MICROANEURYSM FORMATION RATE AS A PREDICTIVE MARKER FOR PROGRESSION TO CLINICALLY SIGNIFICANT MACULAR EDEMA IN NONPROLIFERATIVE DIABETIC RETINOPATHY

Artificial Intelligence (AI) for Early Diagnosis of Retinal Diseases

Artificial intelligence (AI) has emerged as a transformative tool in the field of ophthalmology, revolutionizing disease diagnosis and management. This paper provides a comprehensive overview of AI applications in various retinal diseases, highlighting its potential to enhance screening efficiency, facilitate early diagnosis, and improve patient outcomes. Herein, we elucidate the fundamental concepts of AI, including machine learning (ML) and deep learning (DL), and their application in ophthalmology, underscoring the significance of AI-driven solutions in addressing the complexity and variability of retinal diseases. Furthermore, we delve into the specific applications of AI in retinal diseases such as diabetic retinopathy (DR), age-related macular degeneration (AMD), Macular Neovascularization, retinopathy of prematurity (ROP), retinal vein occlusion (RVO), hypertensive retinopathy (HR), Retinitis Pigmentosa, Stargardt disease, best vitelliform macular dystrophy, and sickle cell retinopathy. We focus on the current landscape of AI technologies, including various AI models, their performance metrics, and clinical implications. Furthermore, we aim to address challenges and pitfalls associated with the integration of AI in clinical practice, including the "black box phenomenon", biases in data representation, and limitations in comprehensive patient assessment. In conclusion, this review emphasizes the collaborative role of AI alongside healthcare professionals, advocating for a synergistic approach to healthcare delivery. It highlights the importance of leveraging AI to augment, rather than replace, human expertise, thereby maximizing its potential to revolutionize healthcare delivery, mitigate healthcare disparities, and improve patient outcomes in the evolving landscape of medicine.

Correlation between Topographic Vessel Density and Retinal Thickness Changes in Patients with Diabetic Macular Edema Treated with Anti-VEGF Therapy: Is It a Suitable OCTA Biomarker?

The objective of this study was to determine the correlation between topographic vessel density (VD) and retinal thickness (RT) reductions induced by vascular endothelial growth factor inhibitors (anti-VEGF) in patients with diabetic macular edema (DME) using optical coherence tomography angiography (OCTA). This was a prospective, interventional case series. VD and RT measurements were separately taken in four parafoveal subfields at baseline and after six months of treatment. This correlation was statistically assessed using Spearman's rho correlation coefficient after adjustment for multiple comparisons. The study included a total of 48 eyes in the final analysis. Mean VD decreased from baseline to month 6 (from 45.2 (±3.5) to 44.6% (±3.2) in the superficial capillary plexus and from 50 (±3.3) to 49% (±3.9) in the deep capillary plexus). Statistically significant reductions in RT were observed in all ETDRS sectors (p < 0.0001). No significant association was found between RT and VD, even when analyzing responders and non-responders separately. After six months of anti-VEGF treatment, no significant correlation was observed between the topographic VD and RT values. These findings suggest that reductions in VD values may not solely result from a reduction in microaneurysms, also being affected by the repositioning of displaced vessels due to edema and a reduction in their caliber. Therefore, VD changes may not be a suitable indirect OCTA biomarker of microaneurysm turnover and treatment response.

Evaluation of microaneurysms as predictors of therapeutic response to anti-VEGF therapy in patients with DME

Administration of intravitreal anti-vascular endothelial growth factor (anti-VEGF) therapy is the first-line therapy for diabetic macular oedema (DME). However, some patients show no or insufficient response to repeated anti-VEGF injections. Therefore, it is necessary to identify factors that can predict this resistance against anti-VEGF treatment. Presence of microaneurysms (MAs) is a predictor of the development and progression of DME, but its relationship with the treatment response to the anti-VEGF agents is not well known. Therefore, we aimed to elucidate the relationship between the distribution of MAs and the response to anti-VEGF therapy in patients with DME. The number of MAs was measured before anti-VEGF therapy in each region using fluorescein angiography, indocyanine green angiography (IA), and optical coherence tomography angiography. Patients with DME were divided into the responder and non-responder groups after three loading phases. Differences in the distribution of MAs between the groups were investigated. Pre-treatment IA revealed more MAs in the nasal area in the non-responder group than in the responder group (10.7 ± 10.7 and 5.7 ± 5.7, respectively, in the nasal macula) (1.4 ± 2.1 and 0.4 ± 0.7, respectively, in the nasal fovea). Whereas, pre-treatment FA and OCTA could not reveal significantly difference between the groups. Detection of MAs in the nasal macula using pre-treatment IA may indicate resistance to anti-VEGF therapy. We recommend the clinicians confirm the presence of MAs in the nasal macula, as shown by IA, as a predictor of therapeutic response to anti-VEGF therapy in patients with treatment naive DME.

Plasma glycation adducts and various RAGE isoforms are intricately associated with oxidative stress and inflammatory markers in type 2 diabetes patients with vascular complications

BACKGROUND

The secondary vascular complications in diabetes mellitus (DM) are contributed by acute as well as inflammatory responses which get activated due to interaction between glycation adducts and respective receptors.

AIM

The present work was performed to understand the relationship between Advanced glycation end products (AGEs)-receptor for advanced glycation end products (RAGE) interaction with oxidative stress and inflammation in vascular complications.

METHODS

For the present work we recruited 103 controls, 200 patients with type 2 DM, and 200 patients with Diabetic complications. Different Plasma glycation adducts (fructosamine, carbonyls, AGEs, β-amyloid content, free amino groups, and free thiol groups); RAGE isoforms, level of antioxidant such as glutathione, catalase activity, nitric oxide level, total antioxidant capacity, and superoxide dismutase activity, as well as oxidative markers, and expression of Nε-carboxymethyl-lysine (CML), different isoforms of RAGE, NF-κB, and inflammatory markers were analyzed.

RESULTS

Glycation adducts were higher in DM patients and more elevated in nephropathy patients where free amino groups and thiol groups lowered as compared to controls. sRAGE levels and expression were increased mainly in nephropathy. CML expression was higher in nephropathy patients. The antioxidant profile indicates a reduced level of different antioxidants while increased lipid peroxidation and intracellular ROS generation in DM and much higher in nephropathy patients. Expression of membrane RAGE, NF-κB, and inflammatory markers showed a remarkably increased level in DM patients with nephropathy.

