Search

To present a deep learning algorithm for segmentation of geographic atrophy (GA) using en face swept-source OCT (SS-OCT) images that is accurate and reproducible for the assessment of GA growth over time.Retrospective review of images obtained as part of a prospective natural history study.Patients with GA (n = 90), patients with early or intermediate age-related macular degeneration (n = 32), and healthy controls (n = 16).An automated algorithm using scan volume data to generate 3 image inputs characterizing the main OCT features of GA-hypertransmission in subretinal pigment epithelium (sub-RPE) slab, regions of RPE loss, and loss of retinal thickness-was trained using 126 images (93 with GA and 33 without GA, from the same number of eyes) using a fivefold cross-validation method and data augmentation techniques. It was tested in an independent set of one hundred eighty 6 × 6-mmThe GA area, enlargement rate of GA area, square root of GA area, and square root of the enlargement rate of GA area measurements were calculated using the automated algorithm and compared with ground truth calculations performed by 2 manual graders. The repeatability of these measurements was determined using intraclass coefficients (ICCs).There were no significant differences in the GA areas, enlargement rates of GA area, square roots of GA area, and square roots of the enlargement rates of GA area between the graders and the automated algorithm. The algorithm showed high repeatability, with ICCs of 0.99 and 0.94 for the GA area measurements and the enlargement rates of GA area, respectively. The repeatability limit for the GA area measurements made by grader 1, grader 2, and the automated algorithm was 0.28, 0.33, and 0.92 mmWhen compared with manual methods, this proposed deep learning-based automated algorithm for GA segmentation using en face SS-OCT images was able to accurately delineate GA and produce reproducible measurements of the enlargement rates of GA.

To determine whether the outer retinal layer (ORL) thickness around geographic atrophy (GA) could serve as a clinical biomarker to predict the annual enlargement rate (ER) of GA.Retrospective analysis of a prospective, observational case series.Eyes with GA were imaged with a swept-source OCT 6 × 6 mm scan pattern. GA lesions were measured from customized en face OCT images and the annual ERs were calculated. The ORL was defined and segmented from the inner boundary of outer plexiform layer (OPL) to the inner boundary of retinal pigment epithelium (RPE) layer. The ORL thickness was measured at different subregions around GA.A total of 38 eyes from 27 participants were included. The same eyes were used for the choriocapillaris (CC) flow deficit (FD) analysis and the RPE to the Bruch membrane (RPE-BM) distance measurements. A negative correlation was observed between the ORL thickness and the GA growth. The ORL thickness in a 300-μm rim around GA showed the strongest correlation with the GA growth (r = -0.457, P = .004). No correlations were found between the ORL thickness and the CC FDs; however, a significant correlation was found between the ORL thickness and the RPE-BM distances around GA (r = -0.398, P = .013).ORL thickness showed a significant negative correlation with annual GA growth, but also showed a significant correlation with the RPE-BM distances, suggesting that they were dependently correlated with GA growth. This finding suggests that the loss of photoreceptors was associated with the formation of basal laminar deposits around GA.

Age-related macular degeneration (AMD) is a leading cause of severe vision loss. With our aging population, it may affect 288 million people globally by the year 2040. AMD progresses from an early and intermediate dry form to an advanced one, which manifests as choroidal neovascularization and geographic atrophy. Conversion to AMD-related exudation is known as progression to neovascular AMD, and presence of geographic atrophy is known as progression to advanced dry AMD. AMD progression predictions could enable timely monitoring, earlier detection and treatment, improving vision outcomes. Machine learning approaches, a subset of artificial intelligence applications, applied on imaging data are showing promising results in predicting progression. Extracted biomarkers, specifically from optical coherence tomography scans, are informative in predicting progression events. The purpose of this mini review is to provide an overview about current machine learning applications in artificial intelligence for predicting AMD progression, and describe the various methods, data-input types, and imaging modalities used to identify high-risk patients. With advances in computational capabilities, artificial intelligence applications are likely to transform patient care and management in AMD. External validation studies that improve generalizability to populations and devices, as well as evaluating systems in real-world clinical settings are needed to improve the clinical translations of artificial intelligence AMD applications.

PURPOSE We sought to determine the earliest diagnostic signs of hydroxychloroquine retinopathy up to the point of clinical recognition. METHODS Retrospective series of 6 patients (5 parafoveal disease; 1 pericentral disease) with at least 3 examinations over 3.5 years or more preceding diagnosis of HCQ retinopathy. Spectral domain optical coherence tomography (sdOCT) cross-sections, fundus autofluorescence (FAF) and visual fields were generated clinically. Stored sdOCT data were re-examined later to generate topographic ellipsoid zone (EZ) maps, minimum intensity (MI) analysis and sequential plots of regional retinal thickness. Retrospective series of six patients (5 parafoveal disease; one pericentral disease) with at least three examinations over 3.5 years or more preceding diagnosis of hydroxychloroquine retinopathy. RESULTS Spectral domain optical coherence tomography cross-sections and fields showed similar sensitivity; fundus autofluorescence was not helpful. In parafoveal cases, EZ topography and minimum intensity analysis were no more reliable. Sequential thickness plots from four parafoveal cases showed dramatic retinal thinning across the posterior pole beginning 4 years to 5 years before clinical diagnosis, with parafoveal regions thinning even faster. The pericentral case showed thinning only outside the central macula. Peripheral EZ loss was more dramatic with EZ topography than sdOCT cross-sections. CONCLUSION Sequential retinal thickness plots reveal definitive thinning years before current diagnostic procedures. We hope that OCT manufacturers will develop software to display such measurements. Ellipsoid zone topography was not more sensitive than sdOCT cross-sections, but important for recognizing pericentral disease.

The purpose of the study was to evaluate the retinal blood flow in patients with acute central serous chorioretinopathy (CSC) over an observational period of 1 month using swept-source optical coherence tomography (SS-OCTA), focusing especially on changes in the area of subretinal fluid (A-SRF). We correlated these findings with conventional indocyanine green angiography (ICGA). ICGA and SS-OCTA images were collected and analyzed of 12 eyes of 12 patients. The A-SRF was annotated and a qualitative analysis of choriocapillaris, the vessel density (VD) and perfusion density (PD) of the retinal superficial capillary plexus (SCP) and the deep capillary plexus (DCP) was performed in A-SRF and the unaffected remaining area (RA). The VD and PD in the DCP were statistically significantly lower in A-SRF than in the RA at baseline. (VD: p = 0.014; PD: p = 0.036). After 1 month, there was a statistically significant difference in the VD and PD of the DCP (VD: p = 0.015; PD: p = 0.014), and for the PD of the SCP between the A-SRF and the RA (p = 0.015), with lower values in the A-SRF. We found low perfused areas in choriocapillaris corresponding to hypofluorescent areas on ICGA. In conclusion there is a difference in VD and VD of the DCP in the area of SRF in acute CSC. These alterations may lead to a chronic change in the microvasculature and potentially to morphological changes.

