ranibizumab with laser was noninferior to ranibizumab monotherapy (mean average BCVA change: 15.4 vs. 15.0 letters; P0.0001). However, addition of laser did not reduce the number of ranibizumab injections (mean injections: 11.4 vs. 11.3; P = 0.4259). A greater reduction in CSFT was seen with ranibizumab with or without laser versus laser monotherapy over 24 months from baseline (ranibizumab monotherapy -224.7 μm, ranibizumab with laser -248.9 μm, laser [monotherapy and combined with ranibizumab from month 6] -197.5 μm). Presence of macular ischemia did not affect BCVA outcome or treatment frequency. There were no reports of neovascular glaucoma or iris neovascularization. No new safety signals were identified.The BRIGHTER study results confirmed the long-term efficacy and safety profile of PRN dosing driven by individualized VA stabilization criteria using ranibizumab 0.5 mg in patients with BRVO. Addition of laser did not lead to better functional outcomes or lower treatment need. The safety results were consistent with the well-established safety profile of ranibizumab.

In this pilot study, we evaluated the potential of computational image analysis of optical coherence tomography (OCT) data to determine the prognosis of patients with diabetic macular edema (DME). Spectral-domain OCT scans with fully automated retinal layer segmentation and segmentation of intraretinal cystoid fluid (IRC) and subretinal fluid of 629 patients receiving anti-vascular endothelial growth factor therapy for DME in a randomized prospective clinical trial were analyzed. The results were used to define 312 potentially predictive features at three timepoints (baseline, weeks 12 and 24) for best-corrected visual acuity (BCVA) at baseline and after one year used in a random forest prediction path. Preliminarily, IRC in the outer nuclear layer in the 3-mm area around the fovea seemed to have the greatest predictive value for BCVA at baseline, and IRC and the total retinal thickness in the 3-mm area at weeks 12 and 24 for BCVA after one year. The overall model accuracy was R2=0.21/0.23 (p

Purpose To investigate the influence of vitreomacular adhesion (VMA) on development of choroidal neovascularization (CNV) in eyes with age-related macular degeneration. Methods In a prospective study, patients with Age-Related Eye Disease Study Category IV age-related macular degeneration underwent standardized examinations, including optical coherence tomography and fluorescein angiography every 3 months for 4 years. Vitreomacular adhesion was evaluated using time- and spectral-domain optical coherence tomography. Development of CNV was detected using fluorescein angiography and optical coherence tomography. Incidences of CNV were compared concerning the presence or absence of VMA. Results Forty-nine patients were available for follow-up according to protocol. Vitreomacular adhesion was present at baseline in 18% (9 of 49) and absent in 82% (40 of 49) of patients. Thirty-seven percent of patients (18 of 49) developed exudative changes during the observation period. In patients with preexisting VMA, de novo development of CNV occurred in 33% (3 of 9). In patients without VMA, 38% developed CNV (15 of 40). Mean interval from baseline to disease progression was 20 ± 19 months in patients with VMA and 22 ± 13 months in patients without VMA. There was no significant difference between the groups regarding rate of CNV development or time to disease progression (P = 0.64). Conclusion No significant influence of VMA on the development of exudative age-related macular degeneration could be found during a 4-year prospective observation of a high-risk cohort.