CONCLUSION

This work provides the first evidence of four different RAGE isoforms in diabetes and in complications. The glycation via the activation of RAGE, oxidative stress, and resultant inflammation plays a crucial role in the development of diabetic complications.

Predicting Post-Therapeutic Visual Acuity and OCT Images in Patients With Central Serous Chorioretinopathy by Artificial Intelligence

To predict visual acuity (VA) and post-therapeutic optical coherence tomography (OCT) images 1, 3, and 6 months after laser treatment in patients with central serous chorioretinopathy (CSC) by artificial intelligence (AI). Real-world clinical and imaging data were collected at Zhongshan Ophthalmic Center (ZOC) and Xiamen Eye Center (XEC). The data obtained from ZOC (416 eyes of 401 patients) were used as the training set; the data obtained from XEC (64 eyes of 60 patients) were used as the test set. Six different machine learning algorithms and a blending algorithm were used to predict VA, and a pix2pixHD method was adopted to predict post-therapeutic OCT images in patients after laser treatment. The data for VA predictions included clinical features obtained from electronic medical records (20 features) and measured features obtained from fundus fluorescein angiography, indocyanine green angiography, and OCT (145 features). The data for OCT predictions included 480 pairs of pre- and post-therapeutic OCT images. The VA and OCT images predicted by AI were compared with the ground truth. In the VA predictions of XEC dataset, the mean absolute errors (MAEs) were 0.074–0.098 logMAR (within four to five letters), and the root mean square errors were 0.096–0.127 logMAR (within five to seven letters) for the 1-, 3-, and 6-month predictions, respectively; in the post-therapeutic OCT predictions, only about 5.15% (5 of 97) of synthetic OCT images could be accurately identified as synthetic images. The MAEs of central macular thickness of synthetic OCT images were 30.15 ± 13.28 μm and 22.46 ± 9.71 μm for the 1- and 3-month predictions, respectively. This is the first study to apply AI to predict VA and post-therapeutic OCT of patients with CSC. This work establishes a reliable method of predicting prognosis 6 months in advance; the application of AI has the potential to help reduce patient anxiety and serve as a reference for ophthalmologists when choosing optimal laser treatments.

Corneal Edema Visualization With Optical Coherence Tomography Using Deep Learning: Proof of Concept

PURPOSE

Optical coherence tomography (OCT) is essential for the diagnosis and follow-up of corneal edema, but assessment can be challenging in minimal or localized edema. The objective was to develop and validate a novel automated tool to detect and visualize corneal edema with OCT.

METHODS

We trained a convolutional neural network to classify each pixel in the corneal OCT images as "normal" or "edema" and to generate colored heat maps of the result. The development set included 199 OCT images of normal and edematous corneas. We validated the model's performance on 607 images of normal and edematous corneas of various conditions. The main outcome measure was the edema fraction (EF), defined as the ratio between the number of pixels labeled as edema and those representing the cornea for each scan. Overall accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve were determined to evaluate the model's performance.

RESULTS

Mean EF was 0.0087 ± 0.01 in the normal scans and 0.805 ± 0.26 in the edema scans (P < 0.0001). Area under the receiver operating characteristic curve for EF in the diagnosis of corneal edema in individual scans was 0.994. The optimal threshold for distinguishing normal from edematous corneas was 6.8%, with an accuracy of 98.7%, sensitivity of 96.4%, and specificity of 100%.

CONCLUSIONS

The model accurately detected corneal edema and distinguished between normal and edematous cornea OCT scans while providing colored heat maps of edema presence.

Associations with sight-threatening diabetic macular oedema among Indigenous adults with type 2 diabetes attending an Indigenous primary care clinic in remote Australia: a Centre of Research Excellence in Diabetic Retinopathy and Telehealth Eye and Associated Medical Services Network study

Objective

To identify factors associated with sight-threatening diabetic macular oedema (STDM) in Indigenous Australians attending an Indigenous primary care clinic in remote Australia.

Methods and analysis

A cross-sectional study design of retinopathy screening data and routinely-collected clinical data among 236 adult Indigenous participants with type 2 diabetes (35.6% men) set in one Indigenous primary care clinic in remote Australia. The primary outcome variable was STDM assessed from retinal images.

Results

Age (median (range)) was 48 (21–86) years, and known diabetes duration (median (range)) was 8.0 (0–24) years. Prevalence of STDM was high (14.8%) and similar in men and women. STDM was associated with longer diabetes duration (11.7 vs 7.9 years, respectively; p<0.001) and markers of renal impairment: abnormal estimated Glomerular Filtration Rate (eGFR) (62.9 vs 38.3%, respectively; p=0.007), severe macroalbuminuria (>300 mg/mmol) (20.6 vs 5.7%, respectively; p=0.014) and chronic kidney disease (25.7 vs 12.2%, respectively; p=0.035). Some clinical factors differed by sex: anaemia was more prevalent in women. A higher proportion of men were smokers, prescribed statins and had increased albuminuria. Men had higher blood pressure, but lower glycated Haemoglobin A1c (HbA1c) levels and body mass index, than women.

Conclusion

STDM prevalence was high and similar in men and women. Markers of renal impairment and longer diabetes duration were associated with STDM in this Indigenous primary care population. Embedded teleretinal screening, known diabetes duration-based risk stratification and targeted interventions may lower the prevalence of STDM in remote Indigenous primary care services.

Trial registration number

Australia and New Zealand Clinical Trials Register: ACTRN 12616000370404.