PurposeTo test whether a single or composite set of parameters evaluated with optical coherence tomography angiography (OCTA), representing retinal capillary closure, can predict non-proliferative diabetic retinopathy (NPDR) staging according to the gold standard ETDRS grading scheme.Methods105 patients with diabetes, either without retinopathy or with different degrees of retinopathy (NPDR up to ETDRS grade 53), were prospectively evaluated using swept-source OCTA (SS-OCTA, PlexElite, Carl Zeiss Meditec) with 15×9 mm and 3×3 mm angiography protocols. Seven-field photographs of the fundus were obtained for ETDRS staging. Eyes from age-matched healthy subjects were also imaged as control.ResultsIn eyes of patients with type 2 diabetes without retinopathy or ETDRS levels 20 and 35, retinal capillary closure was in the macular area, with predominant alterations in the parafoveal retinal circulation (inner ring). Retinal capillary closure in ETDRS stages 43–53 becomes predominant in the retinal midperiphery with vessel density average values of 25.2±7.9 (p=0.001) in ETDRS 43 and 23.5±3.4 (p=0.001) in ETDRS 47–53, when evaluating extended areas of 15×9 protocol. Combination of acquisition protocols 3×3 mm and 15×9 mm, using SS-OCTA, allows discrimination between eyes with mild NPDR (ETDRS 10, 20, 35) and eyes with moderate-to-severe NPDR (ETDRS grades 43–53).ConclusionsRetinal capillary closure, quantified by SS-OCTA, can identify NPDR severity progression. It is located mainly in the perifoveal retinal capillary circulation in the initial stages of NPDR, whereas the retinal midperiphery is predominantly affected in moderate-to-severe NPDR.

Purpose A compensation strategy that was developed to measure the choriocapillaris (CC) flow deficits (FDs) under drusen was tested in eyes with large drusen from age-related macular degeneration (AMD) before and after the drusen spontaneously resolved without evidence of disease progression. Design Prospective, observational consecutive case series. Methods Patients with AMD were enrolled in a prospective swept-source optical coherence tomography (SS-OCT) imaging study. Consecutive eyes with large drusen were followed, and eyes that underwent spontaneous collapse of drusen without evidence of disease progression were identified retrospectively. The drusen-resolved regions were manually outlined. CC FDs were measured using a previously published compensation strategy that adjusted for the decreased signal intensity underlying drusen. Both the percentage of FDs (FD%) and the mean FD sizes (MFDSs) were measured before and after drusen resolution. Results Resolution of drusen was identified in 8 eyes from 8 patients. The average interval between the 2 visits was 7.8 months. The average drusen volumes measured between visits were 0.23 and 0.04 mm3, respectively. After the drusen resolved, the average follow-up time without evidence of disease progression was 10.1 months. When the 2 visits were compared, there were no statistically significant differences in any of the CC parameters within the drusen resolved regions once the compensation strategy was applied (all P values >.22). Conclusions In this naturally occurring experiment in which drusen collapsed without evidence of disease progression, the CC parameters were similar once our compensation strategy was applied both before and after the drusen resolved.

A deep learning algorithm was developed to automatically identify, segment, and quantify geographic atrophy (GA) based on optical attenuation coefficients (OACs) calculated from optical coherence tomography (OCT) datasets. Normal eyes and eyes with GA secondary to age-related macular degeneration were imaged with swept-source OCT using 6 × 6 mm scanning patterns. OACs calculated from OCT scans were used to generate customized composite en face OAC images. GA lesions were identified and measured using customized en face sub-retinal pigment epithelium (subRPE) OCT images. Two deep learning models with the same U-Net architecture were trained using OAC images and subRPE OCT images. Model performance was evaluated using DICE similarity coefficients (DSCs). The GA areas were calculated and compared with manual segmentations using Pearson’s correlation and Bland-Altman plots. In total, 80 GA eyes and 60 normal eyes were included in this study, out of which, 16 GA eyes and 12 normal eyes were used to test the models. Both models identified GA with 100% sensitivity and specificity on the subject level. With the GA eyes, the model trained with OAC images achieved significantly higher DSCs, stronger correlation to manual results and smaller mean bias than the model trained with subRPE OCT images (0.940 ± 0.032 vs 0.889 ± 0.056, p = 0.03, paired t-test, r = 0.995 vs r = 0.959, mean bias = 0.011 mm vs mean bias = 0.117 mm). In summary, the proposed deep learning model using composite OAC images effectively and accurately identified, segmented, and quantified GA using OCT scans.

PURPOSE: To determine age-related changes in choroidal thickness and the volume of choroidal vessels and stroma using automated algorithms based on structural swept source OCT (SS-OCT) scans. DESIGN: Prospective and observational study. PARTICIPANTS: One hundred and forty-four normal subjects with ages ranged from 20 to 88 years. METHODS: A previously reported strategy was used to automatically segment the choroid using SS-OCT structural images. Attenuation correction was applied on B-scans to enhance the choroidal contrast and facilitate more accurate automatic segmentation of the 3D choroidal vessel and stroma. The parameters investigated included mean choroid thickness (MCT), choroidal vessel volume (CVV), choroidal stroma volume (CSV), choroid vascularity index (CVI), and the choroidal stroma-to-vessel volume ratio (CSVR). Correlations between MCT and choroidal vessel metrics of CVV, CSV, CVI, and CSVR were studied. Regional distributions of MCT and CVI were analyzed using a grid centered on the fovea. Age-related changes in MCT, CVV, CSV, and CVI were studied in the whole scanning region, as well as in the sub-regions of the grids. MAIN OUTCOME MEASURES: Age-related changes in MCT, CVV, CSV, and CVI using 6×6 mm and 12×12 mm SS-OCT scans. RESULTS: The automated choroid segmentations were validated against manual segmentations, and MCT measurements were shown to be in good agreement (P

Purpose The purpose of this study was to validate a novel automated swept source optical coherence tomography angiography (SS-OCTA) algorithm to measure elevations of the retinal pigment epithelium (RPE) in eyes with nonexudative age-related macular degeneration (neAMD). Methods Patients with drusen were enrolled in a prospective optical coherence tomography (OCT) study and underwent both spectral domain OCT (SD-OCT) and SS-OCTA imaging at the same visit using the 6 × 6 mm scan patterns. The RPE elevation measurements (square root area and cube root volume) from the SS-OCTA algorithm were compared with the automated validated SD-OCT algorithm on the instrument. Standard deviations of drusen measurements from four repeated scans of another separate set were also calculated to evaluate the reproducibility of the SS-OCTA algorithm. Results A total of 53 eyes from 28 patients were scanned on both instruments. A very strong correlation was found between the measurements from the two algorithms (all r > 0.95), although the measurements of the drusen area and volume were all larger from the SS-OCTA instrument. The reproducibility of the new SS-OCTA algorithm was analyzed using a sample of 66 eyes from 43 patients. The intraclass correlation coefficient (ICC) was greater than 99% from different macular regions for both the square root area and cube root volume measurements. Conclusions A novel automated SS-OCTA algorithm for the quantitative assessment of drusen was validated against the SD-OCT algorithm and was shown to be highly reproducible. Translational Relevance This novel SS-OCTA algorithm provides a strategy to measure the area and volume of drusen to assess disease progression in neAMD.