Purpose To compare the efficacy of intravitreal aflibercept and ranibizumab on the exudative activity of neovascular age-related macular degeneration (nAMD) using optical coherence tomography (OCT) and to correlate morphologic findings with visual acuity (VA) outcomes. Design Post hoc analysis of the prospective VIEW trials. Participants Data of 1815 patients randomized to 0.5 mg ranibizumab every 4 weeks (Q4wks), 2 mg aflibercept Q4wks, or 2 mg aflibercept every 8 weeks (Q8wks). Methods Standardized OCT evaluation was performed by masked reading centers for the presence of intraretinal cystoid fluid (IRC), subretinal fluid (SRF), and pigment epithelial detachment (PED). Rates of feature resolution were compared between drugs and regimen. Associations between morphologic features and VA were analyzed using multivariate modeling. Main Outcome Measures Resolution rates of IRC, SRF, and PED, and associations between morphology and VA. Results At baseline, the proportions of eyes with IRC, SRF, and PED were balanced between the aflibercept and ranibizumab groups. At week 12, IRC resolved in 50% of eyes with both agents. Subretinal fluid resolved in 70% of pooled aflibercept-treated eyes and in 59% of ranibizumab-treated eyes, and PED resolved in 29% and 24% of pooled aflibercept-treated eyes and ranibizumab-treated eyes, respectively. At week 52, IRC resolved in 57% (aflibercept Q4wks), 50% (aflibercept Q8wks), and 52% (ranibizumab) of patients; SRF resolved in 75% (both aflibercept Q4wks/Q8wks) and 66% (ranibizumab) of patients; and PED resolved in 40% (aflibercept Q4wks), 34% (aflibercept Q8wks), and 28% (ranibizumab) of patients. During fixed dosing (weeks 12–52) all exudative features showed synchronized fluctuations after treatment-free visits in the Q8wks aflibercept regimen. During pro re nata dosing (weeks 52–96), greater proportions of patients showed recurrent fluid in all treatment arms. Presence of IRC was generally associated with lower VA at baseline, which translated into poorer final VA outcomes. Conclusions Fluid resolution in all compartments was consistently greater for aflibercept Q4wks than for aflibercept Q8wks and ranibizumab. At week 52, Q8wks aflibercept–treated eyes were, on average, as dry as or drier than with ranibizumab despite the extended treatment interval. Independent of agent or regimen, preexisting morphologic features of the retina at baseline markedly influenced VA outcomes.

To assess the 12-month efficacy and safety profile of an individualized regimen of ranibizumab 0.5 mg driven by stabilization criteria in patients with macular edema secondary to central retinal vein occlusion (CRVO).A 24-month, prospective, open-label, single-arm, multicenter study.Three hundred fifty-seven patients.Patients were treated with monthly ranibizumab 0.5-mg injections (minimum of 3 injections) until stable visual acuity (VA) was maintained for 3 consecutive months. Thereafter, ranibizumab 0.5 mg was dosed as needed if monthly monitoring indicated a loss of VA resulting from disease activity.Mean change from baseline at month 12 in best-corrected VA (BCVA; primary end point) and safety over 12 months. The efficacy of this regimen in subgroups categorized by baseline BCVA score, CRVO duration, or presence of macular ischemia (exploratory analysis).At baseline, the mean BCVA was 53.0 letters and mean CRVO duration was 8.9 months (median, 2.4 months). Ranibizumab 0.5-mg treatment resulted in a statistically significant mean gain in BCVA from baseline at month 12 of 12.3 letters (standard deviation [SD], 16.72 letters; P0.0001). The mean number of ranibizumab injections up to month 12 was 8.1 (SD, 2.77). At month 12, mean BCVA gains were similar with or without macular ischemia at baseline (11.6 vs. 12.1 letters); the mean BCVA gain was higher with baseline CRVO duration of less than 3 months (13.4 letters) than with a longer duration (≥3-9 months, 11.1 letters; ≥9 months, 10.9 letters). Patients with lower baseline BCVA had larger mean BCVA gains at month 12 than those with higher baseline BCVA (≤39/40-59/≥60 and 18.0/12.7/8.9 letters, respectively), although the absolute BCVA at month 12 was higher with higher baseline BCVA. No new ocular or nonocular safety events were observed.An individualized dosing regimen of ranibizumab 0.5 mg driven by stabilization criteria for up to 12 months resulted in significant BCVA gain in a broad population of patients with macular edema secondary to CRVO, including those with macular ischemia at baseline. The safety findings were consistent with those reported in previous ranibizumab studies in patients with CRVO.