The Application of Structural Retinal Biomarkers to Evaluate the Effect of Intravitreal Ranibizumab and Dexamethasone Intravitreal Implant on Treatment of Diabetic Macular Edema

Background: The aim of this study was to compare the therapeutic effect of intravitreal treatment with ranibizumab and dexamethasone using specific swept-source optical coherence tomography retinal biomarkers in patients with diabetic macular edema (DME). Methods: 156 treatment-naïve patients with DME were divided in two groups: 75 patients received 3 monthly intravitreal injections of ranibizumab 0.5 mg (Lucentis^®) (Group 1) and 81 patients received an intravitreal implant of dexamethasone 0.7 mg (Ozurdex^®) (Group 2). Patients were evaluated at baseline (V1), at three months post-treatment in Group 1, and at two months post-treatment in Group 2 (V2). Best-corrected visual acuity (BCVA) and swept source-OCT were recorded at each interval. Changes between V1 and V2 were analyzed using the Wilcoxon test and differences between the two groups of treatment were assessed using the Mann–Whitney test. Multiple regression analysis was performed to evaluate the possible OCT biomarker (CRT, ICR, CT, SND, HRS) as predictive factors for final visual acuity improvement. Results: In both groups, BCVA improved (p-value < 0.0001), and a significant reduction in central retinal thickness, intra-retinal cysts, red dots, hyper-reflective spots (HRS), and serous detachment of neuro-epithelium (SDN) was observed. A superiority of dexamethasone over ranibizumab in reducing the SDN height (p-value = 0.03) and HRS (p-value = 0.01) was documented. Conclusions: Ranibizumab and dexamethasone are effective in the treatment of DME, as demonstrated by functional improvement and morphological biomarker change. DME associated with SDN and HRS represents a specific inflammatory pattern for which dexamethasone appears to be more effective.

Prospective, Single-Center, Six-Month Study of Intravitreal Ranibizumab for Macular Edema with Nonproliferative Diabetic Retinopathy: Effects on Microaneurysm Turnover and Non-Perfused Retinal Area

Purpose

To analyze the effects on microaneurysm (MA) and perifoveal perfusion in nonproliferative diabetic retinopathy (NPDR) patients with macular edema (ME) after early intensive treatment using intravitreal ranibizumab (IVR) injections.

Patients and Methods

Prospectively, 25 eyes of 25 type 2 diabetes mellitus patients with ME were included between August 2016 and February 2019. For 6 months, patients were administered 0.5-mg IVR injections monthly. Ocular evaluation, including best-corrected visual acuity (BCVA; using the Early Treatment Diabetic Retinopathy Study chart), central retinal thickness (CRT; using optical coherence tomography), fundus photography, and fluorescein angiography, was performed for all participants. Results obtained at baseline were compared to those observed after 6 months.

Results

Mean BCVA increased significantly from 67.6±3.29 letters at baseline to 76.36±1.61 letters after 6 months (P=0.002) of IVR therapy. CRT decreased significantly from 479.12±16.66 µm at baseline to 369.12±13.02 µm at 6 months. Similarly, the total number of MAs decreased significantly from 5.68±3.41 to 1.60±1.73 (P<0.0001). MA turnover, calculated by adding the MA formation rate to the MA disappearance rate (both calculated as MA number/month) also decreased significantly from 6.88±3.83 to 1.92±1.75 after treatment (P<0.0001). Perifoveal non-perfused area decreased from 2.517±0.456 mm² at baseline to 2.495±0.293 mm² at 6 months, but the results were not statistically significant (P=0.954).

Conclusion

Treatment with early intensive IVR therapy in NPDR patients with ME not only improved BCVA and CRT but also decreased MA turnover. However, in the study period of 6 months, IVR therapy did not show significant improvement in perifoveal non-perfused area.

Microaneurysms detection in color fundus images using machine learning based on directional local contrast

Background

As one of the major complications of diabetes, diabetic retinopathy (DR) is a leading cause of visual impairment and blindness due to delayed diagnosis and intervention. Microaneurysms appear as the earliest symptom of DR. Accurate and reliable detection of microaneurysms in color fundus images has great importance for DR screening.

Methods

A microaneurysms' detection method using machine learning based on directional local contrast (DLC) is proposed for the early diagnosis of DR. First, blood vessels were enhanced and segmented using improved enhancement function based on analyzing eigenvalues of Hessian matrix. Next, with blood vessels excluded, microaneurysm candidate regions were obtained using shape characteristics and connected components analysis. After image segmented to patches, the features of each microaneurysm candidate patch were extracted, and each candidate patch was classified into microaneurysm or non-microaneurysm. The main contributions of our study are (1) making use of directional local contrast in microaneurysms' detection for the first time, which does make sense for better microaneurysms' classification. (2) Applying three different machine learning techniques for classification and comparing their performance for microaneurysms' detection. The proposed algorithm was trained and tested on e-ophtha MA database, and further tested on another independent DIARETDB1 database. Results of microaneurysms' detection on the two databases were evaluated on lesion level and compared with existing algorithms.

Results

The proposed method has achieved better performance compared with existing algorithms on accuracy and computation time. On e-ophtha MA and DIARETDB1 databases, the area under curve (AUC) of receiver operating characteristic (ROC) curve was 0.87 and 0.86, respectively. The free-response ROC (FROC) score on the two databases was 0.374 and 0.210, respectively. The computation time per image with resolution of 2544×1969, 1400×960 and 1500×1152 is 29 s, 3 s and 2.6 s, respectively.

Conclusions

The proposed method using machine learning based on directional local contrast of image patches can effectively detect microaneurysms in color fundus images and provide an effective scientific basis for early clinical DR diagnosis.