Purpose To investigate the characteristics of treatment-naive nonexudative macular neovascularization (MNV) in age-related macular degeneration before the onset of exudation using swept-source optical coherence tomography angiography. Methods MNV area, choriocapillaris (CC) flow deficits (FDs), vessel area density (VAD), vessel skeleton density (VSD), retinal pigment epithelial detachment (PED) volume, mean choroidal thickness (MCT), and choroid vascularity index (CVI) measurements were assessed at two visits prior to exudation. We compared measurements made at the second visit and the rate of change between visits in eyes with and without exudation. The differences in these parameters between eyes with and without subsequent exudation were summarized with 95% confidence intervals (CIs). Results Twenty-one eyes with nonexudative MNV were identified and followed. Nine eyes developed exudation, and 12 eyes did not develop exudation. Differences between these groups of eyes for all parameters tended to be small, and the 95% CIs largely ruled out any substantial differences. Overall, eyes with exudation had 24% smaller VAD, 20% smaller VSD, and 33% smaller PED volume measurements. No noteworthy differences were observed for MNV area, CC FDs, MCT, or CVI measurements. Conclusions The onset of exudation was correlated with lesions having less vascularity and smaller PED volume measurements, but measurements of MNV area, CC FDs, MCT, and CVI were not correlated with near-term exudation. Investigations are ongoing to further explore these and other anatomic changes as harbingers of near-term exudation.

To examine retinal vessel closure metrics and neurodegenerative changes occurring in the initial stages of nonproliferative diabetic retinopathy (NPDR) and severity progression in a three-year period. Methods: Three-year prospective longitudinal observational cohort of individuals with type 2 diabetes (T2D), one eye per person, using spectral domain-optical coherence tomography (SD-OCT) and OCT-Angiography (OCTA). Eyes were examined four times with one-year intervals. OCTA vessel density maps of the retina were used to quantify vessel closure. Thickness of the ganglion cell + inner plexiform layer (GCL + IPL) was examined to identify retinal neurodegenerative changes. Diabetic retinopathy ETDRS classification was performed using the seven-field ETDRS protocol. Results: A total of 78 eyes/patients, aged 52 to 80 years, with T2D and ETDRS grades from 10 to 47 were followed for 3 years with annual examinations. A progressive increase in retinal vessel closure was observed. Vessel density (VD) showed higher decreases with retinopathy worsening demonstrated by step-changes in ETDRS severity scale (p &lt, 0.001). No clear correlation was observed between neurodegenerative changes and retinopathy progression. Conclusions: Retinal vessel closure in NPDR correlates with DR severity progression. Our findings provide supporting evidence that OCTA metrics of vessel closure may be used as a surrogate for DR severity progression.

We sought to determine the earliest diagnostic signs of hydroxychloroquine retinopathy up to the point of clinical recognition.Retrospective series of 6 patients (5 parafoveal disease; 1 pericentral disease) with at least 3 examinations over 3.5 years or more preceding diagnosis of HCQ retinopathy. Spectral domain optical coherence tomography (sdOCT) cross-sections, fundus autofluorescence (FAF) and visual fields were generated clinically. Stored sdOCT data were re-examined later to generate topographic ellipsoid zone (EZ) maps, minimum intensity (MI) analysis and sequential plots of regional retinal thickness. Retrospective series of six patients (5 parafoveal disease; one pericentral disease) with at least three examinations over 3.5 years or more preceding diagnosis of hydroxychloroquine retinopathy.Spectral domain optical coherence tomography cross-sections and fields showed similar sensitivity; fundus autofluorescence was not helpful. In parafoveal cases, EZ topography and minimum intensity analysis were no more reliable. Sequential thickness plots from four parafoveal cases showed dramatic retinal thinning across the posterior pole beginning 4 years to 5 years before clinical diagnosis, with parafoveal regions thinning even faster. The pericentral case showed thinning only outside the central macula. Peripheral EZ loss was more dramatic with EZ topography than sdOCT cross-sections.Sequential retinal thickness plots reveal definitive thinning years before current diagnostic procedures. We hope that OCT manufacturers will develop software to display such measurements. Ellipsoid zone topography was not more sensitive than sdOCT cross-sections, but important for recognizing pericentral disease.

We propose a hybrid sequential prediction model called “Deep Sequence”, integrating radiomics-engineered imaging features, demographic, and visual factors, with a recursive neural network (RNN) model in the same platform to predict the risk of exudation within a future time-frame in non-exudative AMD eyes. The proposed model provides scores associated with risk of exudation in the short term (within 3 months) and long term (within 21 months), handling challenges related to variability of OCT scan characteristics and the size of the training cohort. We used a retrospective clinical trial dataset that includes 671 AMD fellow eyes with 13,954 observations before any signs of exudation for training and validation in a tenfold cross validation setting. Deep Sequence achieved high performance for the prediction of exudation within 3 months (0.96 ± 0.02 AUCROC) and within 21 months (0.97 ± 0.02 AUCROC) on cross-validation. Training the proposed model on this clinical trial dataset and testing it on an external real-world clinical dataset showed high performance for the prediction within 3-months (0.82 AUCROC) but a clear decrease in performance for the prediction within 21-months (0.68 AUCROC). While performance differences at longer time intervals may be derived from dataset differences, we believe that the high performance and generalizability achieved in short-term predictions may have a high clinical impact allowing for optimal patient follow-up, adding the possibility of more frequent, detailed screening and tailored treatments for those patients with imminent risk of exudation.

PURPOSE: The correlations between enlargement rates (ERs) of geographic atrophy (GA) and choriocapillaris (CC) flow deficits (FDs), mean choroidal thickness (MCT), and choroidal vascularity index (CVI) were investigated using swept source optical coherence tomography (SS-OCT) in age-related macular degeneration (AMD). DESIGN: Retrospective review of prospective, observational case series. METHODS: Eyes with GA from AMD were imaged with SS-OCT (PLEX(®) Elite 9000; Carl Zeiss Meditec, Inc, Dublin, CA) using 6×6 mm scan pattern. GA lesions were identified and measured using custom en face structural images and annual square root ERs of GA were calculated. At baseline, CC FDs from different regions outside of the GA were measured, and MCT and CVI from the entire scan area were measured. All the measurements were performed using previously published and validated algorithms. RESULTS: Thirty-eight eyes from 27 patients were included. The CC FDs within each region around GA lesions were all highly correlated with ERs of GA (All P < 0.005). CVI inside GA region was correlated with the ERs (P = 0.03), while other choroidal measurements had no significant correlation with the ERs of GA (P > 0.06). CONCLUSION: Statistically significant correlations were found between the ERs of GA and CC FD% from the entire scan region outside the GA and not just the region immediately adjacent to the GA. These results suggest that abnormal CC perfusion throughout the macula contributes to disease progression in eyes with GA. CVI inside GA region could also be a potential indicator for the growth of GA.

