Application of Automated Quantification of Fluid Volumes to Anti–VEGF Therapy of Neovascular Age-Related Macular Degeneration

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

OBJECTIVE

Health care systems worldwide are challenged to provide adequate care for the 200 million individuals with age-related macular degeneration (AMD). Artificial intelligence (AI) has the potential to make a significant, positive impact on the diagnosis and management of patients with AMD; however, the development of effective AI devices for clinical care faces numerous considerations and challenges, a fact evidenced by a current absence of Food and Drug Administration (FDA)-approved AI devices for AMD.

PURPOSE

To delineate the state of AI for AMD, including current data, standards, achievements, and challenges.

METHODS

Members of the Collaborative Community on Ophthalmic Imaging Working Group for AI in AMD attended an inaugural meeting on September 7, 2020, to discuss the topic. Subsequently, they undertook a comprehensive review of the medical literature relevant to the topic. Members engaged in meetings and discussion through December 2021 to synthesize the information and arrive at a consensus.

RESULTS

Existing infrastructure for robust AI development for AMD includes several large, labeled data sets of color fundus photography and OCT images; however, image data often do not contain the metadata necessary for the development of reliable, valid, and generalizable models. Data sharing for AMD model development is made difficult by restrictions on data privacy and security, although potential solutions are under investigation. Computing resources may be adequate for current applications, but knowledge of machine learning development may be scarce in many clinical ophthalmology settings. Despite these challenges, researchers have produced promising AI models for AMD for screening, diagnosis, prediction, and monitoring. Future goals include defining benchmarks to facilitate regulatory authorization and subsequent clinical setting generalization.

CONCLUSIONS

Delivering an FDA-authorized, AI-based device for clinical care in AMD involves numerous considerations, including the identification of an appropriate clinical application; acquisition and development of a large, high-quality data set; development of the AI architecture; training and validation of the model; and functional interactions between the model output and clinical end user. The research efforts undertaken to date represent starting points for the medical devices that eventually will benefit providers, health care systems, and patients.

[Artificial intelligence in the management of anti-VEGF treatment: the Vienna fluid monitor in clinical practice]

Der Vienna Fluid Monitor ist ein künstlicher Intelligenz(KI)-Algorithmus zur präzisen Lokalisation und Quantifizierung von retinaler Flüssigkeit. Der Algorithmus soll Klinikern und Klinikerinnen helfen, objektive und genaue Behandlungsentscheidungen bei der antivaskulären endothelialen Wachstumsfaktor(Anti-VEGF)-Therapie von Patienten mit neovaskulärer altersbedingter Makuladegeneration zu treffen. Ziel der Implementierung ist die Optimierung der Patientensicherheit, die Erhaltung der Sehleistung und gleichzeitig die Behandlungslast für das Gesundheitssystem und die Patienten zu verringern.

Artificial Intelligence and Imaging Processing in Optical Coherence Tomography and Digital Images in Uveitis

INTRODUCTION

Computer vision, understood as the area of science that trains computers to interpret digital images through both artificial intelligence (AI) and classical algorithms, has significantly advanced the analysis and interpretation of optical coherence tomography (OCT) in retina research. The aim of this review is to summarise the recent advances of computer vision in imaging processing in uveitis, with a particular focus in optical coherence tomography images.

MATERIAL AND METHODS

Literature review.

RESULTS

The development of computer vision to assist uveitis diagnosis and prognosis is still undergoing, but important efforts have been made in the field.

CONCLUSION

The automatising of image processing in uveitis could be fundamental to establish objective and standardised outcomes for future clinical trials. In addition, it could help to better understand the disease and its progression.

IMPACT OF FLUID COMPARTMENTS ON FUNCTIONAL OUTCOMES FOR PATIENTS WITH NEOVASCULAR AGE-RELATED MACULAR DEGENERATION: A Systematic Literature Review

PURPOSE

Understanding the impact of fluid in different retinal compartments is critical to developing treatment paradigms that optimize visual acuity and reduce treatment burden in neovascular age-related macular degeneration. This systematic review aimed to determine the impact of persistent/new subretinal fluid, intraretinal fluid, and subretinal pigment epithelial fluid on visual acuity over 1 year of treatment.

METHODS

Publication eligibility and data extraction were conducted according to Cochrane methods: 27 of the 1,797 screened records were eligible.

RESULTS

Intraretinal fluid negatively affected visual acuity at baseline and throughout treatment, with foveal intraretinal fluid associated with lower visual acuity than extrafoveal intraretinal fluid. Some studies found that subretinal fluid (particularly subfoveal) was associated with higher visual acuity at Year 1 and longer term, and others suggested subretinal fluid did not affect visual acuity at Years 1 and 2. Data on the effects of subretinal pigment epithelial fluid were scarce, and consensus was not reached. Few studies reported numbers of injections associated with fluid status.

CONCLUSION

To optimally manage neovascular age-related macular degeneration, clinicians should understand the impact of fluid compartments on visual acuity. After initial treatment, antivascular endothelial growth factor regimens that tolerate stable subretinal fluid (if visual acuity is stable/improved) but not intraretinal fluid may enable patients to achieve their best possible visual acuity. Confirmatory studies are required to validate these findings.

The Quantification and Impact of Persistent Retinal Fluid Compartments on Best-Corrected Visual Acuity of Patients With Retinal Vein Occlusion

BACKGROUND AND OBJECTIVES

To evaluate the impact of persistent intraretinal fluid (IRF) and subretinal fluid (SRF) on best visual acuity (BVA) of patients with retinal vein occlusions (RVOs).

PATIENTS AND METHODS

This retrospective cohort study observed 92 treatment-naïve patients with RVO during 12 months of treatment with anti-vascular endothelial growth factor agents. Deep learning was used to quantify IRF and SRF volumes, and linear mixed effects regression modeled the impact on BVA.

RESULTS

Average IRF volume declined -923.1 ± 2,382.5 nL from baseline to 12 months (P < .001). Average SRF volume declined -35.4 ± 223.4 nL from baseline to 12 months (P = .139). linear mixed effects regression modeling disclosed IRF≥ 1,616 nL at all time points predicted a -10.38 letter loss at 12 months (95% CI, -14.58 to -5.9 letters; P < .001). A similar relationship was not found for SRF.

CONCLUSION

Persistent IRF may be an important prognostic biomarker for BVA outcomes in real-world patients with RVO. [Ophthalmic Surg Lasers Imaging Retina. 2022;53:139-147.].

A Delphi study on the clinical management of age-related macular degeneration

PURPOSE

Age-related macular degeneration (AMD) is one of the main causes of blindness and visual impairment worldwide. As achieving a dry macula is one of the main objectives in AMD management, the purpose of this work was to reach a consensus on the relevance of retinal fluid in function, disease activity control and treatment patterns.

METHODS

Forty-seven Portuguese ophthalmologists specialized in AMD participated in a DELPHI panel. Two rounds of presential meetings were conducted and a cut-off of 80% or more of votes was defined to consider answers consensual.

RESULTS

Consensus was reached for 11 out of 18 questions. These questions focused on the impact of anatomical results on visual acuity, standards exams and parameters to assess disease activity, frequency and factors which influence disease activity assessment, criteria to use non-fixed treatment regimens, usefulness of individualized regimens and conditions for treatment interruption. No consensus was obtained for relevance of the different fluid types in AMD prognosis, frequency of fluid presence assessment, factors commonly associated with progression to geographic atrophy, ideal conditions for a fixed treatment regimen, date of first disease activity assessment and parameters to monitor disease activity.

