Dual Consistency Enabled Weakly and Semi-Supervised Optic Disc and Cup Segmentation With Dual Adaptive Graph Convolutional Networks.

Glaucoma is a progressive eye disease that results in permanent vision loss, and the vertical cup to disc ratio (vCDR) in colour fundus images is essential in glaucoma screening and assessment. Previous fully supervised convolution neural networks segment the optic disc (OD) and optic cup (OC) from color fundus images and then calculate the vCDR offline. However, they rely on a large set of labeled masks for training, which is expensive and time-consuming to acquire. To address this, we propose a weakly and semi-supervised graph-based network that investigates geometric associations and domain knowledge between segmentation probability maps (PM), modified signed distance function representations (mSDF), and boundary region of interest characteristics (B-ROI) in three aspects. Firstly, we propose a novel Dual Adaptive Graph Convolutional Network (DAGCN) to reason the long-range features of the PM and the mSDF w.r.t. the regional uniformity. Secondly, we propose a dual consistency regularization-based semi-supervised learning paradigm. The regional consistency between the PM and the mSDF, and the marginal consistency between the derived B-ROI from each of them boost the proposed model's performance due to the inherent geometric associations. Thirdly, we exploit the task-specific domain knowledge via the oval shapes of OD & OC, where a differentiable vCDR estimating layer is proposed. Furthermore, without additional annotations, the supervision on vCDR serves as weakly-supervisions for segmentation tasks. Experiments on six large-scale datasets demonstrate our model's superior performance on OD & OC segmentation and vCDR estimation. The implementation code has been made available.https://github.com/smallmax00/Dual_Adaptive_Graph_Reasoning.

Prevalence and risk factors for age-related macular degeneration in a population-based cohort study of older adults in Northern Ireland using multimodal imaging: NICOLA Study.

PURPOSE: To report prevalence and risk factor associations for age-related macular degeneration (AMD) and AMD features from multimodal retinal grading in a multidisciplinary longitudinal population-based study of aging in Northern Ireland. STUDY DESIGN: Population-based longitudinal cohort study. METHODS: Retinal imaging at the Norther Ireland Cohort for the Longitudinal Aging Study health assessment included stereo Colour Fundus Photography (CFP) (Canon CX-1, Tokyo, Japan) and Spectral-Domain Optical Coherence Tomography (SD-OCT) ((Heidelberg Retinal Angopgraph (HRA)+OCT; Heidelberg Engineering, Heidelberg, Germany). Medical history and demographic information was obtained during a home interview. Descriptive statistics were used to describe the prevalence of AMD and individual AMD features. Multiple imputation followed by multiple regression modelling was used to explore risk factor associations including relationships with AMD genetic risk score. RESULTS: Retinal images from 3386 participants were available for analysis. Mean age of the sample was 63.4 (SD 9.01, range: 36-99). Population weighted prevalence of AMD using colour grading in those over 55 years was: no drusen: 6 0.4%; drusen <63 µm: 15.9%; drusen 63-125 µm: 13.7%; drusen >125 µm or pigmentary changes: 8.3%; late AMD: 1.6%. Prevalence of AMD features in those over 55 years was: OCT drusen 27.5%, complete outer retinal pigment epithelium and outer retinal atrophy (cRORA) on OCT was 4.3%, reticular drusen 3.2% and subretinal drusenoid deposits 25.7%. The genetic risk score was significantly associated with drusen and cRORA but less so for SDD alone and non-significant for hyperpigmentation or vitelliform lesions. CONCLUSIONS: Multimodal imaging-based classification has provided evidence of some divergence of genetic risk associations between classical drusen and SDD. Our findings support an urgent review of current AMD severity classification systems.

UK Biobank retinal imaging grading: methodology, baseline characteristics and findings for common ocular diseases.

BACKGROUND/OBJECTIVES: This study aims to describe the grading methods and baseline characteristics for UK Biobank (UKBB) participants who underwent retinal imaging in 2009-2010, and to characterise individuals with retinal features suggestive of age-related macular degeneration (AMD), glaucoma and retinopathy. METHODS: Non-mydriatic colour fundus photographs and macular optical coherence tomography (OCT) scans were manually graded by Central Administrative Research Facility certified graders and quality assured by clinicians of the Network of Ophthalmic Reading Centres UK. Captured retinal features included those associated with AMD (≥1 drusen, pigmentary changes, geographic atrophy or exudative AMD; either imaging modality), glaucoma (≥0.7 cup-disc ratio, ≥0.2 cup-disc ratio difference between eyes, other abnormal disc features; photographs only) and retinopathy (characteristic features of diabetic retinopathy with or without microaneurysms; either imaging modality). Suspected cases of these conditions were characterised with reference to diagnostic records, physical and biochemical measurements. RESULTS: Among 68,514 UKBB participants who underwent retinal imaging, the mean age was 57.3 years (standard deviation 8.2), 45.7% were men and 90.6% were of White ethnicity. A total of 64,367 participants had gradable colour fundus photographs and 68,281 had gradable OCT scans in at least one eye. Retinal features suggestive of AMD and glaucoma were identified in 15,176 and 2184 participants, of whom 125 (0.8%) and 188 (8.6%), respectively, had a recorded diagnosis. Of 264 participants identified to have retinopathy with microaneurysms, 251 (95.1%) had either diabetes or hypertension. CONCLUSIONS: This dataset represents a valuable addition to what is currently available in UKBB, providing important insights to both ocular and systemic health.