Purpose Optical coherence tomography (OCT) cross-sections have shown limited ellipsoid zone (EZ) improvement in mild hydroxychloroquine (HCQ) retinopathy within a few years after drug cessation. However, the extent, functional significance, and stability of such changes over time remain unclear. Methods We created en face EZ maps using automated pixel-by-pixel segmentation for four patients with early-moderate HCQ toxicity followed for 6-8 years after drug cessation. These maps were compared with OCT cross-sections, fundus autofluorescence, and automated 10-2 visual fields. Results One patient had no EZ line loss; one had stable EZ loss throughout follow-up; two showed 30 to 40% reduction in the area of loss, largely in the first 2 years. This limited recovery mostly occurred in regions where the EZ line was only thinned or fragmented; other similar areas did not improve. Fundus autofluorescence hyperfluorescence and visual fields did not show consistent correlation with topography. Conclusion Anatomic EZ recovery, when present, was restricted to regions of mild damage and did not correlate with fundus autofluorescence or improvement in visual fields. Topographic mapping seemed no more sensitive locally than cross-sectional OCT but may aid detection and longitudinal follow-up of toxicity by showing early damage or changes in the macula that could be missed with individual cross-sections.

Purpose To investigate the involvement of the retinal and choriocapillaris microvasculature in intermediate uveitis on optical coherence tomography angiography (OCT-A). Design Case-control study. Methods Patients and age-matched controls were imaged with swept-source OCT-A. Using ImageJ, superficial and deep retinal vasculature were semi-automatically analyzed for vessel (VD) and skeleton density (SD), vessel diameter index (VDI), and fractal dimension (FD). Choriocapillaris layer was automatically graded for mean signal intensity, signal intensity standard deviation, kurtosis of signal intensity distribution, and flow signal voids. Results Twenty-nine intermediate uveitis eyes and 30 control eyes were included. Both superficial and deep retinal layers showed significant reduction in all OCT-A parameters (eg, superficial retinal layer: 0.31 vs 0.40 VD, 5.6e−8 vs 6.4e−8 SD, 5.4e6 vs 6.1e6 VDI, and 1.78 vs 1.79 FD, respectively, all P Conclusions In intermediate uveitis, reduced vascular density and complexity in superficial as well as deep retinal layers and altered choriocapillaris perfusion are present. Moreover, these findings indicate impairment of the macular microvasculature even in the absence of macular edema. The results of our study may aid in the diagnosis as well as the monitoring of intermediate uveitis.

Background To estimate choriocapillaris flow deficits beyond normal intercapillary distance with swept source optical coherence tomography angiography (SS-OCTA). Methods Subjects were enrolled and repeated SS-OCTA scans were performed using the 3 mm × 3 mm scan pattern. Blood flow was identified using the complex optical microangiography (OMAGc) algorithm. The choriocapillaris (CC) was defined as a 20 µm slab of the flow volume beneath the outer boundary of Bruch's membrane (BM) and was compensated with the corresponding structural image for flow deficits measurement. Flow deficits were segmented based on one mean standard deviation from a normal database. A histogram based thresholding method was developed to remove small flow deficits that were determined by examining intercapillary spacing within normal CC networks. A registration method based on affine and B-spline transformation was utilized for the CC angiogram averaging. Four repeated scans were averaged, and results were compared with and without removal of small flow deficits after averaging a different number of scans (N=1, group 1; N=2, group 2; N=3, group 3 and N=4, group 4). Results Seven normal subjects were enrolled. Intercapillary distance was found to be 24 µm for the CC networks under OCTA, which was used as the threshold to exclude small flow deficits for CC quantification. After averaging, significant reduction in background noise and improvement in continuity of blood vessel networks were observed both on retinal and choriocapillaris angiograms. Flow deficit percentages of the choriocapillaris were significantly reduced with averaging (group 1 vs. group 2: P

Purpose Swept-source optical coherence tomography angiography was used to investigate choroidal changes and their association with pigment epithelial detachments (PEDs) in eyes with polypoidal choroidal vasculopathy (PCV) after treatment with vascular endothelial growth factor (VEGF) inhibitors. Methods Patients with treatment-naïve PCV were included and underwent anti-VEGF therapy. Mean choroidal thickness (MCT), choroidal vascularity index (CVI), and PED volume measurements were obtained before and after treatment. Results Thirty-four treatment-naïve PCV eyes from 33 patients were included. The PED volume decreased after treatment (P < 0.05). The MCT decreased from 223.0 ± 79.6 µm at baseline to 210.9 ± 76.2 µm after treatment (P < 0.001). The CVI at baseline was 0.599 ± 0.024, and the CVI after treatment was 0.602 ± 0.023 (P = 0.16). There was a correlation between the decreased PED volumes and the decreased MCT measurements (r = 0.47; P = 0.006). Also, there was a correlation between the decreased PED volumes and the increased CVI measurements (r = −0.63; P < 0.001). Conclusions In treatment-naïve eyes with PCV, the decreases in PED volumes were correlated with the decrease in MCT and the increase in CVI measurements. We propose that, at baseline, the PCV lesions serve as high-volume arteriovenous shunts between choroidal arterial and venous circulation, causing transudation into the choroidal stroma. We propose that, after treatment, the blood flow through the vascular shunt is reduced, the excess stromal transudation is resorbed, and the exudation from the neovascular lesion is reduced, resulting in thinning of the choroid, resolution of the PEDs, and an increase in the CVI due to the resorption of excess choroidal transudation.