CONCLUSIONS

Consensus was achieved for over half of the questions assessed through this Delphi study. The questions for which no consensus was reached concerned either subjects that need further investigation or monitoring times which are influenced by resource availability. Raising awareness for these issues will allow the improvement of AMD management and treatment.

Detection of oedema on optical coherence tomography images using deep learning model trained on noisy clinical data

PURPOSE

To meet the demands imposed by the continuing growth of the Age-related macular degeneration (AMD) patient population, automation of follow-ups by detecting retinal oedema using deep learning might be a viable approach. However, preparing and labelling data for training is time consuming. In this study, we investigate the feasibility of training a convolutional neural network (CNN) to accurately detect retinal oedema on optical coherence tomography (OCT) images of AMD patients with labels derived directly from clinical treatment decisions, without extensive preprocessing or relabelling.

METHODS

A total of 50 439 OCT images with associated treatment information were retrieved from databases at the Department of Ophthalmology, Rigshospitalet, Copenhagen, Denmark between 01.06.2007 and 01.06.2018. A CNN was trained on the retrieved data with the recorded treatment decisions as labels and validated on a subset of the data relabelled by three ophthalmologists to denote presence of oedema.

RESULTS

Moderate inter-grader agreement on presence of oedema in the relabelled data was found (76.4%). Despite different training and validation labels, the CNN performed on par with inter-grader agreement in detecting oedema on OCT images (AUC 0.97, accuracy 90.9%) and previously published models based on relabelled datasets.

CONCLUSION

The level of performance shown by the current model might make it valuable in detecting disease activity in automated AMD patient follow-up systems. Our approach demonstrates that high accuracy is not necessarily constrained by incongruent training and validation labels. These results might encourage the use of existing clinical databases for development of deep learning based algorithms without labour-intensive preprocessing in the future.

Bacillary Layer Detachment in a Korean Cohort with Neovascular Age-Related Macular Degeneration

PURPOSE

To evaluate the incidence and characteristics of bacillary layer detachment (BALAD) in neovascular age-related macular degeneration (AMD).

METHODS

This retrospective study was performed at Kim's Eye Hospital in South Korea. Patients who were diagnosed with neovascular AMD between January 2017 and December 2017 were included. The incidence of BALAD was compared among different types of macular neovascularization. The best-corrected visual acuity (BCVA) and central retinal thickness (CRT) at diagnosis were compared between patients showing BALAD at diagnosis and those who did not.

RESULTS

Among the 442 patients included, BALAD was observed in 20 patients(4.5%). There was a significant difference in the incidence of BALAD between type 1 MNV(2.7%), type 2 MNV(12.5%), and type 3 MNV (0%)(P<0.001). The BCVA was significantly worse (mean 1.26±0.79 vs 0.62±0.50, P=0.001), and the CRT was significantly greater (mean 648.2±211.1 µm vs 464.0±175.5 µm, P<0.001) in patients with BALAD than in those without it. After anti-vascular endothelial growth factor therapy, all BALADs resolved.

CONCLUSIONS

This study first reported the incidence of the BALAD in neovascular AMD in Korean population. The incidence of BALAD was the highest in type 2 MNVs. BALAD generally develops in eyes with great macular thickness and poor visual acuity.

Optimal approaches and criteria to treat-and-extend regimen implementation for Neovascular age-related macular degeneration: experts consensus in Taiwan

The management of neovascular age-related macular degeneration (nAMD) has taken a major stride forward with the advent of anti-VEGF agents. The treat-and-extend (T&E) approach is a refined management strategy, tailoring to the individual patient’s disease course and treatment outcome. To provide guidance to implementing anti-VEGF T&E regimens for nAMD in resource-limited health care systems, an advisory board was held to discuss and generate expert consensus, based on local and international guidelines, current evidence, as well as local experience and reimbursement policies. In the experts’ opinion, treatment of nAMD should aim to maximize and maintain visual acuity benefits while minimizing treatment burden. Based on current evidence, treatment could be initiated with 3 consecutive monthly injections. After the initial period, treatment interval may be extended by 2 or 4 weeks each time for the qualified patients (i.e. no BCVA loss ≥5 ETDRS letters and dry retina), and a maximum interval of 16 weeks is permitted. For patients meeting the shortening criteria (i.e. any increased fluid with BCVA loss ≥5 ETDRS letters, or presence of new macular hemorrhage or new neovascularization), the treatment interval should be reduced by 2 or 4 weeks each time, with a minimal interval of 4 weeks. Discontinuation of anti-VEGF may be considered for those who have received 2–3 consecutive injections spaced 16 weeks apart and present with stable disease. For these individuals, regular monitoring (e.g. 3–4 months) is recommended and monthly injections should be reinstated upon signs of disease recurrence.

Quantitative assessment of retinal fluid in neovascular age-related macular degeneration under anti-VEGF therapy

The retinal world has been revolutionized by optical coherence tomography (OCT) and anti-vascular endothelial growth factor (VEGF) therapy. The numbers of intravitreal injections are on a constant rise and management in neovascular age-related macular degeneration (nAMD) is mainly driven by the qualitative assessment of macular fluid as detected on OCT scans. The presence of macular fluid, particularly subretinal fluid (SRF) and intraretinal fluid (IRF), has been used to trigger re-treatments in clinical trials and the real world. However, large discrepancies can be found between the evaluations of different readers or experts and especially small amounts of macular fluid might be missed during this process. Pixel-wise detection of macular fluid uses an entire OCT volume to calculate exact volumes of retinal fluid. While manual annotations of such pixel-wise fluid detection are unfeasible in a clinical setting, artificial intelligence (AI) is able to overcome this hurdle by providing real-time results of macular fluid in different retinal compartments. Quantitative fluid assessments have been used for various post hoc analyses of randomized controlled trials, providing novel insights into anti-VEGF treatment regimens. Nonetheless, the application of AI-algorithms in a prospective patient care setting is still limited. In this review, we discuss the use of quantitative fluid assessment in nAMD during anti-VEGF therapy and provide an outlook to novel forms of patient care with the support of AI quantifications.

Modern trends in diagnostics and prediction of results of anti-vascular endothelial growth factor therapy of pigment epithelial detachment in neovascular agerelated macular degeneration using deep machine learning method (literature review)

Detachment of the pigment epithelium is the separation of the basement membrane of the retinal pigment epithelium from the inner collagen layer of Bruch’s membrane, which occurs in 80 % of cases in patients with neovascular age-related macular degeneration. The outcome of anti-VEGF therapy for pigment epithelial detachment may be adherence of the pigment epithelium, the formation of pigment epithelium tear, or preservation of the detachment. The pigment epithelium tear of 3–4th degrees can lead to a sharp decrease in visual acuity.Most retrospective studies confi rm the absence of a proven correlation between anatomical and functional outcomes in the treatment of pigment epithelial detachment in cases of maintaining the integrity of the pigment epithelium monolayer, and therefore the main attention of researchers is focused on studying the morphological features of pigment epithelial detachment during therapy with angiogenesis inhibitors. Modern technologies of spectral optical coherence tomography make it possible to evaluate detailed quantitative parameters of pigment epithelium detachment, such as height, width, maximum linear diameter, area, volume and refl ectivity within the detachment.Groups of Russian and foreign authors identify various biomarkers recorded on optical coherence tomography images. Dynamic registration of such biomarkers expands the ability of clinicians to predict morphological changes in pigment epithelial detachment during anti-VEGF therapy, as well as to optimize treatment regimens to prevent complications in the form of pigment epithelium tear leading to a decrease in visual acuity.Modern methods of deep machine learning and the use of neural networks allow achieving higher accuracy in diff erentiating the types of retinal fluids and automating the quantitative determination of fl uid under the pigment epithelium. These technologies allow achieving a high level of compliance with manual expert assessment and increasing the accuracy and speed of predicting morphological results of treatment of pigment epithelium detachments.

