Search

This systematic review was performed to identify the specifics of an optimal diabetic retinopathy deep learning algorithm, by identifying the best exemplar research studies of the field, whilst highlighting potential barriers to clinical implementation of such an algorithm. Searching five electronic databases (Embase, MEDLINE, Scopus, PubMed, and the Cochrane Library) returned 747 unique records on 20 December 2019. Predetermined inclusion and exclusion criteria were applied to the search results, resulting in 15 highest-quality publications. A manual search through the reference lists of relevant review articles found from the database search was conducted, yielding no additional records. A validation dataset of the trained deep learning algorithms was used for creating a set of optimal properties for an ideal diabetic retinopathy classification algorithm. Potential limitations to the clinical implementation of such systems were identified as lack of generalizability, limited screening scope, and data sovereignty issues. It is concluded that deep learning algorithms in the context of diabetic retinopathy screening have reported impressive results. Despite this, the potential sources of limitations in such systems must be evaluated carefully. An ideal deep learning algorithm should be clinic-, clinician-, and camera-agnostic; complying with the local regulation for data sovereignty, storage, privacy, and reporting; whilst requiring minimum human input. [ABSTRACT FROM AUTHOR]

Objective: To assess diabetes eye service use in New Zealand among people aged ≥15 years by estimating service attendance, biennial screening rate, and disparities in the use of screening and treatment services. Methods: We obtained Ministry of Health data from the National Non-Admitted Patient Collection on diabetes eye service events between 1 July 2006 and 31 December 2019 and sociodemographic and mortality data from the Virtual Diabetes Register and linked these using a unique patient identifier (encrypted National Health Index). We 1) summarized attendance at retinal screening and ophthalmology services, 2) calculated biennial and triennial screening rate, 3) summarized treatment with laser and anti-VEGF and used log-binomial regression to examine associations of all of these with age group, ethnicity, and area-level deprivation. Results: In total, 245,844 people aged ≥15 years had at least one diabetes eye service appointment attended or scheduled; half of these (n = 125,821, 51.2%) attended only retinal screening, one-sixth attended only ophthalmology (n = 35,883, 14.6%) and one-third attended both (n = 78,300, 31.8%). The biennial retinal screening rate was 62.1%, with large regional variation (73.9% in Southern District to 29.2% in West Coast). Compared with NZ Europeans, Māori were approximately twice as likely to never receive diabetes eye care or to access ophthalmology when referred from retinal screening, 9% relatively less likely to receive biennial screening and received the fewest anti-VEGF injections when treatment was commenced. Disparities in service access were also present for Pacific Peoples compared to NZ Europeans, younger and older age groups compared to those aged 50–59 years and those living in areas with higher deprivation. Conclusions: Access to diabetes eye care is suboptimal, with substantial disparity between age groups, ethnicity groups, area level deprivation quintile and across districts. Efforts to improve access to and quality of diabetes eye care services must include strengthening data collection and monitoring. [ABSTRACT FROM AUTHOR]

Deep learning (DL) models have shown promise in detecting chronic kidney disease (CKD) from fundus photographs. However, previous studies have utilized a serum creatinine-only estimated glomerular rate (eGFR) equation to measure kidney function despite the development of more up-to-date methods. In this study, we developed two sets of DL models using fundus images from the UK Biobank to ascertain the effects of using a creatinine and cystatin-C eGFR equation over the baseline creatinine-only eGFR equation on fundus image-based DL CKD predictors. Our results show that a creatinine and cystatin-C eGFR significantly improved classification performance over the baseline creatinine-only eGFR when the models were evaluated conventionally. However, these differences were no longer significant when the models were assessed on clinical labels based on ICD10. Furthermore, we also observed variations in model performance and systemic condition incidence between our study and the ones conducted previously. We hypothesize that limitations in existing eGFR equations and the paucity of retinal features uniquely indicative of CKD may contribute to these inconsistencies. These findings emphasize the need for developing more transparent models to facilitate a better understanding of the mechanisms underpinning the ability of DL models to detect CKD from fundus images. [ABSTRACT FROM AUTHOR]