Purpose Swept-source (SS) OCT angiography (OCTA) was used to determine the prevalence, incidence, and natural history of subclinical macular neovascularization (MNV) in eyes with nonexudative age-related macular degeneration (AMD). Design Prospective, observational, consecutive case series. Participants Patients with intermediate AMD (iAMD) or geographic atrophy (GA) secondary to nonexudative AMD in 1 eye and exudative AMD in the fellow eye. Methods All patients were imaged using both the 3×3 mm and 6×6 mm SS OCTA fields of view (PLEX Elite 9000; Carl Zeiss Meditec, Inc, Dublin, CA). The en face slab used to detect the MNV extended from the outer retina to the choriocapillaris, and projection artifacts were removed using a proprietary algorithm. Main Outcome Measures Prevalence of subclinical MNV and time to exudation with Kaplan-Meier cumulative estimates of exudation at 1 year. Results From August 2014 through March 2017, 160 patients underwent SS OCTA (110 eyes with iAMD and 50 eyes with GA). Swept-source OCTA identified subclinical MNV at the time of first imaging in 23 of 160 eyes, for a prevalence of 14.4%. Six eyes demonstrated subclinical MNV during the follow-up. Of 134 eyes with follow-up visits, a total of 13 eyes demonstrated exudation, and of these 13 eyes, 10 eyes were found to have pre-existing subclinical MNV. By 12 months, the Kaplan-Meier cumulative incidence of exudation for all 134 eyes was 6.8%. For eyes with subclinical MNV at the time of first SS OCTA imaging, the incidence was 21.1%, and for eyes without subclinical MNV, the incidence was 3.6%. There was no difference in the cumulative incidence of exudation from pre-existing MNV in eyes with iAMD or GA ( P = 0.847, log-rank test). After the detection of subclinical MNV, the risk of exudation was 15.2 times (95% confidence interval, 4.2–55.4) greater compared with eyes without subclinical MNV. Conclusions By 12 months, the risk of exudation was greater for eyes with documented subclinical MNV compared with eyes without detectable MNV. For eyes with subclinical MNV, recommendations include more frequent follow-up and home monitoring. Intravitreal therapy is not recommended until prospective studies are performed.

This study explored whether rotational three-dimensional (3D) visualization of optical coherence tomography angiography (OCTA) volume data may yield valuable information regarding type 3 macular neovascularization (MNV). In this retrospective, cross-sectional study, we collected data from 15 eyes (13 patients) with treatment-naïve type 3 MNV in their post-nascent stage and age-related macular degeneration (AMD). Subjects were imaged with the SS-OCT system (PLEX Elite 9000, Carl Zeiss Meditec Inc., Dublin, CA, USA). The OCTA volume data were processed with a prototype volume projection removal algorithm and then analyzed using volumetric visualization techniques in order to obtain a 3D visualization of the region occupied by type 3 MNV. The two-dimensional and three-dimensional OCTA images were investigated. Mean ± SD age was 75.1 ± 7.4 years. BCVA was 0.42 ± 0.21 LogMAR in the study eyes. Considering the cohort of analyzed eyes, on rotational 3D OCTA images, a total of 35 neovascular lesions (vs 22 lesions detected on 2D OCTA images) rising from the deep vascular complex and variably spanning the outer retinal layers and eventually reaching the RPE/sub-RPE space were detected. Nine of 35 lesions had a saccular shape, while the remaining cases had a filiform shape. On rotational 3D OCTA images, these lesions were inclined on the three planes, instead of perpendicular to the RPE/Bruch’s membrane. In conclusion, this study used an algorithm to obtain rotational three-dimensional visualization of type 3 MNV. This approach seems to increase the detection rate for these lesions and to be useful to offer new insight into type 3 MNV.

Purpose Choriocapillaris (CC) imaging of normal eyes with swept-source optical coherence tomographic angiography (SS-OCTA) was performed, and the percentage of CC flow deficits (FD%) and the average area of CC flow deficits (FDa) were compared within the given macular regions. Design A prospective, cross-sectional study. Methods Subjects with normal eyes ranging in age from their 20s through their 80s were imaged with SS-OCTA (PLEX Elite 9000; Carl Zeiss Meditec, Dublin, California, USA) using both 3×3-mm and 6×6-mm macular scan patterns. The CC images were generated using a previously published and validated algorithm. In both 3×3-mm and 6×6-mm scans, the CC FD% and FDa were measured in circular regions centered on the fovea with diameters as 1 mm and 2.5 mm (C1 and C2.5). In 6×6-mm scans, the FD% and FDa were measured within an additional circular region with diameter as 5 mm (C5). The correlations between FD% and FDa from each region were analyzed with Pearson correlation coefficients. Results A total of 164 eyes were analyzed. There was excellent correlation between CC FDa and FD% measurements from each region. In the 3×3-mm scans, the correlations in the C1 and C2.5 regions were 0.83 and 0.90, respectively. In the 6×6-mm scans, the correlations in C1, C2.5, and C5 regions were 0.90, 0.89, and 0.89, respectively. Conclusions When measuring CC FDs, we found excellent correlations between FDa and FD% in regions from 3×3-mm and 6×6-mm scans. Further studies are needed to determine if one parameter is more useful when studying diseased eyes.

Purpose To probabilistically forecast needed anti-vascular endothelial growth factor (anti-VEGF) treatment frequency using volumetric spectral domain-optical coherence tomography (SD-OCT) biomarkers in neovascular age-related macular degeneration from real-world settings. Methods SD-OCT volume scans were segmented with a custom deep-learning-based analysis pipeline. Retinal thickness and reflectivity values were extracted for the central and the four inner Early Treatment Diabetic Retinopathy Study (ETDRS) subfields for six retinal layers (inner retina, outer nuclear layer, inner segments [IS], outer segments [OS], retinal pigment epithelium-drusen complex [RPEDC] and the choroid). Machine-learning models were probed to predict the anti-VEGF treatment frequency within the next 12 months. Probabilistic forecasting was performed using natural gradient boosting (NGBoost), which outputs a full probability distribution. The mean absolute error (MAE) between the predicted versus actual anti-VEGF treatment frequency was the primary outcome measure. Results In a total of 138 visits of 99 eyes with neovascular AMD (96 patients) from two clinical centers, the prediction of future anti-VEGF treatment frequency was observed with an accuracy (MAE [95% confidence interval]) of 2.60 injections/year [2.25-2.96] (R2 = 0.390) using random forest regression and 2.66 injections/year [2.31-3.01] (R2 = 0.094) using NGBoost, respectively. Prediction intervals were well calibrated and reflected the true uncertainty of NGBoost-based predictions. Standard deviation of RPEDC-thickness in the central ETDRS-subfield constituted an important predictor across models. Conclusions The proposed, fully automated pipeline enables probabilistic forecasting of future anti-VEGF treatment frequency in real-world settings. Translational relevance Prediction of a probability distribution allows the physician to inspect the underlying uncertainty. Predictive uncertainty estimates are essential to highlight cases where human-inspection and/or reversion to a fallback alternative is warranted.