Impact of intra- and subretinal fluid on vision based on volume quantification in the HARBOR trial

PURPOSE

To investigate the functional associations of intra- and subretinal fluid volumes at baseline, after the loading dose as well as fluid change after the first injection with best corrected visual acuity (BCVA) in patients with neovascular AMD (nAMD) who received an anti-VEGF treatment over 24 months.

DESIGN

Post-hoc analysis of a phase III, randomized, multicenter trial, in which ranibizumab was administered monthly or in a PRN regimen (HARBOR).

PARTICIPANTS

Study eyes of 1094 treatment-naïve nAMD patients.

METHODS

Intraretinal (IRF) and subretinal (SRF) fluid volumes were segmented automatically on monthly SD-OCT images. Fluid volumes and changes thereof were included as covariates into longitudinal mixed effects models, modeling BCVA trajectories.

MAIN OUTCOME MEASURES

BCVA estimates corresponding to baseline, follow-up and persistent IRF/SRF volumes following the loading dose; BCVA estimates of change in fluid volumes following the first injection; marginal and conditional R2.

RESULTS

Analysis 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/100nl). After the first injection, BCVA increased by +2.13 letters/100nl decrease of 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/100nl). After the first injection, SRF decrease was associated with significant vision gain (+5.88 letters/100nl). Foveal fluid correlated more with BCVA than parafoveal IRF/SRF.

CONCLUSION

While 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.

IMPACT OF RESIDUAL SUBRETINAL FLUID VOLUMES ON TREATMENT OUTCOMES IN A SUBRETINAL FLUID-TOLERANT TREAT-AND-EXTEND REGIMEN

PURPOSE

To investigate associations between residual subretinal fluid (rSRF) volumes, quantified using artificial intelligence and treatment outcomes in a subretinal fluid (SRF)-tolerant treat-and-extend (T&E) regimen in neovascular age-related macular degeneration.

METHODS

Patients enrolled in the prospective, multicenter FLUID study randomized in an SRF-tolerant T&E regimen were examined by spectral-domain optical coherence tomography and tested for best-corrected visual acuity (BCVA). Intraretinal fluid and SRF volumes were quantified using artificial intelligence tools. In total, 375 visits of 98 patients were divided into subgroups: extended intervals despite rSRF and extended intervals without fluid. Associations between BCVA change, SRF volume, subgroups, and treatment intervals were estimated using linear mixed models.

RESULTS

In extended intervals despite rSRF, increased SRF was associated with reduced BCVA at the next visit in the central 1 mm (-0.138 letters per nL; P = 0.014) and 6 mm (-0.024 letters per nL; P = 0.049). A negative association between increased interval and BCVA change was found for rSRF in 1 mm and 6 mm (-0.250 and -0.233 letter per week interval, respectively; both P < 0.001). Extended intervals despite rSRF had significantly higher SRF volumes in the central 6 mm at the following visit (P = 0.002).

CONCLUSION

Artificial intelligence-based analysis of extended visits despite rSRF demonstrated increasing SRF volumes associated with BCVA loss at the consecutive visit. This negative association contributes to the understanding of rSRF volumes on treatment outcomes in neovascular age-related macular degeneration.

Updates in deep learning research in ophthalmology

Ophthalmology has been one of the early adopters of artificial intelligence (AI) within the medical field. Deep learning (DL), in particular, has garnered significant attention due to the availability of large amounts of data and digitized ocular images. Currently, AI in Ophthalmology is mainly focused on improving disease classification and supporting decision-making when treating ophthalmic diseases such as diabetic retinopathy, age-related macular degeneration (AMD), glaucoma and retinopathy of prematurity (ROP). However, most of the DL systems (DLSs) developed thus far remain in the research stage and only a handful are able to achieve clinical translation. This phenomenon is due to a combination of factors including concerns over security and privacy, poor generalizability, trust and explainability issues, unfavorable end-user perceptions and uncertain economic value. Overcoming this challenge would require a combination approach. Firstly, emerging techniques such as federated learning (FL), generative adversarial networks (GANs), autonomous AI and blockchain will be playing an increasingly critical role to enhance privacy, collaboration and DLS performance. Next, compliance to reporting and regulatory guidelines, such as CONSORT-AI and STARD-AI, will be required to in order to improve transparency, minimize abuse and ensure reproducibility. Thirdly, frameworks will be required to obtain patient consent, perform ethical assessment and evaluate end-user perception. Lastly, proper health economic assessment (HEA) must be performed to provide financial visibility during the early phases of DLS development. This is necessary to manage resources prudently and guide the development of DLS.

Quantitative effect of subretinal fluid and intraretinal edema on visual acuity in uveitic cystoid macular edema

Background

The effect of subretinal fluid (SRF) in uveitic cystoid macular edema (CME) is not fully understood. This study evaluates the quantitative effect of SRF and intraretinal thickness on visual acuity in eyes with uveitic CME. We separately measured SRF and intraretinal area on Optical Coherence Tomography (OCT) to determine the associations of each component with visual acuity and response to treatment.

Main text

Medical records were reviewed of patients with CME presenting to the University of Colorado uveitis clinic from January 2012 to May 2019. All available OCTs were reviewed to classify eyes as either having only CME or CME with SRF. Intraretinal area was manually measured using Image J along the central 1-mm section of B-scan OCT spanning from the internal limiting membrane to the outer most portion of the outer retina including both cysts and retinal tissue. SRF cross-sectional area was measured spanning from the outermost portion of the outer retina to retinal pigment epithelium. Response to treatment was assessed one to four months after presentation. Eyes with CME secondary to structural or non-inflammatory causes were excluded.

Forty-seven (50.5%) eyes had CME alone and 46 (49.5%) eyes had SRF with CME. Measured SRF cross-sectional area was not associated (p = 0.21) with LogMAR at presentation. Conversely, intraretinal area was strongly correlated with visual acuity in eyes with SRF (p < 0.001) and without SRF (p < 0.001). Following treatment, there was a significant decrease in intraretinal area for both groups (p < 0.001), with a larger decrease in the SRF group compared to the non-SRF group (p = 0.001). Similarly, logMAR improved in both groups (p = 0.008 for SRF eyes and p = 0.005 for non-SRF eyes), but the change was more prominent in the SRF group (p = 0.06).

Conclusions

There was no direct association observed between the amount of SRF and visual acuity. In contrast, increased intraretinal area was significantly associated with decreased visual acuity. This relationship between intraretinal thickening and visual acuity may explain differences observed in response to treatment between SRF and non-SRF eyes, with a larger decrease in the intraretinal cross-sectional area in SRF eyes associated with a greater improvement in logMAR visual acuity.

Artificial intelligence-based predictions in neovascular age-related macular degeneration

PURPOSE OF REVIEW

Predicting treatment response and optimizing treatment regimen in patients with neovascular age-related macular degeneration (nAMD) remains challenging. Artificial intelligence-based tools have the potential to increase confidence in clinical development of new therapeutics, facilitate individual prognostic predictions, and ultimately inform treatment decisions in clinical practice.

RECENT FINDINGS

To date, most advances in applying artificial intelligence to nAMD have focused on facilitating image analysis, particularly for automated segmentation, extraction, and quantification of imaging-based features from optical coherence tomography (OCT) images. No studies in our literature search evaluated whether artificial intelligence could predict the treatment regimen required for an optimal visual response for an individual patient. Challenges identified for developing artificial intelligence-based models for nAMD include the limited number of large datasets with high-quality OCT data, limiting the patient populations included in model development; lack of counterfactual data to inform how individual patients may have fared with an alternative treatment strategy; and absence of OCT data standards, impairing the development of models usable across devices.