To compare the 6-month efficacy and safety profile of an individualized stabilization criteria-driven pro re nata (PRN) regimen of ranibizumab 0.5 mg with or without laser versus laser alone in patients with visual impairment due to macular edema secondary to branch retinal vein occlusion (BRVO).A 24-month, prospective, open-label, randomized, active-controlled, multicenter, phase IIIb study.A total of 455 patients.Eligible patients were randomized 2:2:1 to receive ranibizumab (n = 183), ranibizumab with laser (n = 180), or laser only (n = 92). Patients treated with ranibizumab with or without laser received a minimum of 3 initial monthly ranibizumab injections until visual acuity (VA) stabilization, and VA-based PRN dosing thereafter. In the ranibizumab with laser and laser-only groups, laser was given at the investigator's discretion at a minimum interval of 4 months and if VA was79 letters.Mean change from baseline at month 6 in best-corrected visual acuity (BCVA) (primary end point) and central subfield thickness, and safety over 6 months. Exploratory objectives were to evaluate the influence of baseline BCVA, disease duration, and ischemia on BCVA outcomes at month 6.Baseline mean BCVA was 57.7 letters, and mean BRVO duration was 9.9 months. Ranibizumab with or without laser was superior to laser only in improving mean BCVA from baseline at month 6 (14.8 and 14.8 vs. 6.0 letters; both P0.0001; primary end point met). Patients with a shorter BRVO duration at baseline had a higher mean BCVA gain than those with a longer BRVO duration. Patients with a poor baseline VA had a better BCVA gain than those with a higher baseline VA, although final BCVA was lower in those with poor baseline VA. In the ranibizumab with or without laser groups, the presence of some macular ischemia at baseline did not influence mean BCVA gains. There were no new ocular or nonocular safety events.Ranibizumab with an individualized VA-based regimen, with or without laser, showed statistically significant superior improvement in BCVA compared with laser alone in patients with BRVO. Overall, there were no new safety events other than those reported in previous studies.

A framework including data selection, task assignment and annotation combination stages for a confidence based benchmark dataset for retinal image processing is proposed.A novel task assignment is used to remove data and reader biases.The annotation of readers is combined based on their accuracy and performance.The framework is used to build a confidence based benchmark dataset for cyst segmentation.The generated benchmark can be used to reliably evaluate cyst segmentation methods. Background and objectivesThe lack of benchmark data in computational ophthalmology contributes to the challenging task of applying disease assessment and evaluate performance of machine learning based methods on retinal spectral domain optical coherence tomography (SD-OCT) scans. Presented here is a general framework for constructing a benchmark dataset for retinal image processing tasks such as cyst, vessel, and subretinal fluid segmentation and as a result, a benchmark dataset for cyst segmentation has been developed. MethodFirst, a dataset captured by different SD-OCT vendors with different numbers of scans and pathology qualities are selected. Then a robust and intelligent method is used to evaluate performance of readers, partitioning the dataset into subsets. Subsets are then assigned to complementary readers for annotation with respect to a novel confidence based annotation protocol. Finally, reader annotations are combined based on their performance to generate final annotations. ResultThe generated benchmark dataset for cyst segmentation comprises 26 SD-OCT scans with differing cyst qualities, collected from 4 different SD-OCT vendors to cover a wide variety of data. The dataset is partitioned into three subsets which are annotated by complementary readers based on a confidence based annotation protocol. Experimental results show annotations of complementary readers are combined efficiently with respect to their performance, generating accurate annotations. ConclusionOur results facilitate the process of generating benchmark datasets. Moreover the generated benchmark data set for cyst segmentation can be used reliably to train and test machine learning based methods.

We propose a method to predict treatment response patterns based on spatio-temporal disease signatures extracted from longitudinal spectral domain optical coherence tomography (SD-OCT) images. We extract spatio-temporal disease signatures describing the underlying retinal structure and pathology by transforming total retinal thickness maps into a joint reference coordinate system. We formulate the prediction as a multi-variate sparse generalized linear model regression based on the aligned signatures. The algorithm predicts if and when recurrence of the disease will occur in the future. Experiments demonstrate that the model identifies predictive and interpretable features in the spatio-temporal signature. In initial experiments recurrence vs. non-recurrence is predicted with a ROC AuC of 0.99. Based on observed longitudinal morphology changes and a time-to-event based Cox regression model we predict the time to recurrence with a mean absolute error (MAE) of 1.25 months comparing favorably to elastic net regression (1.34 months), demonstrating the benefit of a spatio-temporal survival model.