Importance Sensitive outcome measures for disease progression are needed for treatment trials in geographic atrophy (GA) secondary to age-related macular degeneration (AMD). Objective To quantify photoreceptor degeneration outside regions of GA in eyes with nonexudative AMD, to evaluate its association with future GA progression, and to characterize its spatio-temporal progression. Design, Setting, and Participants Monocenter cohort study (Directional Spread in Geographic Atrophy [NCT02051998]) and analysis of data from a normative data study at a tertiary referral center. One hundred fifty-eight eyes of 89 patients with a mean (SD) age of 77.7 (7.1) years, median area of GA of 8.87 mm2(IQR, 4.09-15.60), and median follow-up of 1.1 years (IQR, 0.52-1.7 years), as well as 93 normal eyes from 93 participants. Exposures Longitudinal spectral-domain optical coherence tomography (SD-OCT) volume scans (121 B-scans across 30° × 25°) were segmented with a deep-learning pipeline and standardized in a pointwise manner with age-adjusted normal data (z scores). Outer nuclear layer (ONL), photoreceptor inner segment (IS), and outer segment (OS) thickness were quantified along evenly spaced contour lines surrounding GA lesions. Linear mixed models were applied to assess the association between photoreceptor-related imaging features and GA progression rates and characterize the pattern of photoreceptor degeneration over time. Main Outcomes and Measures Association of ONL thinning with follow-up time (after adjusting for age, retinal topography [zscore], and distance to the GA boundary). Results The study included 158 eyes of 89 patients (51 women and 38 men) with a mean (SD) age of 77.7 (7.1) years. The fully automated B-scan segmentation was accurate (dice coefficient, 0.82; 95% CI, 0.80-0.85; compared with manual markings) and revealed a marked interpatient variability in photoreceptor degeneration. The ellipsoid zone (EZ) loss-to-GA boundary distance and OS thickness were prognostic for future progression rates. Outer nuclear layer and IS thinning over time was significant even when adjusting for age and proximity to the GA boundary (estimates of −0.16 μm/y; 95% CI, −0.30 to −0.02; and −0.17 μm/y; 95% CI, −0.26 to −0.09). Conclusions and Relevance Distinct and progressive alterations of photoreceptor laminae (exceeding GA spatially) were detectable and quantifiable. The degree of photoreceptor degeneration outside of regions of retinal pigment epithelium atrophy varied markedly between eyes and was associated with future GA progression. Macula-wide photoreceptor laminae thinning represents a potential candidate end point to monitor treatment effects beyond mere GA lesion size progression.

PURPOSE: Swept source optical coherence tomography angiography (SS-OCTA) was used to study the prevalence, incidence, and natural history of subclinical macular neovascularization (MNV) in eyes with unilateral non-exudative age-related macular degeneration (AMD). DESIGN: Prospective cohort study. METHODS: Patients were imaged using 3×3 mm and 6×6 mm SS-OCTA scan patterns. MNV was detected using the outer retina to choriocapillaris en face slab. Prevalence and incidence of subclinical MNV, Kaplan-Meier cumulative estimates for the overall risk of exudation, and the association between neovascular lesion size and the risk of exudation were assessed through 2 years. RESULTS: From August 2014 through March 2018, 227 patients (154 intermediate and 73 late AMD eyes) underwent SS-OCTA imaging. Thirty eyes (13.2%) had subclinical MNV at first imaging and 12 eyes (8.9%) developed subclinical MNV during follow-up. Of the 191 eyes with more than one visit, 19 developed exudation. Fourteen of these eyes had pre-existing subclinical MNV. The incidence of exudation from the time of first detection of any subclinical MNV was 34.5%. The relative risk of exudation after detection of subclinical MNV was 13.6 times greater (95% confidence interval, 4.9–37.7) than in the absence of subclinical MNV (P

Age-related macular degeneration (AMD) is a major cause of vision loss for elderly people. Drusen and geographic atrophy are two kinds of pathological changes associated with AMD. This chapter presents several methods that achieve abnormal region segmentation for quantitative analysis or enhanced visualization for better qualitative assessment.

PURPOSE: Swept source optical coherence tomography angiography (SS-OCTA) was used to measure the age-dependent changes in macular choriocapillaris (CC) flow deficits (FDs) in normal eyes. DESIGN: A prospective, cross-sectional study. METHODS: Subjects with normal eyes ranging in age from their 20s to their 80s were imaged using a 100-kHz SS-OCTA instrument (PLEX(®) Elite 9000, Carl Zeiss Meditec). Both 3×3 mm and 6×6 mm scans were used to image the macular CC. Visualization of the CC and quantification of FDs were performed using a previously validated algorithm. The percentage of FDs (FD%) in the central 1 mm circle (C(1)), 1.5 mm rim (R(1.5)), and 2.5 mm circle (C(2.5)) from the 3×3 mm and 6×6 mm scans and FD% in the 2.5 mm rim (R(2.5)) and 5 mm circle (C(5)) from the 6×6 mm scans were measured and correlated with age and axial length. RESULTS: A total of 164 subjects were enrolled with at least 10 subjects from each decade of life. No meaningful correlations were found between FD% and axial length (∣r∣ < 0.30). FD% in all fields increased with increasing age (all r > 0.50; all P < 0.001); however, the greatest increases were found in the central macula C(1) regions and the smallest increases in the peripheral macula R(2.5) regions. CONCLUSIONS: In normal aging, the FD% increased with age across the central 5 mm of the macula, but the greatest increase was found in the central 1 mm region of the macula.

Purpose: To investigate the choroidal blood flow in areas within and adjacent to retinal pigment epithelium (RPE) atrophy secondary to late-onset Stargardt disease (STGD1) and age-related macular degeneration (AMD). Methods: A total of 43 eyes (23 STGD1 and 20 AMD) of patients with RPE atrophy and 25 eyes of healthy controls without ocular pathology underwent multimodal imaging including optical coherence tomography angiography (OCT-A; PLEX Elite 9000 Swept-Source OCT). Using an exploratory approach, choriocapillaris and deeper choroid OCT-A slabs were evaluated in order to detect differences between STGD1 and AMD. The magnitude of absence-of-flow signal (AFS) was investigated in terms of area-fraction and size-frequency distribution. Results: Qualitative and quantitative analysis of areas of RPE atrophy revealed more pronounced rarefaction of the choriocapillaris flow signal in STGD1 as compared to AMD (AFS area fraction: 33.15% ± 6.86% vs. 31.68% ± 8.39%; P = 0.517), while outside RPE atrophy rarefaction was less pronounced in STGD1 (AFS area fraction: 17.41% ± 5.67% vs. 21.59% ± 6.90%; P < 0.001), to the level of nonsignificance compared to controls (13.27% ± 2.99%, P = 0.368). Given this discrepancy, the ratio of the AFS area fraction within/outside of RPE atrophy could be used to differentiate between STGD1 and AMD with 65.0% sensitivity and 92.3% specificity. Conclusions: Using OCT-A, comparison of choroidal flow signal within and outside the area of RPE atrophy revealed distinct differences between STGD1 and AMD, potentially implicating a differential role of the choroid in the pathogenesis of RPE atrophy in these two diseases.