SUMMARY

Artificial intelligence has the potential to enable powerful prognostic tools for a complex nAMD treatment landscape; however, additional work remains before these tools are applicable to informing treatment decisions for nAMD in clinical practice.

Predictors of Visual Acuity After Treatment of Neovascular Age-Related Macular Degeneration - Current Perspectives

Visual acuity is a key outcome measure in the treatment of neovascular age-related macular degeneration (nAMD) using anti-vascular endothelial growth factor agents. Large variations in visual responses between individuals within clinical trials and real-world studies may relate to underlying differences in patient and treatment factors. Most notably, a better baseline visual acuity, younger age and smaller choroidal neovascularization lesion size have been strongly associated with achieving better visual outcomes. In addition, there is emerging evidence for other roles including genetic factors and anatomical variables such as fluid status. Apart from patient-related factors, treatments that favor a higher number of injections tend to provide better visual outcomes. Overall, the identification of predictive factors does not currently play an essential role in the clinical management of patients with nAMD. However, they have allowed for the understanding that early detection, timely management and close monitoring of the disease are required to achieve optimal visual outcomes. Further investigation into predictive factors alongside the development of novel therapeutic agents may one day provide a means to accurately predict patient outcomes. Treatment regimens that offer flexible dosing patterns such as the treat-and-extend strategy currently provide a degree of personalization during treatment.

Validation and Clinical Applicability of Whole-Volume Automated Segmentation of Optical Coherence Tomography in Retinal Disease Using Deep Learning

Question

Is deep learning–based segmentation of macular disease in optical coherence tomography (OCT) suitable for clinical use?

Findings

In this diagnostic study of OCT data from 173 patients with age-related macular degeneration or diabetic macular edema, model segmentations qualitatively ranked better or comparable for clinical applicability to 1 or more expert grader segmentations in 127 scans (73%) by a panel of 3 retinal specialists. Scans with high quantitative accuracy scores were not reliably associated with higher rankings.

Meaning

These findings suggest that qualitative evaluation adds to quantitative approaches when assessing clinical applicability of segmentation tools and clinician satisfaction in practice.

Importance

Quantitative volumetric measures of retinal disease in optical coherence tomography (OCT) scans are infeasible to perform owing to the time required for manual grading. Expert-level deep learning systems for automatic OCT segmentation have recently been developed. However, the potential clinical applicability of these systems is largely unknown.

Objective

To evaluate a deep learning model for whole-volume segmentation of 4 clinically important pathological features and assess clinical applicability.

Design, Setting, Participants

This diagnostic study used OCT data from 173 patients with a total of 15 558 B-scans, treated at Moorfields Eye Hospital. The data set included 2 common OCT devices and 2 macular conditions: wet age-related macular degeneration (107 scans) and diabetic macular edema (66 scans), covering the full range of severity, and from 3 points during treatment. Two expert graders performed pixel-level segmentations of intraretinal fluid, subretinal fluid, subretinal hyperreflective material, and pigment epithelial detachment, including all B-scans in each OCT volume, taking as long as 50 hours per scan. Quantitative evaluation of whole-volume model segmentations was performed. Qualitative evaluation of clinical applicability by 3 retinal experts was also conducted. Data were collected from June 1, 2012, to January 31, 2017, for set 1 and from January 1 to December 31, 2017, for set 2; graded between November 2018 and January 2020; and analyzed from February 2020 to November 2020.

Main Outcomes and Measures

Rating and stack ranking for clinical applicability by retinal specialists, model-grader agreement for voxelwise segmentations, and total volume evaluated using Dice similarity coefficients, Bland-Altman plots, and intraclass correlation coefficients.

Results

Among the 173 patients included in the analysis (92 [53%] women), qualitative assessment found that automated whole-volume segmentation ranked better than or comparable to at least 1 expert grader in 127 scans (73%; 95% CI, 66%-79%). A neutral or positive rating was given to 135 model segmentations (78%; 95% CI, 71%-84%) and 309 expert gradings (2 per scan) (89%; 95% CI, 86%-92%). The model was rated neutrally or positively in 86% to 92% of diabetic macular edema scans and 53% to 87% of age-related macular degeneration scans. Intraclass correlations ranged from 0.33 (95% CI, 0.08-0.96) to 0.96 (95% CI, 0.90-0.99). Dice similarity coefficients ranged from 0.43 (95% CI, 0.29-0.66) to 0.78 (95% CI, 0.57-0.85).

Conclusions and Relevance

This deep learning–based segmentation tool provided clinically useful measures of retinal disease that would otherwise be infeasible to obtain. Qualitative evaluation was additionally important to reveal clinical applicability for both care management and research.

The Challenges of Treating Neovascular Age-Related Macular Degeneration

Age-related macular degeneration (AMD) is one of the main causes of visual impairment and blindness in patients over 60 years in developed countries. Whilst no effective form of therapy is available for the dry form of AMD, intravitreal application of anti-VEGF substances is able to prevent the progression of neovascular AMD (nAMD) in most cases. Aside from the drugs ranibizumab, aflibercept and brolucizumab, other agents such as bevacizumab are often used off-label in order to save expense. The treatment intervals have also been refined, so as to reduce the burden on patients and health care systems. After fixed injection intervals, the pro re nata-regimen has been developed. Each month, it is decided whether the patient receives intravitreal injections based on fixed criteria. In the treat and extend-protocol, patients receive injections on each visit, but the intervals between injections vary due to the clinical outcomes. The observe-and-plan regime allows scheduling of the injection intervals in blocks, for three consecutive injections at a time. However, results of real-world studies were not able to reproduce those obtained in the pivotal studies. A high number of visits and fear of the injection procedure impose a burden on patients, that is mostly accepted due to fear of vision loss. Caregivers also complain of loss of productivity and income from having to provide regular support to patients. Health care systems worldwide are affected by increasing treatment numbers and the costs involved. The treatment of nAMD constitutes an achievement for modern medicine. However, despite the challenges, it must be evaluated and reviewed repeatedly in order to provide the best therapy for patients.

An observational clinical study of the influence of phacoemulsification on choroidal neovascular membrane activity in age related macular degeneration

Background

Thousands of phacoemulsification surgeries are performed on eyes with age-related macular degeneration (AMD) complicated by choroidal neovascular membrane (CNV) in the United Kingdom each year. As populations age this number is expected to rise. Controversy over phacoemulsification’s influence on CNV activity limits the information which clinicians and these patients use to decide on surgery. This observational study aims to resolve this controversy by reporting on intravitreal injection (IVI) frequency as a pragmatic marker of CNV activity in a large cohort.

Methods

A cohort of eyes with AMD complicated by CNV (n = 327) that underwent cataract surgery at a single tertiary centre from 2014 to 2019 were identified. These cases were matched by interval since CNV diagnosis at a specified ‘time zero’ within the follow-up of pseudophakic eyes with AMD (n = 327). Data concerning demographics, visual acuity (VA) and intravitreal injection frequency before and after ‘time zero’/phacoemulsification were collected.

Results

Following ‘time zero’/phacoemulsification’ the mean reduction in annual IVI frequency was 0.6 injections/year (95% CI 0.4,0.9) and 0.4 injections/year (95% CI 0.1,0.7) in the comparison and phacoemulsification cohorts respectively. The mean VA gain 12 months after phacoemulsification in the intervention cohort was 11.3 (95% CI 9.2,13.4) early treatment of diabetic retinopathy study (ETDRS) letters, with 214 eyes (65.4%) having gained ≥5 ETDRS letters after surgery.