The aim of this study is to correlate small dot hyper‐reflective foci (HRF) observed in spectral domain optical coherence tomography (SD‐OCT) scans of an animal model of hyperglycaemia with focal electroretinography (fERG) response and immunolabelling of retinal markers. The eyes of an animal model of hyperglycaemia showing signs of diabetic retinopathy (DR) were imaged using SD‐OCT. Areas showing dot HRF were further evaluated using fERG. Retinal areas enclosing the HRF were dissected and serially sectioned, stained and labelled for glial fibrillary acidic protein (GFAP) and a microglial marker (Iba‐1). Small dot HRF were frequently seen in OCT scans in all retinal quadrants in the inner nuclear layer or outer nuclear layer in the DR rat model. Retinal function in the HRF and adjacent areas was reduced compared with normal control rats. Microglial activation was detected by Iba‐1 labelling and retinal stress identified by GFAP expression in Müller cells observed in discrete areas around small dot HRF. Small dot HRF seen in OCT images of the retina are associated with a local microglial response. This study provides the first evidence of dot HRF correlating with microglial activation, which may allow clinicians to better evaluate the microglia‐mediated inflammatory component of progressive diseases showing HRF. [ABSTRACT FROM AUTHOR]

Background: Clinical ophthalmological guidelines encourage the assessment of potential benefits and harms when deciding whether to perform elective ophthalmology procedures during the COVID‐19 pandemic, in order to minimize the risk of disease transmission. Method: We performed probability calculations to estimate COVID‐19 infection status and likelihood of disease transmission among neovascular age‐related macular degeneration patients and health‐care workers during anti‐VEGF procedures, at various community prevalence levels of COVID‐19. We then applied the expected burden of COVID‐19 illness and death expressed through health‐adjusted life‐years (HALYs) lost. We compared these results to the expected disease burden of severe visual impairment if sight protecting anti‐VEGF injections were not performed. Results: Our calculations suggest a wide range of contexts where the benefits of treatment to prevent progression to severe visual impairment or blindness are greater than the expected harms to the patient and immediate health care team due to COVID‐19. For example, with appropriate protective equipment the benefits of treatment outweigh harms when the chance of progression to severe visual impairment is >0.044% for all scenarios where COVID‐19 prevalence was 1/1000, even when the attack rate in the clinical setting is very high (5‐43%). Conclusion: Unless COVID‐19 prevalence is very high, the reduced disease burden from avoiding visual impairment outweighs the expected HALYs lost from COVID‐19 transmission. This finding is driven by the fact that HALYs lost when someone suffers severe visual impairment for 5 years are equivalent to nearly 400 moderate cases of infectious disease lasting 2 weeks each. [ABSTRACT FROM AUTHOR]

Convolutional Neural Networks (CNNs) have become a prominent method of AI implementation in medical classification tasks. Grading Diabetic Retinopathy (DR) has been at the forefront of the development of AI for ophthalmology. However, major obstacles remain in the generalization of these CNNs onto real-world DR screening programs. We believe these difficulties are due to use of 1) small training datasets (<5,000 images), 2) private and 'curated' repositories, 3) locally implemented CNN implementation methods, while 4) relying on measured Area Under the Curve (AUC) as the sole measure of CNN performance. To address these issues, the public EyePACS Kaggle Diabetic Retinopathy dataset was uploaded onto Microsoft Azure™ cloud platform. Two CNNs were trained; 1 a "Quality Assurance", and 2. a "Classifier". The Diabetic Retinopathy classifier CNN (DRCNN) performance was then tested both on 'un-curated' as well as the 'curated' test set created by the "Quality Assessment" CNN model. Finally, the sensitivity of the DRCNNs was boosted using two post-training techniques. Our DRCNN proved to be robust, as its performance was similar on 'curated' and 'un-curated' test sets. The implementation of 'cascading thresholds' and 'max margin' techniques led to significant improvements in the DRCNN's sensitivity, while also enhancing the specificity of other grades. [ABSTRACT FROM AUTHOR]

Pseudophakic cystoid macular oedema (PCMO) remains a significant cause of compromised postoperative vision in contemporary cataract surgery. Well‐established risk factors include intraoperative complications such as posterior capsule rupture and preoperative factors including: diabetes mellitus, uveitis, retinal vein occlusion, epiretinal membrane. The role of topical glaucoma medications in PCMO continues to be debated. Current treatment strategies largely target suppression of inflammation. Topical NSAIDs remain the mainstay in prophylaxis and treatment of PCMO. Topical corticosteroids are commonly used as monotherapy or in combination with NSAIDs. Unfortunately, high‐quality trials are notably lacking for other PCMO treatment modalities such as: periocular corticosteroids, orbital floor triamcinolone, intravitreal triamcinolone, corticosteroid implants, intravitreal bevacizumab and pars‐plana vitrectomy. A lack of consistency in defining PCMO and resolution of PCMO explains why even large systematic reviews may come to contradictory conclusions. This review explores the varied contemporary evidence‐base in relation to the aetiology, diagnosis, prophylaxis and treatment of PCMO. [ABSTRACT FROM AUTHOR]