Long-term Retinal Morphology and Functional Associations in Treated Neovascular Age-Related Macular Degeneration: Findings from the Inhibition of VEGF in Age-Related Choroidal Neovascularisation Trial.

PURPOSE: To describe the frequency of long-term morphologic features and their relationships with visual function in participants who exited the Inhibition of VEGF in Age-Related Choroidal Neovascularisation (IVAN; ISRCTN92166560) trial. DESIGN: Multicenter cohort study up to 7 years after enrollment. PARTICIPANTS: Patients enrolled in the IVAN trial, excluding participants who died or withdrew during the trial. METHODS: Multimodal fundus images, best-corrected visual acuity (BCVA), and low-luminance visual acuity (LLVA) were obtained for a subset of 199 participants who attended a research visit. Clinical sites (n = 20) also provided all visual acuity and clinical information from usual care records for 532 participants and submitted the most recent color, OCT, and other fundus images for 468 participants to a reading center. MAIN OUTCOME MEASURES: Assessed the following from the most recent images: intralesional macular atrophy (ILMA) within the footprint of the neovascular lesion; hyperreflective material (HRM); intraretinal fluid (IRF); subretinal fluid (SRF); pigment epithelial detachment (PED); and disorganized retinal outer layers (DROLs). Cross-sectional relationships between morphologic features and BCVA/LLVA were estimated. RESULTS: Intralesional macular atrophy was present in 31.8% of the study eyes at IVAN exit (mean follow-up, 1.96 years) and 89.5% at the most recent imaging visit (mean follow-up, 6.18 years). Hyperreflective material, IRF, SRF, PED, and DROLs were present in 78.8%, 47.7%, 7.6%, 94.5%, and 55% of the study eyes, respectively. In the subset with complete imaging data, in eyes without DROL, the BCVA was worst in the thinnest outer fovea tertile (thinnest minus middle and thickest tertiles, -19.7 and -19.5 letters, respectively), whereas in eyes with DROL, the BCVA was worst in the thickest (thinnest and middle tertiles minus thickest, 12.5 and 12.2, respectively). Regression models showed that the presence of ILMA and HRM was independently associated with BCVA (22 letters worse [95% confidence interval {CI}, -11.2 to -32.8; P < 0.001] and 9.8 letters worse [95% CI, -0.1 to -19.4; P = 0.047], respectively). Subretinal fluid and foveal PED were associated with better BCVA (5.9 letters [95% CI, -7.9 to 19.7; P = 0.399] and 6.4 letters [95% CI, -1.1 to 14.0; P = 0.094], respectively). The model with LLVA was similar. A sensitivity analysis involving the entire eligible cohort yielded similar estimates. CONCLUSIONS: Macular atrophy and HRM were common after 7 years of follow-up and strongly associated with visual outcomes.

Assessment of the Vitreomacular Interface Using High-Resolution OCT in a Population-Based Cohort Study of Older Adults.

PURPOSE: To describe the prevalence of vitreomacular interface (VMI) features and their associated risk factors in the Northern Ireland Cohort for the Longitudinal Study of Ageing (NICOLA) Study. DESIGN: Cross-sectional population-based study. PARTICIPANTS: Noninstitutionalized Northern Irish adults 40 years of age or older. METHODS: Using geographic stratification, a representative sample of people in Northern Ireland was invited to participate in the NICOLA Study. SD OCT images of participants were graded for vitreomacular traction (VMT), macular hole (MH), and epiretinal membrane (ERM) according to the International Vitreomacular Traction Study Group. A subsample was graded in more detail to estimate the prevalence of VMA and VMA area detailing size and location of VMA. Descriptive analysis and risk factors for each VMI feature were determined using generalized estimating equations. Results were standardized to the Northern Ireland population census (2011). MAIN OUTCOME MEASURES: Cohort profile, standardized prevalence, and risk factor associations of each VMI feature. RESULTS: Three thousand three hundred fifty-one NICOLA participants had gradable SD OCT images available for at least 1 eye. The prevalence of VMT was 0.5% (CI, 0.31%-0.70%), that for MH was 0.3% (CI, 0.23%-0.52%), and that for ERM was 7.6% (CI, 7.0%-8.3%). A detailed VMA analysis was performed on a subsample consisting of the first 1481 participants. The prevalence of VMA was 22.6% (CI, 21.1-24.2), and VMA area ranged from 0.25 to 42.7 mm2 (mean, 12.53 mm2; standard deviation, 6.90 mm2). In multivariate analyses, increased age was associated with an increased odds ratio (OR) of VMT, MH, and ERM. VMA area was positively associated with younger age and normal blood pressure. ERM and MH were present more often in more myopic eyes, associated with an increase in levels of high-density lipoprotein (HDL) cholesterol and triglycerides. CONCLUSIONS: The epidemiologic characteristics of VMI features indicated that VMI interactions throughout life are age dependent. Vitreous separation reduced to a greater extent in the horizontal meridians compared with the vertical, differing from previous studies. Future longitudinal studies of the evolution of these VMI changes over time would be of great interest.