Conclusions

Phacoemulsification has no clinically significant impact on the activity of pre-existent CNV secondary to AMD. Phacoemulsification should be offered to patients with AMD and cataract that limits vision, regardless of CNV activity.

AI-based monitoring of retinal fluid in disease activity and under therapy

Retinal fluid as the major biomarker in exudative macular disease is accurately visualized by high-resolution three-dimensional optical coherence tomography (OCT), which is used world-wide as a diagnostic gold standard largely replacing clinical examination. Artificial intelligence (AI) with its capability to objectively identify, localize and quantify fluid introduces fully automated tools into OCT imaging for personalized disease management. Deep learning performance has already proven superior to human experts, including physicians and certified readers, in terms of accuracy and speed. Reproducible measurement of retinal fluid relies on precise AI-based segmentation methods that assign a label to each OCT voxel denoting its fluid type such as intraretinal fluid (IRF) and subretinal fluid (SRF) or pigment epithelial detachment (PED) and its location within the central 1-, 3- and 6-mm macular area. Such reliable analysis is most relevant to reflect differences in pathophysiological mechanisms and impacts on retinal function, and the dynamics of fluid resolution during therapy with different regimens and substances. Yet, an in-depth understanding of the mode of action of supervised and unsupervised learning, the functionality of a convolutional neural net (CNN) and various network architectures is needed. Greater insight regarding adequate methods for performance, validation assessment, and device- and scanning-pattern-dependent variations is necessary to empower ophthalmologists to become qualified AI users. Fluid/function correlation can lead to a better definition of valid fluid variables relevant for optimal outcomes on an individual and a population level. AI-based fluid analysis opens the way for precision medicine in real-world practice of the leading retinal diseases of modern times.

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

PURPOSE

We sought to develop and validate a deep learning model for segmentation of 13 features associated with neovascular and atrophic age-related macular degeneration (AMD).

DESIGN

Development and validation of a deep-learning model for feature segmentation.

METHODS

Data for model development were obtained from 307 optical coherence tomography volumes. Eight experienced graders manually delineated all abnormalities in 2712 B-scans. A deep neural network was trained with these data to perform voxel-level segmentation of the 13 most common abnormalities (features). For evaluation, 112 B-scans from 112 patients with a diagnosis of neovascular AMD were annotated by 4 independent observers. The main outcome measures were Dice score, intraclass correlation coefficient, and free-response receiver operating characteristic curve.

RESULTS

On 11 of 13 features, the model obtained a mean Dice score of 0.63 ± 0.15, compared with 0.61 ± 0.17 for the observers. The mean intraclass correlation coefficient for the model was 0.66 ± 0.22, compared with 0.62 ± 0.21 for the observers. Two features were not evaluated quantitatively because of a lack of data. Free-response receiver operating characteristic analysis demonstrated that the model scored similar or higher sensitivity per false positives compared with the observers.

CONCLUSIONS

The quality of the automatic segmentation matches that of experienced graders for most features, exceeding human performance for some features. The quantified parameters provided by the model can be used in the current clinical routine and open possibilities for further research into treatment response outside clinical trials.

Deep Learning-Based Optical Coherence Tomography and Optical Coherence Tomography Angiography Image Analysis: An Updated Summary

ABSTRACT

Deep learning (DL) is a subset of artificial intelligence based on deep neural networks. It has made remarkable breakthroughs in medical imaging, particularly for image classification and pattern recognition. In ophthalmology, there are rising interests in applying DL methods to analyze optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA) images. Studies showed that OCT and OCTA image evaluation by DL algorithms achieved good performance for disease detection, prognosis prediction, and image quality control, suggesting that the incorporation of DL technology could potentially enhance the accuracy of disease evaluation and the efficiency of clinical workflow. However, substantial issues, such as small training sample size, data preprocessing standardization, model robustness, results explanation, and performance cross-validation, are yet to be tackled before deploying these DL models in real-time clinics. This review summarized recent studies on DL-based image analysis models for OCT and OCTA images and discussed the potential challenges of clinical deployment and future research directions.

Automated Quantitative Assessment of Retinal Fluid Volumes as Important Biomarkers in Neovascular Age-Related Macular Degeneration

PURPOSE

To evaluate retinal fluid volume data extracted from optical coherence tomography (OCT) scans by artificial intelligence algorithms in the treatment of neovascular age-related macular degeneration (NV-AMD).

DESIGN

Perspective.

METHODS

A review was performed of retinal image repository datasets from diverse clinical settings.

SETTINGS

Clinical trial (HARBOR) and trial follow-on (Age-Related Eye Disease Study 2 10-year Follow-On); real-world (Belfast and Tel-Aviv tertiary centers).

PATIENTS

24,362 scans of 1,095 eyes (HARBOR); 4,673 of 880 (Belfast); 1,470 of 132 (Tel-Aviv); 511 of 511 (Age-Related Eye Disease Study 2 10-year Follow-On). ObservationProcedures: Vienna Fluid Monitor or Notal OCT Analyzer applied to macular cube scans. OutcomeMeasures: Intraretinal fluid (IRF), subretinal fluid (SRF), and pigment epithelial detachment (PED) volumes.

RESULTS

The fluid volumes measured in neovascular AMD were expressed efficiently in nanoliters. Large ranges that differed by population were observed at the treatment-naïve stage: 0-3,435 nL (IRF), 0-5,018 nL (SRF), and 0-10,022 nL (PED). Mean volumes decreased rapidly and consistently with anti-vascular endothelial growth factor therapy. During maintenance therapy, mean IRF volumes were highest in Tel-Aviv (100 nL), lower in Belfast and HARBOR-Pro Re Nata, and lowest in HARBOR-monthly (21 nL). Mean SRF volumes were low in all: 30 nL (HARBOR-monthly) and 48-49 nL (others).

CONCLUSIONS

Quantitative measures of IRF, SRF, and PED are important biomarkers in NV-AMD. Accurate volumes can be extracted efficiently from OCT scans by artificial intelligence algorithms to guide the treatment of exudative macular diseases. Automated fluid monitoring identifies fluid characteristics in different NV-AMD populations at baseline and during follow-up. For consistency between studies, we propose the nanoliter as a convenient unit. We explore the advantages of using these quantitative metrics in clinical practice and research.

Aflibercept therapy for exudative age-related macular degeneration resistant to bevacizumab and ranibizumab

Background

Despite the good outcomes achieved with intravitreal angiogenic therapy, a subset of neovascular age-related macular degeneration (AMD) patients experience resistance to therapy after repeated injections. Switching drugs could offer benefit to this group of patients.

Purpose

To determine visual and anatomical outcomes in a cohort of neovascular AMD patients resistant to repeated injections of bevacizumab/ranibizumab after switching to aflibercept therapy.

Methods

This was a retrospective chart review of patients who had a diagnosis of neovascular AMD and persistent intraretinal (IRF) and/or subretinal fluid (SRF) on optical coherence tomography (OCT) for at least 3 months despite monthly bevacizumab and/or ranibizumab injections prior to transition to aflibercept. We reviewed patients’ records and OCT images obtained at baseline, 1, 3, 6 and 12 months after transition to aflibercept. Data collected included demographics, best-corrected visual acuity (BCVA), number of injections received and the occurrence of any adverse events. Studied OCT parameters included central macular thickness (CMT) values and the presence or absence of SRF, IRF and/or pigment epithelial detachment (PED) at each visit.