Abstract: Diabetic retinopathy (DR) is a vascular disease of the neuroretina characterised by hyperglycaemia and inflammation. Current DR therapies target late-stage vascular defects and there is evidence to suggest that they contribute to geographic atrophy and retinal ganglion cell death long term. Therefore, alternative treatments that target common upstream disease mechanisms are needed. Recent studies have shown that connexin43 hemichannel blockers can reduce inflammation and prevent vessel leak in brain and spinal cord lesions. The aim of this study was to evaluate the effectiveness of a connexin43 hemichannel blocker (Peptide5) in a mouse model of DR in which pro-inflammatory cytokines, IL-1β and TNF-α, were intravitreally injected into non-obese diabetic (NOD, hyperglycaemic) mice. Fundus and optical coherence tomography images were taken to evaluate vessel dilation and beading as well as retinal and vitreous hyper-reflective foci (HRF). Immunohistochemistry was performed to assess levels of astrogliosis, microgliosis and inflammasome activation. Results showed that Peptide5 injection lowered the incidence of vessel dilation and beading, decreased the severity of vitreous and retinal HRF, and reduced sub-retinal fluid accumulation compared to the vehicle group. Furthermore, Peptide5 led to reduced connexin43 and GFAP upregulation, inhibited microglial infiltration into the outer nuclear layer and prevented upregulation of inflammasome markers compared to vehicle. The present study provides evidence in support of Peptide5, and connexin43 hemichannel block in general, as a potential upstream approach for the treatment of DR.Key messages: Connexin43 is upregulated in a novel mouse model of diabetic retinopathy (DR).Connexin43 hemichannel block inhibits inflammation and inflammasome activation.Connexin43 hemichannel block prevents the development of clinical DR signs.Connexin43 hemichannel block is a potential upstream approach for DR treatment. [ABSTRACT FROM AUTHOR]

Diabetic retinopathy is a vascular disease of the retina characterised by hyperglycaemic and inflammatory processes. Most animal models of diabetic retinopathy are hyperglycaemia-only models that do not account for the significant role that inflammation plays in the development of the disease. In the present study, we present data on the establishment of a new animal model of diabetic retinopathy that incorporates both hyperglycaemia and inflammation. We hypothesized that inflammation may trigger and worsen the development of diabetic retinopathy in a hyperglycaemic environment. Pro-inflammatory cytokines, IL-1β and TNF-α, were therefore injected into the vitreous of non-obese diabetic (NOD) mice. CD1 mice were used as same genetic background controls. Fundus and optical coherence tomography images were obtained before (day 0) as well as on days 2 and 7 after intravitreal cytokine injection to assess vessel dilation and beading, retinal and vitreous hyper-reflective foci and retinal thickness. Astrogliosis and microgliosis were assessed using immunohistochemistry. Results showed that intravitreal cytokines induced vessel dilation, beading, severe vitreous hyper-reflective foci, retinal oedema, increased astrogliosis and microglia upregulation in diabetic NOD mice. Intravitreal injection of inflammatory cytokines into the eyes of diabetic mice therefore appears to provide a new model of diabetic retinopathy that could be used for the study of disease progression and treatment strategies. [ABSTRACT FROM AUTHOR]

Purpose: To create an ensemble of Convolutional Neural Networks (CNNs), capable of detecting and stratifying the risk of progressive age-related macular degeneration (AMD) from retinal photographs. Design: Retrospective cohort study. Methods: Three individual CNNs are trained to accurately detect 1) advanced AMD, 2) drusen size and 3) the presence or otherwise of pigmentary abnormalities, from macular centered retinal images were developed. The CNNs were then arranged in a "cascading" architecture to calculate the Age-related Eye Disease Study (AREDS) Simplified 5-level risk Severity score (Risk Score 0 – Risk Score 4), for test images. The process was repeated creating a simplified binary "low risk" (Scores 0– 2) and "high risk" (Risk Score 3– 4) classification. Participants: There were a total of 188,006 images, of which 118,254 images were deemed gradable, representing 4591 patients, from the AREDS1 dataset. The gradable images were split into 50%/25%/25% ratios for training, validation and test purposes. Main Outcome Measures: The ability of the ensemble of CNNs using retinal images to predict an individual's risk of experiencing progression of their AMD based on the AREDS 5-step Simplified Severity Scale. Results: When assessed against the 5-step Simplified Severity Scale, the results generated by the ensemble of CNN's achieved an accuracy of 80.43% (quadratic kappa 0.870). When assessed against a simplified binary (Low Risk/High Risk) classification, an accuracy of 98.08%, sensitivity of ≥ 85% and specificity of ≥ 99% was achieved. Conclusion: We have created an ensemble of neural networks, trained on the AREDS 1 dataset, that is able to accurately calculate an individual's score on the AREDS 5-step Simplified Severity Scale for AMD. If the results presented were replicated, then this ensemble of CNNs could be used as a screening tool that has the potential to significantly improve health outcomes by identifying asymptomatic individuals who would benefit from AREDS2 macular supplements. [ABSTRACT FROM AUTHOR]