Practice Guidelines for Ocular Telehealth-Diabetic Retinopathy, Third Edition

Contributors The following document and appendices represent the third edition of the Practice Guidelines for Ocular Telehealth-Diabetic Retinopathy. These guidelines were developed by the Diabetic Retinopathy Telehealth Practice Guidelines Working Group. This working group consisted of a large number of subject matter experts in clinical applications for telehealth in ophthalmology. The editorial committee consisted of Mark B. Horton, OD, MD, who served as working group chair and Christopher J. Brady, MD, MHS, and Jerry Cavallerano, OD, PhD, who served as cochairs. The writing committees were separated into seven different categories. They are as follows: 1.Clinical/operational: Jerry Cavallerano, OD, PhD (Chair), Gail Barker, PhD, MBA, Christopher J. Brady, MD, MHS, Yao Liu, MD, MS, Siddarth Rathi, MD, MBA, Veeral Sheth, MD, MBA, Paolo Silva, MD, and Ingrid Zimmer-Galler, MD. 2.Equipment: Veeral Sheth, MD (Chair), Mark B. Horton, OD, MD, Siddarth Rathi, MD, MBA, Paolo Silva, MD, and Kristen Stebbins, MSPH. 3.Quality assurance: Mark B. Horton, OD, MD (Chair), Seema Garg, MD, PhD, Yao Liu, MD, MS, and Ingrid Zimmer-Galler, MD. 4.Glaucoma: Yao Liu, MD, MS (Chair) and Siddarth Rathi, MD, MBA. 5.Retinopathy of prematurity: Christopher J. Brady, MD, MHS (Chair) and Ingrid Zimmer-Galler, MD. 6.Age-related macular degeneration: Christopher J. Brady, MD, MHS (Chair) and Ingrid Zimmer-Galler, MD. 7.Autonomous and computer assisted detection, classification and diagnosis of diabetic retinopathy: Michael Abramoff, MD, PhD (Chair), Michael F. Chiang, MD, and Paolo Silva, MD.

Automated and Computer-Assisted Detection, Classification, and Diagnosis of Diabetic Retinopathy

Background: The introduction of artificial intelligence (AI) in medicine has raised significant ethical, economic, and scientific controversies. Introduction: Because an explicit goal of AI is to perform processes previously reserved for human clinicians and other health care personnel, there is justified concern about the impact on patient safety, efficacy, equity, and liability. Discussion: Systems for computer-assisted and fully automated detection, triage, and diagnosis of diabetic retinopathy (DR) from retinal images show great variation in design, level of autonomy, and intended use. Moreover, the degree to which these systems have been evaluated and validated is heterogeneous. We use the term DR AI system as a general term for any system that interprets retinal images with at least some degree of autonomy from a human grader. We put forth these standardized descriptors to form a means to categorize systems for computer-assisted and fully automated detection, triage, and diagnosis of DR. The components of the categorization system include level of device autonomy, intended use, level of evidence for diagnostic accuracy, and system design. Conclusion: There is currently minimal empirical basis to assert that certain combinations of autonomy, accuracy, or intended use are better or more appropriate than any other. Therefore, at the current stage of development of this document, we have been descriptive rather than prescriptive, and we treat the different categorizations as independent and organized along multiple axes.

How and why SGLT2 inhibitors should be explored as potential treatment option in diabetic retinopathy: clinical concept and methodology

Patients suffering from type 2 diabetes are at an increased risk of developing classical microvascular complications such as retinopathy, neuropathy, and nephropathy, which represent a significant health burden. Tight control of blood glucose, blood pressure, and serum cholesterol reduce the risk of microvascular complications but effective pharmacologically targeted treatment options for the treatment and prevention of diabetic microangiopathy are still lacking. Pharmacological inhibition of sodium glucose cotransporter 2 (SGLT2) might have the potential to directly protect against microvascular complications and could represent a potential treatment option. Randomized controlled clinical proof of concept trials are needed to investigate a potential central role of SGLT2 inhibitors in the prevention of diabetic microangiopathy and its classical clinical complications of retinopathy, neuropathy, and nephropathy.

Artificial intelligence for diabetic retinopathy screening: a review

Diabetes is a global eye health issue. Given the rising in diabetes prevalence and ageing population, this poses significant challenge to perform diabetic retinopathy (DR) screening for these patients. Artificial intelligence (AI) using machine learning and deep learning have been adopted by various groups to develop automated DR detection algorithms. This article aims to describe the state-of-art AI DR screening technologies that have been described in the literature, some of which are already commercially available. All these technologies were designed using different training datasets and technical methodologies. Although many groups have published robust diagnostic performance of the AI algorithms for DR screening, future research is required to address several challenges, for examples medicolegal implications, ethics, and clinical deployment model in order to expedite the translation of these novel technologies into the healthcare setting.

Longitudinal Relationship Between Retinal Diabetic Neurodegeneration and Progression of Diabetic Retinopathy in Patients With Type 2 Diabetes

PURPOSE

To investigate the longitudinal relationship between diabetic retinal neurodegeneration and the progression of diabetic retinopathy (DR) by measuring macular ganglion cell-inner plexiform layer (mGCIPL) thickness in patients with type 2 diabetes (T2DM).

DESIGN

Retrospective cohort study.

METHODS

T2DM patients with no DR or mild nonproliferative DR (NPDR) followed up for ≥4 years were included in this study. DR was graded according to retinal photography, and mean parafoveal mGCIPL thickness was measured using optical coherence tomography with at least a 6-month interval from baseline. Hazard ratios (HR) for predicting 2-step progression and development of proliferative DR (PDR) were calculated using Cox proportional hazard modeling using baseline clinical factors.

RESULTS

Of 87 eyes of T2DM patients, 39 (44.8%) exhibited 2-step DR progression and 6 (6.9%) experienced progression to PDR. Patients with DR progression exhibited longer T2DM duration, thinner mGCIPL, greater mGCIPL thinning rate, severe cardiac autonomic neuropathy (CAN), lower peripheral nerve-conduction velocity, and higher glycated hemoglobin A1c level. Multivariate regression modeling revealed that baseline mGCIPL thickness (HR = 0.94), mGCIPL thinning rate (HR = 1.924), CAN score (HR = 1.248), and conduction velocity of peripheral nerves (HR = 0.894) were significant predictive factors for DR progression (area under the curve = 0.92).

CONCLUSION

Progressive loss of mGCIPL is an independent risk factor for progression in early-stage DR. Further assessment of autonomic and peripheral nerve functions can increase sensitivity in predicting aggravation of DR in patients with T2DM.

Microaneurysm turnover is a predictor of diabetic retinopathy progression

Aim

To analyse retinopathy phenotypes and microaneurysm (MA) turnover in mild non-proliferative diabetic retinopathy (NPDR) as predictors of progression to diabetic central-involved macular oedema (CIMO) in patients with type 2 diabetes mellitus (DM) in two different ethnic populations.

Methods

205 patients with type 2 DM and mild NPDR were followed in a prospective observational study for 2 years or until development of CIMO, in two centres from different regions of the world. Ophthalmological examinations, including best-corrected visual acuity (BCVA), fundus photography with RetmarkerDR analysis, and optical coherence tomography (OCT), were performed at baseline and 6 12 and 24 months.