Results

We included 53 eyes of 48 patients. Mean change in BCVA from baseline was 0.05 ± 0.13 (P = 0.01) at M1, 0.04 ± 0.16 (P = 0.08) at M3, 0.01 ± 0.22 (P = 0.9) at M6, and 0.02 ± 0.28 (P = 1) at M12, while the mean change in CMT from baseline was 64 ± 75 μm (P < 0.0001) at M1, 42 ± 85 μm (P = 0.002) at M3, 47 ± 69 μm (P < 0.0001) at M6, and 46 ± 99 μm (P = 0.001) at M12. The percentage of eyes with SRF decreased from 77.4% at baseline to 39.6% at M1, then increased to 47.2% at M3, then decreased to 43.4% at M6, and to 41.5% at M12 (All p < 0.001, compared to baseline). Compared to baseline, there was a statistically significant decrease in the percentage of eyes having IRF from 47.2 to 20.8% at M1 (p < 0.001), 30.2% at M3, 24.5% at M6 and 26.4% at M12 (p < 0.01, each). The number of bevacizumab and/or ranibizumab injections (7.36 ± 1.85) was significantly higher than that of aflibercept (6.47 ± 2.45, p = 0.001). A significant direct relationship between CMT reduction and BCVA improvement was demonstrated at M1 (p = 0.01, r = 0.36), M3 (p = 0.03, r = 0.30) and M12 (p = 0.03, r = 0.30). Eyes with IRF had significantly poorer BCVA than eyes without IRF at baseline (p = 0.02) and M3 (p = 0.04).

Conclusion

Switching to intravitreal aflibercept therapy in a cohort of neovascular AMD patients resistant to chronic bevacizumab and/or ranibizumab injections can lead to significant visual improvement in the short term and sustained reduction of central macular thickness over 1 year of followup.

Morphometric analysis of fibrovascular pigment epithelial detachments treated with ranibizumab and aflibercept

PURPOSE

To assess fibrovascular pigment epithelial detachments (PED) and their response to two different anti-VEGF agents using optical coherence tomography (OCT) morphometric analysis.

METHODS

Seventy-three consecutive, treatment-naïve eyes with fibrovascular PED (>125 μm) treated with ranibizumab or aflibercept were retrospectively included. A custom-made software was used to manually segment and calculate PED maximum height, base area, volume and internal reflectivity at baseline, after three injections and 1 year.

RESULTS

Visual acuity (VA) change was 2 ETDRS letters ± 7.6 after three injections and 3.2 ETDRS letters ± 10.3 at 1 year. There was no significant difference between VA changes amongst the two drugs. At 1 year, anti-VEGF treatment resulted in a mean reduction of 125 μm in maximum PED height, of 2.26 mm² in base area and of 0.54 mm³ in volume with a corresponding increase in reflectivity. These changes were more prominent in the aflilbercept group. The observed PED and VA changes at year 1 were strongly correlated with their values at baseline and after three injections.

CONCLUSIONS

Anti-VEGF treatment resulted in a reduction of all PED dimensions and a corresponding increase in optical reflectivity. Higher, larger and more hypo-reflective PEDs demonstrated a better anatomical response, especially with aflibercept, but this was not correlated with VA.

Quantification of Fluid Resolution and Visual Acuity Gain in Patients With Diabetic Macular Edema Using Deep Learning: A Post Hoc Analysis of a Randomized Clinical Trial

Importance

Large amounts of optical coherence tomographic (OCT) data of diabetic macular edema (DME) are acquired, but many morphologic features have yet to be identified and quantified.

Objective

To examine the volumetric change of intraretinal fluid (IRF) and subretinal fluid (SRF) in DME during anti-vascular endothelial growth factor treatment using deep learning algorithms.

Design, Setting, and Participants

This post hoc analysis of a randomized clinical trial, the Diabetic Retinopathy Clinical Research Network (protocol T), assessed 6945 spectral-domain OCT volume scans of 570 eyes from 570 study participants with DME. The original trial was performed from August 21, 2012, to October 18, 2018. This analysis was performed from December 7, 2017, to January 15, 2020.

Interventions

Participants were treated according to a predefined, standardized protocol with aflibercept, ranibizumab, or bevacizumab with or without deferred laser.

Main Outcomes and Measures

The association of treatment with IRF and SRF volumes and best-corrected visual acuity (BCVA) during 12 months using deep learning algorithms.

Results

Among the 570 study participants (302 [53%] male; 369 [65%] white; mean [SD] age, 43.4 [12.6] years), the mean fluid volumes in the central 3 mm were 448.6 nL (95% CI, 412.3-485.0 nL) of IRF and 36.9 nL (95% CI, 27.0-46.7 nL) of SRF at baseline and 161.2 nL (95% CI, 135.1-187.4 nL) of IRF and 4.4 nL (95% CI, 1.7-7.1 nL) of SRF at 12 months. The presence of SRF at baseline was associated with a worse baseline BCVA Early Treatment Diabetic Retinopathy Study (ETDRS) score of 63.2 (95% CI, 60.2-66.1) (approximate Snellen equivalent of 20/63 [95% CI, 20/50-20/63]) in eyes with SRF vs 66.9 (95% CI, 65.7-68.1) (approximate Snellen equivalent, 20/50 [95% CI, 20/40-20/50]) without SRF (P < .001) and a greater gain in ETDRS score (0.5; 95% CI, 0.3-0.8) every 4 weeks during follow-up in eyes with SRF at baseline vs 0.4 (95% CI, 0.3-0.5) in eyes without SRF at baseline (P = .02) when adjusted for baseline BCVA. Aflibercept was associated with greater reduction of IRF volume compared with bevacizumab after the first injection (difference, 79.8 nL; 95% CI, 5.3-162.5 nL; P < .001) and every 4 weeks thereafter (difference, 10.4 nL; 95% CI, 0.7-20.0 nL; P = .004). Ranibizumab was associated with a greater reduction of IRF after the first injection compared with bevacizumab (difference, 75.2 nL; 95% CI, 1.4-154.7 nL; P < .001).

Conclusions and Relevance

Automated segmentation of fluid in DME revealed that the presence of SRF was associated with lower baseline BCVA but with good response to anti-vascular endothelial growth factor therapy. These automated spectral-domain OCT analyses may be used clinically to assess anatomical change during therapy.

Trial Registration

ClinicalTrials.gov Identifier: NCT01627249.

Digital health during COVID-19: lessons from operationalising new models of care in ophthalmology

The COVID-19 pandemic has resulted in massive disruptions within health care, both directly as a result of the infectious disease outbreak, and indirectly because of public health measures to mitigate against transmission. This disruption has caused rapid dynamic fluctuations in demand, capacity, and even contextual aspects of health care. Therefore, the traditional face-to-face patient-physician care model has had to be re-examined in many countries, with digital technology and new models of care being rapidly deployed to meet the various challenges of the pandemic. This Viewpoint highlights new models in ophthalmology that have adapted to incorporate digital health solutions such as telehealth, artificial intelligence decision support for triaging and clinical care, and home monitoring. These models can be operationalised for different clinical applications based on the technology, clinical need, demand from patients, and manpower availability, ranging from out-of-hospital models including the hub-and-spoke pre-hospital model, to front-line models such as the inflow funnel model and monitoring models such as the so-called lighthouse model for provider-led monitoring. Lessons learnt from operationalising these models for ophthalmology in the context of COVID-19 are discussed, along with their relevance for other specialty domains.