Background: Bevacizumab is the only agent that many people can afford, yet there are only limited data on whether it improves macular oedema (MO) secondary to retinal vein occlusion (RVO) in real‐world clinical practice. Here we studied 12‐month real‐world treatment outcomes of bevacizumab for RVO‐related MO. Methods: This was a multicentre, observational study analysing 12‐month data from the Fight Retinal Blindness! (FRB) database. We studied treatment‐naïve eyes with MO secondary to RVO commencing bevacizumab therapy between June 2009 and June 2019. Visual acuity (VA) and central subfield thickness (CST) were measured at baseline, 6 and 12 months. The primary outcome was a change in VA from baseline to 12 months. Results: Two hundred and twenty treatment naive eyes were analyzed. The baseline VA for BRVO was better than CRVO (55.8 vs. 42.6 LogMAR letters) and this gap widened over the 12‐month period, with a 12‐month VA change of +14.0 (95% CI 11.1, 16.8) letters for BRVO and + 11.9 (95% CI 6.4, 17.4) for CRVO. The mean CST at baseline was 511 μm for BRVO and 627 μm for CRVO, falling at 12 months by −155 μm (−190, −121) in BRVO and −198 μm (−252, −145) in CRVO. The median number of injections for BRVO and CRVO completers was 7 (5, 9). Conclusions: Bevacizumab can be an effective treatment of RVO‐MO in a real‐world setting with outcomes approaching those reported by the seminal clinical trials. The functional and anatomical outcomes of intravitreal therapy were better for BRVO than CRVO. [ABSTRACT FROM AUTHOR]

Purpose: The main purpose of the study was to report the estimated incidence, cumulative rate, risk factors and outcomes of submacular haemorrhage (SMH) with loss of vision in neovascular age‐related macular degeneration (nAMD) receiving intravitreal injections (IVT) of vascular endothelial growth factor (VEGF) inhibitor in routine clinical practice. Methods: Retrospective analysis of treatment‐naïve eyes receiving IVTs of VEGF inhibitors (ranibizumab, aflibercept or bevacizumab) for nAMD from 1 January 2010 to 31 December 2020 that were tracked the Fight Retinal Blindness! registry. Estimated incidence, cumulative rate and hazard ratios (HR) of SMH with loss of vision during treatment were measured using the Poisson regression, Kaplan–Meier survival curves and Cox proportional hazard models. Results: We identified 7642 eyes (6425 patients) with a total of 135 095 IVT over a 10‐year period. One hundred five eyes developed SMH with loss of vision with a rate of 1 per 1283 injections (0.08% 95% confidence interval [95% CI] [0.06; 0.09]). The estimated incidence [95% CI] was 4.6 [3.8; 5.7] SMH with loss of vision per year per 1000 treated patients during the study. The cumulative [95% CI] rate of SMH per patient did not increase significantly with each successive injection (p = 0.947). SMH cases had a mean VA drop of around 6 lines at diagnosis, which then improved moderately to a 4‐line loss at 1 year. Conclusions: Submacular haemorrhage (SMH) with loss of vision is an uncommon complication that can occur at any time in eyes treated for nAMD in routine clinical practice, with only limited recovery of vision 1 year later. [ABSTRACT FROM AUTHOR]