Results

158 eyes/patients reached either the study endpoint, CIMO (24) or performed the last study visit (24-month visit) without developing CIMO (134). From the eyes/patients in analysis, 27 eyes (17.1%) progressed to more advanced ETDRS (Early Treatment Diabetic Retinopathy Study) levels: 6 progressed to mild NPDR (level 35), 15 progressed to moderate NPDR (level 43), 5 progressed to moderately severe NPDR (level 47) and 1 progressed to high risk PDR (level 71). Worsening in ETDRS level is associated with phenotype C (p=0.005). From the 130 eyes/patients with a low MA turnover, 18 (13.8%) eyes/patients had an increase in ETDRS level, and from the 19 eyes/patients with a high MA turnover, 9 (47.4%) had an increase in ETDRS level (p<0.001).

Conclusion

Eyes in the initial stages of diabetic retinopathy show different phenotypes with different risks for progression to CIMO. In phenotype C, MA turnover correlates with ETDRS grading worsening and development of CIMO.

Deep learning applications in ophthalmology

PURPOSE OF REVIEW

To describe the emerging applications of deep learning in ophthalmology.

RECENT FINDINGS

Recent studies have shown that various deep learning models are capable of detecting and diagnosing various diseases afflicting the posterior segment of the eye with high accuracy. Most of the initial studies have centered around detection of referable diabetic retinopathy, age-related macular degeneration, and glaucoma.

SUMMARY

Deep learning has shown promising results in automated image analysis of fundus photographs and optical coherence tomography images. Additional testing and research is required to clinically validate this technology.

Subclinical Macular Edema as a Predictor of Progression to Central-Involved Macular Edema in Type 2 Diabetes

PURPOSE

The aim of this study was to examine the relationship between subclinical diabetic macular edema (SCME) and the development of central-involved macular edema (CIME) in patients with diabetes mellitus type-2 and mild nonproliferative diabetic retinopathy (NPDR), from 2 populations of different ethnicities.

METHODS

Two hundred and five patients with diabetes mellitus type-2 and mild NPDR with no prior laser or intravitreal treatment were followed for 2 years or until the development of CIME. Ophthalmological examinations, including BCVA, fundus photography with RetmarkerDR analysis, and optical coherence tomography were performed at baseline and months 6, 12, and 24.

RESULTS

One hundred and fifty eight eyes/patients reached either the study endpoint, CIME (n = 24), or performed the 24-month visit without developing CIME (n = 134). Fifty eyes/patients had SCME at baseline (31.6%). Of these 50 eyes, 16 (32.0%) developed CIME, whereas of the 108 eyes with normal retinal thickness (RT) at baseline, only 8 (7.4%) developed CIME (p < 0.001). Patients with increased RT in the central subfield at baseline showed a 12-fold risk of progression to CIME compared with patients without SCME.

CONCLUSIONS

In patients with mild NPDR, the presence of SCME is a good predictor of progression to CIME.

An observational study to assess if automated diabetic retinopathy image assessment software can replace one or more steps of manual imaging grading and to determine their cost-effectiveness

BACKGROUND

Diabetic retinopathy screening in England involves labour-intensive manual grading of retinal images. Automated retinal image analysis systems (ARIASs) may offer an alternative to manual grading.

OBJECTIVES

To determine the screening performance and cost-effectiveness of ARIASs to replace level 1 human graders or pre-screen with ARIASs in the NHS diabetic eye screening programme (DESP). To examine technical issues associated with implementation.

DESIGN

Observational retrospective measurement comparison study with a real-time evaluation of technical issues and a decision-analytic model to evaluate cost-effectiveness.

SETTING

A NHS DESP.

PARTICIPANTS

Consecutive diabetic patients who attended a routine annual NHS DESP visit.

INTERVENTIONS

Retinal images were manually graded and processed by three ARIASs: iGradingM (version 1.1; originally Medalytix Group Ltd, Manchester, UK, but purchased by Digital Healthcare, Cambridge, UK, at the initiation of the study, purchased in turn by EMIS Health, Leeds, UK, after conclusion of the study), Retmarker (version 0.8.2, Retmarker Ltd, Coimbra, Portugal) and EyeArt (Eyenuk Inc., Woodland Hills, CA, USA). The final manual grade was used as the reference standard. Arbitration on a subset of discrepancies between manual grading and the use of an ARIAS by a reading centre masked to all grading was used to create a reference standard manual grade modified by arbitration.

MAIN OUTCOME MEASURES

Screening performance (sensitivity, specificity, false-positive rate and likelihood ratios) and diagnostic accuracy [95% confidence intervals (CIs)] of ARIASs. A secondary analysis explored the influence of camera type and patients' ethnicity, age and sex on screening performance. Economic analysis estimated the cost per appropriate screening outcome identified.

RESULTS

A total of 20,258 patients with 102,856 images were entered into the study. The sensitivity point estimates of the ARIASs were as follows: EyeArt 94.7% (95% CI 94.2% to 95.2%) for any retinopathy, 93.8% (95% CI 92.9% to 94.6%) for referable retinopathy and 99.6% (95% CI 97.0% to 99.9%) for proliferative retinopathy; and Retmarker 73.0% (95% CI 72.0% to 74.0%) for any retinopathy, 85.0% (95% CI 83.6% to 86.2%) for referable retinopathy and 97.9% (95% CI 94.9 to 99.1%) for proliferative retinopathy. iGradingM classified all images as either 'disease' or 'ungradable', limiting further iGradingM analysis. The sensitivity and false-positive rates for EyeArt were not affected by ethnicity, sex or camera type but sensitivity declined marginally with increasing patient age. The screening performance of Retmarker appeared to vary with patient's age, ethnicity and camera type. Both EyeArt and Retmarker were cost saving relative to manual grading either as a replacement for level 1 human grading or used prior to level 1 human grading, although the latter was less cost-effective. A threshold analysis testing the highest ARIAS cost per patient before which ARIASs became more expensive per appropriate outcome than human grading, when used to replace level 1 grader, was Retmarker £3.82 and EyeArt £2.71 per patient.