Predicting Incremental and Future Visual Change in Neovascular Age-Related Macular Degeneration Using Deep Learning

PURPOSE

To evaluate the predictive usefulness of quantitative imaging biomarkers, acquired automatically from OCT scans, of cross-sectional and future visual outcomes of patients with neovascular age-related macular degeneration (AMD) starting anti-vascular endothelial growth factor (VEGF) therapy.

DESIGN

Retrospective cohort study.

PARTICIPANTS

Treatment-naive, first-treated eyes of patients with neovascular AMD between 2007 and 2017 at Moorfields Eye Hospital (a large, United Kingdom single center) undergoing anti-VEGF therapy.

METHODS

Automatic segmentation was carried out by applying a deep learning segmentation algorithm to 137 379 OCT scans from 6467 eyes of 3261 patients with neovascular AMD. After applying selection criteria, 926 eyes of 926 patients were analyzed.

MAIN OUTCOME MEASURES

Correlation coefficients (R² values) and mean absolute error (MAE) between quantitative OCT (qOCT) parameters and cross-sectional visual function, as well as the predictive value of these parameters for short-term visual change, that is, incremental visual acuity (VA) resulting from an individual injection, as well as VA at distant time points (up to 12 months after baseline).

RESULTS

Visual acuity at distant time points could be predicted: R² = 0.80 (MAE, 5.0 Early Treatment Diabetic Retinopathy Study [ETDRS] letters) and R² = 0.7 (MAE, 7.2 ETDRS letters) after injection at 3 and at 12 months after baseline (P < 0.001 for both), respectively. Best performing models included both baseline qOCT parameters and treatment response. Furthermore, we present proof-of-principle evidence that the incremental change in VA from an injection can be predicted: R² = 0.14 (MAE, 5.6 ETDRS letters) for injection 2 and R² = 0.11 (MAE, 5.0 ETDRS letters) for injection 3 (P < 0.001 for both).

CONCLUSIONS

Automatic segmentation enables rapid acquisition of quantitative and reproducible OCT biomarkers with potential to inform treatment decisions in the care of neovascular AMD. This furthers development of point-of-care decision-aid systems for personalized medicine.

Digital Technology for AMD Management in the Post-COVID-19 New Normal

PURPOSE

The COVID-19 pandemic has put strain on healthcare systems and the availability and allocation of healthcare manpower, resources and infrastructure. With immediate priorities to protect the health and safety of both patients and healthcare service providers, ophthalmologists globally were advised to defer nonurgent cases, while at the same time managing sight-threatening conditions such as neovascular Age-related Macular Degeneration (AMD). The management of AMD patients both from a monitoring and treatment perspective presents a particular challenge for ophthalmologists. This review looks at how these pressures have encouraged the acceptance and speed of adoption of digitalization.

DESIGN AND METHODS

A literature review was conducted on the use of digital technology during COVID-19 pandemic, and on the transformation of medicine, ophthalmology and AMD screening through digitalization.

RESULTS

In the management of AMD, the implementation of artificial intelligence and "virtual clinics" have provided assistance in screening, diagnosis, monitoring of the progression and the treatment of AMD. In addition, hardware and software developments in home monitoring devices has assisted in self-monitoring approaches.

CONCLUSIONS

Digitalization strategies and developments are currently ongoing and underway to ensure early detection, stability and visual improvement in patients suffering from AMD in this COVID-19 era. This may set a precedence for the post COVID-19 new normal where digital platforms may be routine, standard and expected in healthcare delivery.

[Different types localisation of retinal fluid as prognostic biomarkers in the choice of anti-VEGF therapy for age-related macular degeneration]

Age-related macular degeneration is an advanced chronic disease and the main cause of vision loss in geriatric patients. Optical coherence tomography (OCT) is a modern method of retinal imaging allowing to detect different types of fluid: intraretinal fluid (IRF), subretinal fluid (SRF) and fluid under pigment epithelial detachment (PED). Finding relevant imaging biomarkers is necessary for identification of basic activity criteria of the disease, choosing treatment algorithms, determining treatment duration and termination criteria, and predicting the outcomes. Presence of IRF is associated with poor functional outcomes. Its presence is an indication for early beginning of treatment aimed at full resorption of the fluid with further possible careful extension of anti-VEGF therapy intervals with a regular follow-up. Degenerative intraretinal cysts developing in the background of subretinal fibrosis in absence of choroidal neovascularization (CNV) should be a sign for discontinuation of anti-VEGF therapy due to the lack of targets. Presence of SRF is associated with favorable outcomes and good treatment prognosis and is not a barrier to the extension of treatment intervals even up to the maximum of 16 weeks as described in existing randomized controlled trials, on the condition of no other CNV activity. PED with active CNV is one of the biomarkers that reveal the need for long-term aggressive therapy. In case of its size gain, it is necessary to restart the anti-VEGF treatment to prevent visual loss in the long-term. Combination of different fluid types is a sign of lasting disease history with a poor outcome prognosis. In this case, anti-VEGF treatment should be started as soon as possible with long-term fixed regimen or Treat-and-extend (T&E) with minimal suitable interval for the patient and precise monitoring of the condition of retina until complete suppression of activity. Developing a personalized approach in each case plays an important role in preserving visual functions.

Evaluation of the Safety and Efficacy of Selective Retina Therapy Laser Treatment in Patients with Central Serous Chorioretinopathy

Purpose

To assess the safety and efficacy of selective retina therapy (SRT) using a Q-switched neodymium-doped yttrium lithium fluoride laser with feedback systems in patients with idiopathic central serous chorioretinopathy (CSC).

Methods

This randomized clinical trial enrolled patients having at least 3-month symptom of CSC. From month 3 visit, all subjects in both groups were eligible for SRT retreatment if they showed persistent or recurrent subretinal fluid (SRF). The primary outcome was complete resolution of SRF by optical coherence tomography at 3 months after treatment. The secondary outcomes were changes in SRF, central macular thickness (CMT) and best-corrected visual acuity at the 1-, 3-, and 6-month examinations.

Results

Sixty-eight CSC patients were included (SRT, 31; control, 37). After 1 and 3 months, complete resolution of SRF was achieved in 25.8% and 54.8% of SRT group and 17.6% and 35.1% of controls. The differences were not statistically significant (p = 0.424 and p = 0.142, respectively). However, mixed model for repeated measures analyses showed that the reduction of SRF and CMT were observed earlier in SRT group than in the sham group (least squares mean difference, −59.7 μm; 95% confidence interval, −98.2 to −21.2; p = 0.0029; least squares mean difference −67.0 μm; 95% confidence interval, −104.8 to −29.2; p = 0.0007, respectively). Significant reduction of SRF (≥50% reduction from baseline) was more frequently observed in SRT group (80.6%) than the sham group (44.1%) at month 1 (p = 0.007). Early reduction of SRF and CMT was more abundant in SRT group with symptom duration less than 6 months. Treatment related serious adverse events were not observed.

Conclusions

SRT using a Q-switched neodymium-doped yttrium lithium fluoride laser with feedback system was safe in this trial and effective for early resolution of SRF in the CSC patients. Early intervention with SRT can be a safe alternative for patients with acute symptomatic CSC.

ANALYSIS OF FLUID VOLUME AND ITS IMPACT ON VISUAL ACUITY IN THE FLUID STUDY AS QUANTIFIED WITH DEEP LEARNING

PURPOSE

To investigate quantitative differences in fluid volumes between subretinal fluid (SRF)-tolerant and SRF-intolerant treat-and-extend regimens for neovascular age-related macular degeneration and analyze the association with best-corrected visual acuity.