Intralesional Macular Atrophy in Anti-Vascular Endothelial Growth Factor Therapy for Age-Related Macular Degeneration in the IVAN Trial.

PURPOSE: To report on the development and progression of macular atrophy (MA) and its relationship with morphologic and functional measures in study and fellow eyes in the Inhibition of vascular endothelial growth factor (VEGF) in Age-related Choroidal Neovascularisation trial. DESIGN: Reading center analysis of data from a randomized controlled trial. PARTICIPANTS: Participants with previously untreated neovascular age-related macular degeneration (nAMD) in the study eye. METHODS: Color, fluorescein angiography (FA) and OCT images acquired at baseline and during the 2-year follow-up were graded systematically for presence of MA. Regression models were constructed to explore relationships between MA and lesion morphology and vision measures (best-corrected distance and near acuity, reading speed and index, contrast sensitivity). MAIN OUTCOME MEASURES: Primary outcome was development of intralesional MA (≥175 µm greatest linear dimension of choroidal vessels seen on FA and/or color, aided by OCT) lying within the maximum footprint of the neovascular lesion. RESULTS: Study eye data were available for 594 of 610 participants; 57 (9.6%) showed intralesional MA at baseline. Incident intralesional MA occurred in 24.4% by the final visit and extralesional MA in only 1.54%. In fellow eyes, an established nAMD lesion was present at baseline in 248 of whom 42 (16.9%) showed intralesional MA at baseline and 32 (12.9%) developed incident intralesional MA. The odds of incident intralesional MA by final visit were lower in study eyes that had ≥50% classic CNV at baseline (odds ratio [OR], 0.39; 95% confidence interval [CI], 0.19-0.80; P = 0.010), subretinal fluid at final visit (OR, 0.41; 95% CI, 0.25-0.76; P = 0.004), or pigment epithelial detachment at final visit (OR, 0.40; 95% CI, 0.21-0.74; P = 0.004). Secondary analyses of incident or progressed intralesional MA in study eyes supported these findings, with odds increasing if the fellow eye had baseline intralesional MA (OR, 2.43; 95% CI, 1.09-5.44; P = 0.030). No significant associations were observed between development of intralesional MA and any other morphologic or visual function measure. CONCLUSIONS: Macular atrophy frequently develops within an nAMD lesion in eyes receiving anti-VEGF therapy over 2 years. No associations between incident MA and drug or treatment frequency or visual function were detected, providing some reassurance to clinicians; however, the longer-term effects remain unknown.

Confocal infrared imaging with optical coherence tomography provides superior detection of a number of common macular lesions compared to colour fundus photography.

PURPOSE: To compare diagnostic accuracy of confocal infrared reflectance (IR), with and without optical coherence tomography (OCT), to colour fundus photography (CFP) in the Northern Ireland Cohort for the Longitudinal Study of Ageing (NICOLA) Study. METHODS: Cross-sectional observational study of participants in NICOLA. CFP, IR and IR/OCT of 640 eyes were graded for hard, soft and reticular pseudodrusen; geographic atrophy; choroidal neovascularisation; naevus; epiretinal membrane; and haemorrhages. Test characteristics (sensitivity and specificity) for each imaging modality with respect to each retinal feature were calculated. RESULTS: With CFP as the reference standard, sensitivity of IR by itself ranged from 75% for RPD to 93.5% for hard drusen and specificity was above 90% for all features except hard drusen (71.7%). For IR combined with OCT, sensitivity ranged from 80% for choroidal neovascularisation to 96.5% for hard drusen. When IR alone was the reference standard, CFP sensitivity was high for naevi (97.5%) but reduced markedly for epiretinal membrane (48.5%). When the combination of IR and OCT was the reference standard, sensitivity for CFP was least for epiretinal membrane (31.5%), low for geographic atrophy and reticular pseudodrusen (77.8% and 76.2% respectively) and high for all other lesion types. CONCLUSION: Our findings support the use of confocal IR with OCT as a screening tool for a variety of features of macular disease in community optometric practice.