LIMITATIONS

The non-randomised study design limited the health economic analysis but the same retinal images were processed by all ARIASs in this measurement comparison study.

CONCLUSIONS

Retmarker and EyeArt achieved acceptable sensitivity for referable retinopathy and false-positive rates (compared with human graders as reference standard) and appear to be cost-effective alternatives to a purely manual grading approach. Future work is required to develop technical specifications to optimise deployment and address potential governance issues.

FUNDING

The National Institute for Health Research (NIHR) Health Technology Assessment programme, a Fight for Sight Grant (Hirsch grant award) and the Department of Health's NIHR Biomedical Research Centre for Ophthalmology at Moorfields Eye Hospital and the University College London Institute of Ophthalmology.

Biomarkers of Diabetic Retinopathy

Diabetic retinopathy (DR), a leading cause of acquired vision loss, is a microvascular complication of diabetes. While traditional risk factors for diabetic retinopathy including longer duration of diabetes, poor blood glucose control, and dyslipidemia are helpful in stratifying patient's risk for developing retinopathy, many patients without these traditional risk factors develop DR; furthermore, there are persons with long diabetes duration who do not develop DR. Thus, identifying biomarkers to predict DR or to determine therapeutic response is important. A biomarker can be defined as a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention. Incorporation of biomarkers into risk stratification of persons with diabetes would likely aid in early diagnosis and guide treatment methods for those with DR or with worsening DR. Systemic biomarkers of DR include serum measures including genomic, proteomic, and metabolomics biomarkers. Ocular biomarkers including tears and vitreous and retinal vascular structural changes have also been studied extensively to prognosticate the risk of DR development. The current studies on biomarkers are limited by the need for larger sample sizes, cross-validation in different populations and ethnic groups, and time-efficient and cost-effective analytical techniques. Future research is important to explore novel DR biomarkers that are non-invasive, rapid, economical, and accurate to help reduce the incidence and progression of DR in people with diabetes.

Automated diabetic retinopathy imaging in Indian eyes: a pilot study

Aim:

To evaluate the efficacy of an automated retinal image grading system in diabetic retinopathy (DR) screening.

Materials and Methods:

Color fundus images of patients of a DR screening project were analyzed for the purpose of the study. For each eye two set of images were acquired, one centerd on the disk and the other centerd on the macula. All images were processed by automated DR screening software (Retmarker). The results were compared to ophthalmologist grading of the same set of photographs.

Results:

5780 images of 1445 patients were analyzed. Patients were screened into two categories DR or no DR. Image quality was high, medium and low in 71 (4.91%), 1117 (77.30%) and 257 (17.78%) patients respectively. Specificity and sensitivity for detecting DR in the high, medium and low group were (0.59, 0.91); (0.11, 0.95) and (0.93, 0.14).

Conclusion:

Automated retinal image screening system for DR had a high sensitivity in high and medium quality images. Automated DR grading software's hold promise in future screening programs.

Clinical biomarkers and molecular basis for optimized treatment of diabetic retinopathy: current status and future prospects

Diabetic retinopathy is a highly specific microvascular complication of diabetes and a leading cause of blindness worldwide. It is triggered by hyperglycemia which causes increased oxidative stress leading to an adaptive inflammatory assault to the neuroretinal tissue and microvasculature. Prolonged hyperglycemia causes increased polyol pathway flux, increased formation of advanced glycation end-products, abnormal activation of signaling cascades such as activation of protein kinase C (PKC) pathway, increased hexosamine pathway flux, and peripheral nerve damage. All these changes lead to increased oxidative stress and inflammatory assault to the retina resulting in structural and functional changes. In addition, neuroretinal alterations affect diabetes progression. The most effective way to manage diabetic retinopathy is by primary prevention such as hyperglycemia control. While the current mainstay for the management of severe and proliferative diabetic retinopathy is laser photocoagulation, its role is diminishing with the development of newer drugs including corticosteroids, antioxidants, and antiangiogenic and anti-VEGF agents which work as an adjunct to laser therapy or independently. The current pharmacotherapy of diabetic retinopathy is incomplete as a sole treatment option in view of limited efficacy and short-term effect. There is a definite clinical need to develop new pharmacological therapies for diabetic retinopathy, particularly ones which would be effective through the oral route and help recover lost vision. The increasing understanding of the mechanisms of diabetic retinopathy and its biomarkers is likely to help generate better and more effective medications.

Automated Diabetic Retinopathy Screening and Monitoring Using Retinal Fundus Image Analysis

BACKGROUND

Diabetic retinopathy (DR)-a common complication of diabetes-is the leading cause of vision loss among the working-age population in the western world. DR is largely asymptomatic, but if detected at early stages the progression to vision loss can be significantly slowed. With the increasing diabetic population there is an urgent need for automated DR screening and monitoring. To address this growing need, in this article we discuss an automated DR screening tool and extend it for automated estimation of microaneurysm (MA) turnover, a potential biomarker for DR risk.

METHODS

The DR screening tool automatically analyzes color retinal fundus images from a patient encounter for the various DR pathologies and collates the information from all the images belonging to a patient encounter to generate a patient-level screening recommendation. The MA turnover estimation tool aligns retinal images from multiple encounters of a patient, localizes MAs, and performs MA dynamics analysis to evaluate new, persistent, and disappeared lesion maps and estimate MA turnover rates.

RESULTS

The DR screening tool achieves 90% sensitivity at 63.2% specificity on a data set of 40 542 images from 5084 patient encounters obtained from the EyePACS telescreening system. On a subset of 7 longitudinal pairs the MA turnover estimation tool identifies new and disappeared MAs with 100% sensitivity and average false positives of 0.43 and 1.6 respectively.

CONCLUSIONS

The presented automated tools have the potential to address the growing need for DR screening and monitoring, thereby saving vision of millions of diabetic patients worldwide.