METHODS

Macular fluid (SRF and intraretinal fluid) was quantified on optical coherence tomography volumetric scans using a trained and validated deep learning algorithm. Fluid volumes and complete resolution was automatically assessed throughout the study. The impact of fluid location and volumes on best-corrected visual acuity was computed using mixed-effects regression models.

RESULTS

Baseline fluid quantifications for 348 eyes from 348 patients were balanced (all P > 0.05). No quantitative differences in SRF/intraretinal fluid between the treatment arms was found at any study-specific time point (all P > 0.05). Compared with qualitative assessment, the proportion of eyes without SRF/intraretinal fluid did not differ between the groups at any time point (all P > 0.05). Intraretinal fluid in the central 1 mm and SRF in the 1-mm to 6-mm macular area were negatively associated with best-corrected visual acuity (-2.8 letters/100 nL intraretinal fluid, P = 0.007 and -0.20 letters/100 nL SRF, P = 0.005, respectively).

CONCLUSION

Automated fluid quantification using artificial intelligence allows objective and precise assessment of macular fluid volume and location. Precise determination of fluid parameters will help improve therapeutic efficacy of treatment in neovascular age-related macular degeneration.

Automated quantification of macular fluid in retinal diseases and their response to anti-VEGF therapy

AIM

To objectively assess disease activity and treatment response in patients with retinal vein occlusion (RVO), neovascular age-related macular degeneration (nAMD) and centre-involved diabetic macular oedema (DME), using artificial intelligence-based fluid quantification.

METHODS

Posthoc analysis of 2311 patients (11 151 spectral-domain optical coherence tomography volumes) from five clinical, multicentre trials, who received a flexible antivascular endothelial growth factor (anti-VEGF) therapy over a 12-month period. Fluid volumes were measured with a deep learning algorithm at baseline/months 1, 2, 3 and 12, for three concentric circles with diameters of 1, 3 and 6 mm (fovea, paracentral ring and pericentral ring), as well as four sectors surrounding the fovea (superior, nasal, inferior and temporal).

RESULTS

In each disease, at every timepoint, most intraretinal fluid (IRF) per square millimetre was present at the fovea, followed by the paracentral ring and pericentral ring (p<0.0001). While this was also the case for subretinal fluid (SRF) in RVO/DME (p<0.0001), patients with nAMD showed more SRF in the paracentral ring than at the fovea up to month 3 (p<0.0001). Between sectors, patients with RVO/DME showed the highest IRF volumes temporally (p<0.001/p<0.0001). In each disease, more SRF was consistently found inferiorly than superiorly (p<0.02). At month 1/12, we measured the following median reductions of initial fluid volumes. For IRF: RVO, 95.9%/97.7%; nAMD, 91.3%/92.8%; DME, 37.3%/69.9%. For SRF: RVO, 94.7%/97.5%; nAMD, 98.4%/99.8%; DME, 86.3%/97.5%.

CONCLUSION

Fully automated localisation and quantification of IRF/SRF over time shed light on the fluid dynamics in each disease. There is a specific anatomical response of IRF/SRF to anti-VEGF therapy in all diseases studied.

Principal Cause of Poor Visual Acuity after Neovascular Age-Related Macular Degeneration: Age-Related Eye Disease Study 2 Report Number 23

PURPOSE

To analyze the principal cause for poor vision in eyes with best-corrected visual acuity (BCVA) of 20/200 or worse 2 years after neovascular age-related macular degeneration (nAMD).

DESIGN

Prospective cohort study of participants enrolled in a clinical trial of oral supplements.

PARTICIPANTS

Age-Related Eye Disease Study 2 (AREDS2) participants whose eyes began anti-vascular endothelial growth factor (VEGF) therapy for incident nAMD and had data available at 2 years.

METHODS

Participants underwent refracted BCVA testing, ophthalmoscopic examination, and fundus photography at baseline and annual visits. Self-reports of anti-VEGF injections were collected.

MAIN OUTCOME MEASURES

Principal cause of BCVA of 20/200 or worse at 2 years, detected on fundus photography grading.

RESULTS

Of the 594 eligible eyes, the number with BCVA of 20/200 or worse at 2 years was 56 (9.4%). Mean BCVA was 14.9 letters (standard deviation [SD], 12.3 letters; Snellen equivalent, 20/500), versus 70.1 letters (SD, 12.8 letters; Snellen equivalent, 20/40) in the other group. Of the 55 eyes with fundus photography available at 2 years, 33 (60.0%) had central macular atrophy and 22 (40.0%) had central subretinal fibrosis assessed as the principal cause for poor vision. The group with poor BCVA had a higher proportion of non-White participants (8.9% vs. 1.7%; P = 0.006), lower BCVA 2 years earlier (mean, 38.0 letters [SD, 26.7 letters; Snellen equivalent, 20/160] vs. 71.8 letters (SD, 11.9 letters; Snellen equivalent, 20/40]; P < 0.0001), higher proportion with macular atrophy 2 years earlier (26.8% vs. 12.3%; P = 0.003), higher proportion with macular hemorrhage (25.5% vs. 13.2%; P = 0.014), and fewer anti-VEGF injections (7.6 vs. 10.2; P = 0.001).

CONCLUSIONS

Visual acuity data and fundus photography were obtained in a clinical trial environment, but were related to anti-VEGF therapy given in routine clinical practice. At 2 years after starting anti-VEGF therapy, almost 1 in 10 eyes showed BCVA at the level of legal blindness. From fundus photography grading, the cause of poor vision appeared to be macular atrophy in 60% and subretinal fibrosis in 40%. These data may be useful in understanding the long-term limits to good vision in nAMD.

Quantitative Analysis of OCT for Neovascular Age-Related Macular Degeneration Using Deep Learning

PURPOSE

To apply a deep learning algorithm for automated, objective, and comprehensive quantification of OCT scans to a large real-world dataset of eyes with neovascular age-related macular degeneration (AMD) and make the raw segmentation output data openly available for further research.

DESIGN

Retrospective analysis of OCT images from the Moorfields Eye Hospital AMD Database.

PARTICIPANTS

A total of 2473 first-treated eyes and 493 second-treated eyes that commenced therapy for neovascular AMD between June 2012 and June 2017.

METHODS

A deep learning algorithm was used to segment all baseline OCT scans. Volumes were calculated for segmented features such as neurosensory retina (NSR), drusen, intraretinal fluid (IRF), subretinal fluid (SRF), subretinal hyperreflective material (SHRM), retinal pigment epithelium (RPE), hyperreflective foci (HRF), fibrovascular pigment epithelium detachment (fvPED), and serous PED (sPED). Analyses included comparisons between first- and second-treated eyes by visual acuity (VA) and race/ethnicity and correlations between volumes.

MAIN OUTCOME MEASURES

Volumes of segmented features (mm³) and central subfield thickness (CST) (μm).

RESULTS

In first-treated eyes, the majority had both IRF and SRF (54.7%). First-treated eyes had greater volumes for all segmented tissues, with the exception of drusen, which was greater in second-treated eyes. In first-treated eyes, older age was associated with lower volumes for RPE, SRF, NSR, and sPED; in second-treated eyes, older age was associated with lower volumes of NSR, RPE, sPED, fvPED, and SRF. Eyes from Black individuals had higher SRF, RPE, and serous PED volumes compared with other ethnic groups. Greater volumes of the majority of features were associated with worse VA.

CONCLUSIONS

We report the results of large-scale automated quantification of a novel range of baseline features in neovascular AMD. Major differences between first- and second-treated eyes, with increasing age, and between ethnicities are highlighted. In the coming years, enhanced, automated OCT segmentation may assist personalization of real-world care and the detection of novel structure-function correlations. These data will be made publicly available for replication and future investigation by the AMD research community.