Topical non-steroidal anti-inflammatory drugs are not the mainstay of prophylaxis and treatment for pseudophakic cystoid macular oedema: Response We thank Levitz, Lim, Grzybowski and Kim, well-known and prolific correspondents on scientific publications, for their interest in, and kind comments upon, our recent publication[1] on pseudophakic cystoid macular oedema (PCMO); "The review article does allude to the challenges faced in providing advice on prophylaxis of PCMO" and "The authors have provided a significant review of a complex topic".[2] However, in relation to other comments, we feel the correspondents may have confused the semantics of what we believe is a balanced review, stating they would "Like to highlight several aspects of the quoted literature which may be misconstrued". Topical NSAIDs are not the mainstay of prophylaxis and treatment for pseudophakic cystoid macular oedema. [Extracted from the article]

Background: We assessed the proportion of eyes with neovascular age‐related macular degeneration (nAMD) in routine clinical practice that reach ≥14 week treatment intervals and their outcomes. Method: We analysed data from the Fight Retinal Blindness! (FRB!) Project database, a prospectively designed registry of 'real‐world' outcomes. Treatment‐naive eyes starting vascular endothelial growth factor (VEGF) inhibitors for nAMD from 1st January 2006 were included. Eyes were defined to have reached the ≥14 week treatment interval if they received ≥2 consecutive injections at treatment intervals of ≥14 week but not exceeding 26 weeks. Outcomes were reported in a subgroup of eyes that had 12 months of follow‐up from reaching this interval. Results: Of the 3907 treatment‐naïve eyes that started treatment during the identified periods on a treat‐and‐extend regimen and received at least 8 injections over the first 2 years, 402 (10%) eyes received at least 2 consecutive injections at an interval of ≥14 week during their follow‐up. Fifty‐two percent of these eyes maintained vision to 12 months, however only 40% stayed at this interval and 25% of the lesions reactivated. Conclusion: We found that only 10% of eyes with nAMD were extended beyond a 13‐week injection interval and that over half had returned to a shorter interval by 12 months. Eyes that stayed at this extended treatment interval maintained stable vision. More data on the outcomes of eyes treated with intervals longer than 3 months are required to establish whether emerging VEGF inhibitors provide a more sustained effect than the currently available drugs. [ABSTRACT FROM AUTHOR]

Importance: There are limited data on real‐world outcomes of cataract surgery in eyes receiving intravitreal treatments for diabetic macular oedema (DMO). Background: Cataract surgery may exacerbate oedema in some eyes with DMO resulting in inferior outcomes. Design: Matched, case‐controlled retrospective study of observational data in routine clinical practice. Participants: Eyes receiving intravitreal treatments for DMO tracked in the Fight Retinal Blindness! Registry. Methods: Eyes that underwent cataract surgery were identified and matched 1:1 with phakic controls also receiving intravitreal injections for DMO. We also assessed potential factors that were associated with better visual acuity (VA) outcomes. Main Outcome Measures: Change in VA 6 months after cataract surgery. Results: Cataract surgery was identified in 208 eyes of 156 patients of which 147 eyes had 6 months of observations before and after surgery. The mean VA 6 months after surgery improved by 10.6 letters and was similar to their matched phakic controls (68.8 vs 69.2 letters; P = 0.8). Mean CST both 6 months before (341 μm) and after (360 μm) surgery were similar (P = 0.08). However, these eyes had thicker maculae and they received more injections than their matched phakic controls both before and after surgery. Eyes with worse VA before surgery and those that had received intravitreal treatment in the 4 weeks preceding surgery were more likely to gain vision. Conclusions and Relevance: Visual outcomes of cataract surgery in eyes receiving intravitreal therapy for DMO were reasonably better. Their maculae were thicker and required more injections in the 6 months before and after surgery than their phakic controls. [ABSTRACT FROM AUTHOR]

Background and Objective. To determine if using a multi-input deep learning approach in the image analysis of optical coherence tomography (OCT), OCT angiography (OCT-A), and colour fundus photographs increases the accuracy of a CNN to diagnose intermediate dry age-related macular degeneration (AMD). Patients and Methods. Seventy-five participants were recruited and divided into three cohorts: young healthy (YH), old healthy (OH), and patients with intermediate dry AMD. Colour fundus photography, OCT, and OCT-A scans were performed. The convolutional neural network (CNN) was trained on multiple image modalities at the same time. Results. The CNN trained using OCT alone showed a diagnostic accuracy of 94%, whilst the OCT-A trained CNN resulted in an accuracy of 91%. When multiple modalities were combined, the CNN accuracy increased to 96% in the AMD cohort. Conclusions. Here we demonstrate that superior diagnostic accuracy can be achieved when deep learning is combined with multimodal image analysis. [ABSTRACT FROM AUTHOR]