Retinal Imaging Techniques for Diabetic Retinopathy Screening

Due to the increasing prevalence of diabetes mellitus, demand for diabetic retinopathy (DR) screening platforms is steeply increasing. Early detection and treatment of DR are key public health interventions that can greatly reduce the likelihood of vision loss. Current DR screening programs typically employ retinal fundus photography, which relies on skilled readers for manual DR assessment. However, this is labor-intensive and suffers from inconsistency across sites. Hence, there has been a recent proliferation of automated retinal image analysis software that may potentially alleviate this burden cost-effectively. Furthermore, current screening programs based on 2-dimensional fundus photography do not effectively screen for diabetic macular edema (DME). Optical coherence tomography is becoming increasingly recognized as the reference standard for DME assessment and can potentially provide a cost-effective solution for improving DME detection in large-scale DR screening programs. Current screening techniques are also unable to image the peripheral retina and require pharmacological pupil dilation; ultra-widefield imaging and confocal scanning laser ophthalmoscopy, which address these drawbacks, possess great potential. In this review, we summarize the current DR screening methods using various retinal imaging techniques, and also outline future possibilities. Advances in retinal imaging techniques can potentially transform the management of patients with diabetes, providing savings in health care costs and resources.

Effects of sub-Tenon's capsule triamcinolone acetonide injection combined with microaneurysm photocoagulation on diabetic macular edema

Purpose

To compare the effects of sub-Tenon’s capsule triamcinolone acetonide (STTA) injection to that of STTA injection combined with microaneurysm photocoagulation (MAPC; STTA + MAPC) on eyes with diabetic macular edema (DME).

Patients and methods

The medical records of 138 eyes of 138 patients with DME treated by either STTA or STTA + MAPC were reviewed. The degree of DME was determined by the optical coherence tomographic features: patients with serous retinal detachment (SRD+; 38 eyes) and patients without SRD (non-SRD; 100 eyes). The central macular thickness (CMT) and the best-corrected visual acuity (BCVA) were measured periodically for 6 months after the treatments.

Results

The BCVA was significantly improved in the non-SRD group after STTA + MAPC. The CMT was significantly improved in all groups and improved considerably more in the non-SRD group than in the SRD+ group after STTA + MAPC.

Conclusion

Our findings indicate that MAPC has an additive effect in the non-SRD type.

Risk Factors for Refractory Diabetic Macular Oedema after Sub-Tenon's Capsule Triamcinolone Acetonide Injection

The purpose of this study is to identify the risk factors for a recurrence or persistence of diabetic macular oedema (DME) after a sub-Tenon's capsule triamcinolone acetonide (STTA) injection. The medical records of 124 patients (124 eyes) treated by STTA were reviewed. The age, sex, HbA1c level, best-corrected visual acuity, central macular thickness, insulin use, pioglitazone use, systemic hypertension, serous retinal detachment, proteinuria, panretinal photocoagulation, microaneurysm photocoagulation (MAPC), subthreshold micropulse diode laser photocoagulation (SMDLP), cataract surgery, and history of vitrectomy were examined by logistic regression analysis. Procedures of MAPC and SMDLP were significantly associated with DME treated with STTA (P = 0.0315, P = 0.04, resp.). However, a history of vitrectomy was found to have significantly fewer recurrences or persistent DME after STTA (P = 0.0464). In conclusion, patients who required combined MAPC or SMDLP with a STTA injection had significantly higher refractoriness to STTA, but postvitrectomy may prevent the recurrence or persistence of DME after STTA injection.

Compound Danshen Dripping Pill for Treating Early Diabetic Retinopathy: A Randomized, Double-Dummy, Double-Blind Study

This randomized, double-dummy, double-blind study was to observe the therapeutic effects of compound Danshen dripping pill (CDDP) in treating early diabetic retinopathy (DR). All the 57 type 2 diabetes cases in nonproliferative diabetic retinopathy (NPDR) stage were divided into two groups randomly: 28 cases treated with CDDP as the treated group and 29 cases treated with calcium dobesilate as the control group. The best corrected visual acuity (BCVA) in the treated group was significantly improved after treatment when compared to that before treatment (P < 0.05). Mean defect (MD) of visual field, hemorrhage area of the fundus, microaneurysm number, fluorescent leakage area, and capillary nonperfusion area evaluated by visual field, fundus photography, and fundus fluorescein angiography in the treated group had the same results as BCVA. However, there was no statistical difference in each index between the two groups. No obvious adverse events with clinical significance occurred. Our present study showed that CDDP has a similar improvement and safety to calcium dobesilate for NPDR. In future DR treatments, CDDP may function as the auxiliary drug.

Optical Coherence Tomography Angiography in Diabetic Retinopathy: A Prospective Pilot Study

PURPOSE

To evaluate how optical coherence tomography (OCT) angiography depicts clinical fundus findings in patients with diabetic retinopathy (DR).

DESIGN

Prospective study evaluating imaging technology.

METHODS

Forty-seven eyes of 25 patients with DR were scanned using a high-speed 840-nm-wavelength spectral-domain optical coherence tomography instrument (RTVue XR Avanti; Optovue, Inc, Fremont, California, USA). Blood flow was detected using the split-spectrum amplitude-decorrelation angiography algorithm. Fluorescein angiography (FA) images were also obtained in all eyes and the ability to visualize microaneurysms, retinal nonperfused areas, and neovascularization was compared with that of the en face OCT angiograms.

RESULTS

In 42 eyes, microaneurysms detected by FA near the macula appeared as focally dilated saccular or fusiform capillaries on OCT angiograms of the superficial and/or deep capillary plexus. Retinal nonperfused areas visualized by FA appeared as lesions with no or sparse capillaries on OCT angiograms. Area measurement of retinal nonperfusion near the macula in 7 eyes revealed a difference between the extent of nonperfused areas in superficial and deep plexuses. In 4 eyes, the vascular structures of neovascularization at the optic disc were clearly visualized on OCT angiograms. Decreases and re-increases of flow in new vessels were quantified in an eye treated with anti-vascular endothelial growth factor.

CONCLUSIONS

OCT angiography can clearly visualize microaneurysms and retinal nonperfused areas and enables closer observation of each layer of the retinal capillaries. Quantitative information on new vessels can also be obtained. OCT angiography may be clinically useful to evaluate the microvascular status and therapeutic effect of treatments for DR.