Purpose To achieve reproducible imaging of the choriocapillaris and associated flow voids using swept-source OCT angiography (SS-OCTA). Methods Subjects were enrolled and SS-OCTA was performed using the 3 × 3 mm scan pattern. Blood flow was identified using the complex optical microangiography (OMAG) algorithm. The choriocapillaris was defined as a slab from the outer boundary of Bruch's membrane (BM) to approximately 20 μm below BM. Compensation for the shadowing effect caused by the RPE and BM complex on the choriocapillaris angiogram was achieved by using the structural information from the same slab. A thresholding method to calculate the percentage of flow voids from a region was developed based on a normal database. Results Twenty normal subjects and 12 subjects with drusen were enrolled. SS-OCTA identified the choriocapillaris in normal subjects as a lobular plexus of capillaries in the central macula and the lobular arrangement became more evident toward the periphery. In all eyes, signal compensation resulted in fewer choriocapillaris flow voids with improved repeatability of measurements. The best repeatability for the measurement was achieved by using 1 standard deviation (SD) for the thresholding strategy. Conclusions SS-OCTA can image the choriocapillaris in vivo, and the repeatability of flow void measurements is high in the presence of drusen. The ability to image the choriocapillaris and associated flow voids should prove useful in understanding disease onset, progression, and response to therapies.

Importance Risk factors associated with the development of neovascular age-related macular degeneration (AMD) have been identified. However, population size and methods to integrate imaging, genetic, and demographic factors associated with conversion to neovascular AMD are limited, specifically when treatment is administered in 1 eye. Objective To determine the imaging, genetic, and demographic factors associated with conversion from nonneovascular to neovascular AMD in fellow eyes. Design, Setting, and Participants This post hoc secondary analysis of the 24-month phase 3 multicenter, double-masked, active treatment-controlled HARBOR trial included 686 fellow eyes with nonneovascular AMD at baseline. Imaging features describing the presence, number, extent, density, and relative reflectivity of drusen were automatically extracted from spectral-domain optical coherence tomography scans. Genetic analysis included 34 single-nucleotide polymorphisms. Least absolute shrinkage and selection operator regression was performed to narrow imaging features. Survival analysis and Cox proportional hazards regression were performed to determine the association of the selected imaging features and genetic and demographic factors with conversion to neovascular AMD. Data were collected from November 2016 through October 2017 and analyzed from October 2017 through October 2018. Exposure Nonneovascular AMD in the fellow eye. Main Outcomes and Measures Features associated with conversion to neovascular AMD. Hazard ratios (HRs) and their 95% CIs were calculated. Results Among the 686 fellow eyes included in the analysis (406 [59.2%] women; mean [SD] age, 78.12 [8.28] years), 154 (22.4%) converted to neovascular AMD. Female sex was significantly associated with conversion to neovascular AMD (HR, 1.57; 95% CI, 1.11-2.20; P = .009). After controlling for demographic and treatment effects, drusen area within 3 mm of the fovea (HR, 1.45; 95% CI, 1.24-1.69; HR for 1-SD increase, 1.36 [95% CI, 1.20-1.54]) and mean drusen reflectivity (HR, 3.97; 95% CI, 1.11-14.18; HR for 1-SD increase, 1.32 [95% CI, 1.02-1.71]) were significantly associated with conversion to neovascular AMD. In addition, 1 genetic variant (rs61941274) was found to be associated with conversion to neovascular AMD. Conclusions and Relevance Two imaging features (total en face area of drusen restricted to a circular area 3 mm from the fovea and mean drusen reflectivity) and 1 genetic variant ( ACAD10 locus) were associated with conversion to neovascular AMD. Drusen characteristics may be associated with conversion to neovascular AMD despite treatment in 1 eye. Trial Registration ClinicalTrials.gov identifier:NCT00891735

Purpose To develop an automated method based on saliency map of the retinal thickness map to determine foveal center in spectral-domain optical coherence tomography (SD-OCT) images. Methods This paper proposes an automatic method for the detection of the foveal center in SD-OCT images. Initially, a retinal thickness map is generated by considering the axial distance between the internal limiting membrane (ILM) and the Bruch's membrane (BM). Both the ILM and BM boundaries are automatically segmented by a known retinal segmentation technique. The macular foveal region is identified as a salient feature in the retinal thickness map, and segmented by the saliency detection method based on a human vision attention model. Finally, the foveal center is identified by searching for the lowest point from the determined macular fovea region. Results Experimental results in 39 scans from 35 healthy eyes and 58 scans from 29 eyes diagnosed with several stages of age-related macular degeneration (AMD), from mild or intermediate stages to severe dry or wet stages, demonstrated that the proposed method achieves good performance. The mean radial distance error of the automatically detected foveal center locations when compared to consensus manual determination established by repeated sessions from two expert readers was 52 ± 56μm for the normal eyes and 73 ± 63μm for AMD eyes. Conclusions The proposed algorithm was more effective for detecting the foveal center automatically in SD-OCT images than the state-of-art methods. This article is protected by copyright. All rights reserved.

This work introduces and evaluates an automated intra-retinal segmentation method for spectral-domain optical coherence (SD-OCT) retinal images. While quantitative assessment of retinal features in SD-OCT data is important, manual segmentation is extremely time-consuming and subjective. We address challenges that have hindered prior automated methods, including poor performance with diseased retinas relative to healthy retinas, and data smoothing that obscures image features such as small retinal drusen. Our novel segmentation approach is based on the iterative adaptation of a weighted median process, wherein a three-dimensional weighting function is defined according to image intensity and gradient properties, and a set of smoothness constraints and pre-defined rules are considered. We compared the segmentation results for 9 segmented outlines associated with intra-retinal boundaries to those drawn by hand by two retinal specialists and to those produced by an independent state-of-the-art automated software tool in a set of 42 clinical images (from 14 patients). These images were obtained with a Zeiss Cirrus SD-OCT system, including healthy, early or intermediate AMD, and advanced AMD eyes. As a qualitative evaluation of accuracy, a highly experienced third independent reader blindly rated the quality of the outlines produced by each method. The accuracy and image detail of our method was superior in healthy and early or intermediate AMD eyes (98.15% and 97.78% of results not needing substantial editing) to the automated method we compared against. While the performance was not as good in advanced AMD (68.89%), it was still better than the manual outlines or the comparison method (which failed in such cases). We also tested our method's performance on images acquired with a different SD-OCT manufacturer, collected from a large publicly available data set (114 healthy and 255 AMD eyes), and compared the data quantitatively to reference standard markings of the internal limiting membrane and inner boundary of retinal pigment epithelium, producing a mean unsigned positioning error of 6.04 ± 7.83µm (mean under 2 pixels). Our automated method should be applicable to data from different OCT manufacturers and offers detailed layer segmentations in healthy and AMD eyes.