Learning representative computational models from medical imaging data requires large training data sets. Often, voxel-level annotation is unfeasible for sufficient amounts of data. An alternative to manual annotation, is to use the enormous amount of knowledge encoded in imaging data and corresponding reports generated during clinical routine. Weakly supervised learning approaches can link volume-level labels to image content but suffer from the typical label distributions in medical imaging data where only a small part consists of clinically relevant abnormal structures. In this paper we propose to use a semantic representation of clinical reports as a learning target that is predicted from imaging data by a convolutional neural network. We demonstrate how we can learn accurate voxel-level classifiers based on weak volume-level semantic descriptions on a set of 157 optical coherence tomography (OCT) volumes. We specifically show how semantic information increases classification accuracy for intraretinal cystoid fluid (IRC), subretinal fluid (SRF) and normal retinal tissue, and how the learning algorithm links semantic concepts to image content and geometry.

To establish the predictive value of defined retinal morphologic parameters on visual outcomes and re-treatment needs in patients with neovascular age-related macular degeneration (nAMD) receiving ranibizumab treatment.Post hoc analysis of a prospective, 12-month, multicenter, phase IIIb trial.Three hundred fifty-three treatment-naïve patients with nAMD.Available data from 319 treatment-naïve patients receiving ranibizumab 0.3 mg monthly (frequent regimen; n = 102) or ranibizumab 0.3 or 0.5 mg quarterly (pooled 0.3/0.5 mg = infrequent regimen; n = 217) were analyzed to assess the correlations between baseline retinal morphologic parameters and best-corrected visual acuity (BCVA) change (structure-function correlations). The BCVA was measured at monthly visits. Optical coherence tomography scans were acquired monthly for quantitative measures of the central retinal thickness and qualitative assessment of retinal morphologic features. Assessed morphologic parameters included intraretinal cystoid fluid (IRC), subretinal fluid (SRF), pigment epithelial detachment, and vitreomacular interface configuration classification comprising vitreomacular adhesion and posterior vitreous detachment (PVD). An analysis of covariance was conducted to evaluate the impact of retinal morphologic features on BCVA change at month 12.Change in BCVA from baseline to month 12 compared between frequent and infrequent treatment arms.Relevant predictive factors for BCVA change at month 12 were baseline SRF (P = 0.05), PVD (P = 0.03), IRC (P = 0.05), treatment frequency (P0.01), and BCVA (P0.01). The presence of both SRF and PVD at baseline was associated with similar BCVA gains regardless of treatment frequency (mean difference in BCVA gains at month 12 of +2.6 letters in favor of infrequent treatment). Subretinal fluid was present in 71% of patients, and PVD was present in 64% of patients.In patients with both SRF and PVD at baseline, similar BCVA outcomes were observed regardless of treatment frequency. These patients may require less frequent treatments compared with patients without SRF, without PVD, or without either who may require more frequent injections for maintenance of vision. This finding may have implications in clinical practice by helping to tailor an individualized re-treatment interval in nAMD patients.

Purpose To identify the spatial distribution of exudative features of choroidal neovascularization in neovascular age-related macular degeneration (nAMD) based on the localization of intraretinal cystoid fluid (IRC), subretinal fluid (SRF), and pigment-epithelial detachment (PED). Methods This retrospective cross-sectional study included spectral-domain optical coherence tomography volume scans (6 × 6 mm) of 1341 patients with treatment-naive nAMD. IRC, SRF, and PED were detected on a per-voxel basis using fully automated segmentation algorithms. Two subsets of 37 volumes each were manually segmented to validate the automated results. The spatial correspondence of components was quantified by computing proportions of IRC-, SRF-, or PED-presenting A-scans simultaneously affected by the respective other pathomorphologic components on a per-patient basis. The median across the population is reported. Odds ratios between pairs of lesions were calculated and tested for significance pixel wise. Results Automated image segmentation was successful in 1182 optical coherence tomography volumes, yielding more than 61 million A-scans for analysis. Overall, 81% of eyes showed IRC, 95% showed SRF, and 92% showed PED. IRC-presenting A-scans also showed SRF in a median 2.5%, PED in 32.9%. Of the SRF-presenting A-scans, 0.3% demonstrated IRC, 1.4% PED. Of the PED-presenting A-scans, 5.2% contained IRC, 2.0% SRF. Similar patterns were observed in the manually segmented subsets and via pixel-wise odds ratio analysis. Conclusions Automated analyses of large-scale datasets in a cross-sectional study of 1182 patients with active treatment-naive nAMD demonstrated low spatial correlation of SRF with IRC and PED in contrast to increased colocalization of IRC and PED. These morphological associations may contribute to our understanding of functional deficits in nAMD.