The clinical implications of recent studies on the structure and function of the retinal microvasculature in diabetes

The retinal blood vessels provide the opportunity to study early structural and functional changes in the microvasculature prior to clinically significant microvascular and macrovascular complications of diabetes. Advances in digital retinal photography and computerised assessment of the retinal vasculature have provided more objective and precise measurements of retinal vascular changes. Clinic- and population-based studies have reported that these quantitatively measured retinal vascular changes (e.g. retinal arteriolar narrowing and venular widening) are associated with preclinical structural changes in other microvascular systems (e.g. infarct in the cerebral microcirculation), as well as diabetes and diabetic complications, suggesting that they are markers of early microvascular dysfunction. In addition, there are new retinal imaging techniques to further assess alterations in retinal vascular function (e.g. flicker-induced vasodilatory response, blood flow and oxygen saturation) in diabetes and complications that result from the effects of chronic hyperglycaemia, inflammation and endothelial dysfunction. In this review, we summarise the latest findings on the relationships between quantitatively measured structural and functional retinal vascular changes with diabetes and diabetic complications. We also discuss clinical implications and future research to evaluate whether detection of retinal vascular changes has additional value beyond that achieved with methods currently used to stratify the risk of diabetes and its complications.

Imbalance of the nerve growth factor and its precursor as a potential biomarker for diabetic retinopathy

Our previous studies have demonstrated that diabetes-induced oxidative stress alters homeostasis of retinal nerve growth factor (NGF) resulting in accumulation of its precursor, proNGF, at the expense of NGF which plays a critical role in preserving neuronal and retinal function. This imbalance coincided with retinal damage in experimental diabetes. Here we test the hypothesis that alteration of proNGF and NGF levels observed in retina and vitreous will be mirrored in serum of diabetic patients. Blood and vitreous samples were collected from patients (diabetic and nondiabetic) undergoing vitrectomy at Georgia Regents University under approved IRB. Levels of proNGF, NGF, and p75^NTR shedding were detected using Western blot analysis. MMP-7 activity was also assayed. Diabetes-induced proNGF expression and impaired NGF expression were observed in vitreous and serum. Vitreous and sera from diabetic patients (n = 11) showed significant 40.8-fold and 3.6-fold increases, respectively, compared to nondiabetics (n = 9). In contrast, vitreous and sera from diabetic patients showed significant 44% and 64% reductions in NGF levels, respectively, compared to nondiabetics. ProNGF to NGF ratios showed significant correlation between vitreous and serum. Further characterization of diabetes-induced imbalance in the proNGF to NGF ratio will facilitate its utility as an early biomarker for diabetic complications.

Automated retinal image analysis for diabetic retinopathy in telemedicine

There will be an estimated 552 million persons with diabetes globally by the year 2030. Over half of these individuals will develop diabetic retinopathy, representing a nearly insurmountable burden for providing diabetes eye care. Telemedicine programmes have the capability to distribute quality eye care to virtually any location and address the lack of access to ophthalmic services. In most programmes, there is currently a heavy reliance on specially trained retinal image graders, a resource in short supply worldwide. These factors necessitate an image grading automation process to increase the speed of retinal image evaluation while maintaining accuracy and cost effectiveness. Several automatic retinal image analysis systems designed for use in telemedicine have recently become commercially available. Such systems have the potential to substantially improve the manner by which diabetes eye care is delivered by providing automated real-time evaluation to expedite diagnosis and referral if required. Furthermore, integration with electronic medical records may allow a more accurate prognostication for individual patients and may provide predictive modelling of medical risk factors based on broad population data.

Microaneurysm count as a predictor of long-term progression in diabetic retinopathy in young patients with type 1 diabetes: the Danish Cohort of Pediatric Diabetes 1987 (DCPD1987)

PURPOSE

To investigate microaneurysm (MA) count as a predictor of long-term progression of diabetic retinopathy (DR) in young patients with type 1 diabetes mellitus (T1DM).

METHODS

We examined 185 patients with T1DM at baseline (1995) and at follow-up (2011). At baseline, mean age and duration of diabetes were 20.6 and 12.9 years, respectively. Two-field (1995) and seven-field (2011) fundus photographs were taken in accordance with the European Diabetes Study Group (EURODIAB) and the Early Treatment Diabetic Retinopathy Study (ETDRS) protocol, respectively. DR was graded in accordance to the ETDRS protocol, allowing for non-standard photography at baseline. Baseline MAs were counted; patients without DR and those with MAs only were included. Multivariable logistic regressions were performed to investigate MA-count as a predictor of two-step progression, progression to proliferative DR (PDR), and incident diabetic macula edema (DME).

RESULTS

We included 138 patients (138 eyes). Of these, 58 had no retinopathy and 80 had MAs only. At follow-up, rates of two-step progression of DR, progression to PDR and incident DME were 52.9, 21.7, and 10.1 %, respectively. In logistic regression models, MA count was able to predict progression to PDR (OR: 1.51 per MA; 95 % CI: [1.04-2.20]) and DME (OR: 1.69 per MA; 95 % CI: [1.05-2.77]), but not two-step progression (OR 0.91 per MA, 95 % CI: [0.64-1.31]).

CONCLUSIONS

In younger patients with T1DM, MA count predicts long-term incidence of PDR and DME. This demonstrates that early DR is a warning sign of late retinopathy complications and that the number of MAs is an important factor for long-term outcome.

Phenotypes and biomarkers of diabetic retinopathy

Diabetic retinopathy (DR) remains a major cause of blindness as the prevalence of diabetes is expected to approximately double globally between 2000 and 2030. DR progresses over time at different rates in different individuals with only a limited number developing significant vision loss due to the two major vision-threatening complications, clinically significant macular edema and proliferative retinopathy. Good metabolic control is important to prevent and delay progression, but whereas some patients escape vision loss even with poor control, others develop vision loss despite good metabolic control. Our research group has been able to identify three different DR phenotypes characterized by different dominant retinal alterations and different risks of progression to vision-threatening complications. Microaneurysm turnover has been validated as a prognostic biomarker of development of clinically significant macular edema, whereas subclinical macular edema identified by OCT and mfERG appear to be also good candidates as organ-specific biomarkers of DR. Hemoglobin A1c remains the only confirmed systemic prognostic biomarker of DR progression. The availability of biomarkers of DR progression and the identification of different phenotypes of DR with different risks for development of vision-threatening complications offers new perspectives for understanding DR and for its personalized management.