Importance Retinopathy is a known risk of long-term use of hydroxychloroquine sulfate. However, whether the inner as well as outer retina are involved before retinopathy develops and whether changes in the retina might signal impending toxic effects during screening remain unknown. Objective To determine the degree of inner and outer retinal involvement in short- and long-term use of hydroxychloroquine before the development of retinopathy. Design, Setting, and Participants This retrospective medical record review of spectral-domain optical coherence tomography (SD-OCT) findings was performed at an academic medical center. Thirty-two patients without retinopathy and with high-quality SD-OCT images were studied. Twenty-seven patients were age matched (49-65 years old) for comparison of retinal layers among patients who used the drug less than 5 years (n = 12) or longer than 15 years (n = 15) at the initial visit. Populations were also defined (without age limitation) for comparison of change during 25 to 52 months of follow-up among patients with initial use of less than 5 years (n = 7) or longer than 15 years (n = 8). Data were collected from 2010 to 2015. Main Outcomes and Measures Measurements of inner and outer retinal thickness in SD-OCT images using commercial software and a Stanford pixel-by-pixel segmentation software that also provided topographic maps of thickness dimensions and change. Results Thirty-two patients (5 men and 27 women) were included in the analysis. Measurements of inner retinal thickness between short- and long-term hydroxychloroquine users (n = 27) in different retinal regions, and during a median 39 months of follow-up (n = 15), showed no statistically significant differences or change. Similarly, no significant differences or changes were identified in outer retinal thickness except for the final visit of 1 patient who developed focal parafoveal thinning, a toxic effect of hydroxychloroquine use. Cirrus ganglion cell analysis measurements were inaccurate in the presence of outer retinal damage. Conclusions and Relevance The inner retina appears not to be involved in hydroxychloroquine-induced retinopathy to any clinically relevant degree within the limitations of our sample size. No clinically apparent warning of outer retinal damage was seen in the SD-OCT images of long-term hydroxychloroquine users until the actual appearance of focal retinopathy. Early detection of hydroxychloroquine-induced retinopathy is known to prevent visual acuity loss and serious progression after the therapy is stopped, and these data suggest that screening should seek distinct new areas of retinopathy (shown by topographic thickness maps) rather than long-term progressive thinning.

Purpose To develop a predictive model based on quantitative characteristics of geographic atrophy (GA) to estimate future potential regions of GA growth. Design Progression study and predictive model. Participants One hundred eighteen spectral-domain (SD) optical coherence tomography (OCT) scans of 38 eyes in 29 patients. Methods Imaging features of GA quantifying its extent and location, as well as characteristics at each topographic location related to individual retinal layer thickness and reflectivity, the presence of pathologic features (like reticular pseudodrusen or loss of photoreceptors), and other known risk factors of GA growth, were extracted automatically from 118 SD OCT scans of 38 eyes from 29 patients collected over a median follow-up of 2.25 years. We developed and evaluated a model to predict the magnitude and location of GA growth at given future times using the quantitative features as predictors in 3 possible scenarios. Main Outcome Measures Potential regions of GA growth. Results In descending order of out-of-bag feature importance, the most predictive SD OCT biomarkers for predicting the future regions of GA growth were thickness loss of bands 11 through 14 (5.66), reflectivity of bands 11 and 12 (5.37), thickness of reticular pseudodrusen (5.01), thickness of bands 5 through 11 (4.82), reflectivity of bands 7 through 11 (4.78), GA projection image (4.73), increased minimum retinal intensity map (4.59), and GA eccentricity (4.49). The predicted GA regions in the 3 tested scenarios resulted in a Dice index mean ± standard deviation of 0.81±0.12, 0.84±0.10, and 0.87±0.06, respectively, when compared with the observed ground truth. Considering only the regions without evidence of GA at baseline, predicted regions of future GA growth showed relatively high Dice indices of 0.72±0.18, 0.74±0.17, and 0.72±0.22, respectively. Predictions and actual values of GA growth rate and future GA involvement in the central fovea showed high correlations. Conclusions Experimental results demonstrated the potential of our predictive model to predict future regions where GA is likely to grow and to identify the most discriminant early indicator (thickness loss of bands 11 through 14) of regions susceptible to GA growth.

Age-related macular degeneration (AMD) is the leading cause of blindness among elderly individuals. Geographic atrophy (GA) is a phenotypic manifestation of the advanced stages of non-exudative AMD. Determination of GA extent in SD-OCT scans allows the quantification of GA-related features, such as radius or area, which could be of important value to monitor AMD progression and possibly identify regions of future GA involvement. The purpose of this work is to develop an automated algorithm to segment GA regions in SD-OCT images. An en face GA fundus image is generated by averaging the axial intensity within an automatically detected sub-volume of the three dimensional SD-OCT data, where an initial coarse GA region is estimated by an iterative threshold segmentation method and an intensity profile set, and subsequently refined by a region-based Chan-Vese model with a local similarity factor. Two image data sets, consisting on 55 SD-OCT scans from twelve eyes in eight patients with GA and 56 SD-OCT scans from 56 eyes in 56 patients with GA, respectively, were utilized to quantitatively evaluate the automated segmentation algorithm. We compared results obtained by the proposed algorithm, manual segmentation by graders, a previously proposed method, and experimental commercial software. When compared to a manually determined gold standard, our algorithm presented a mean overlap ratio (OR) of 81.86% and 70% for the first and second data sets, respectively, while the previously proposed method OR was 72.60% and 65.88% for the first and second data sets, respectively, and the experimental commercial software OR was 62.40% for the second data set.

To enhance the rapid assessment of geographic atrophy (GA) across the macula in a single projection image generated from three-dimensional (3D) spectral-domain optical coherence tomography (SD-OCT) scans by introducing a novel restricted summed-area projection (RSAP) technique.We describe a novel en face GA visualization technique, the RSAP, by restricting the axial projection of SD-OCT images to the regions beneath the Bruch's membrane (BM) boundary and also considering the choroidal vasculature's influence on GA visualization. The technique analyzes the intensity distribution beneath the retinal pigment epithelium (RPE) layer to fit a cross-sectional surface in the sub-RPE region. The area is taken as the primary GA projection. A median filter is then adopted to smooth the generated GA projection image. The RSAP technique was evaluated in 99 3D SD-OCT data sets from 27 eyes of 21 patients presenting with advanced nonexudative age-related macular degeneration and GA. We used the mean difference between GA and background regions and GA separability metric to measure GA contrast and distinction in the generated images, respectively. We compared our results with two existing GA projection techniques, the summed-voxel projection (SVP) and Sub-RPE Slab techniques.Comparative results demonstrate that the RSAP technique is more effective in displaying GA than the SVP and Sub-RPE Slab. The average of the mean difference between GA and background regions and the GA separability based on SVP, Sub-RPE Slab, and RSAP were 0.129/0.880, 0.238/0.919, and 0.276/0.938, respectively.The RSAP technique was more effective for GA visualization than the conventional SVP and Sub-RPE Slab techniques. Our technique decreases choroidal vasculature influence on GA projection images by analyzing the intensity distribution characteristics in sub-RPE regions. The generated GA projection image with the RSAP technique has improved contrast and distinction.Our method for automated generation of GA projection images from SD-OCT images may improve the visualization of the macular abnormalities and the management of GA.