Dataset Information

Methods and reproducibility of grading optimized digital color fundus photographs in the Age-Related Eye Disease Study 2 (AREDS2 Report Number 2).

ABSTRACT: To establish continuity with the grading procedures and outcomes from the historical data of the Age-Related Eye Disease Study (AREDS), color photographic imaging and evaluation procedures for the assessment of age-related macular degeneration (AMD) were modified for digital imaging in the AREDS2. The reproducibility of the grading of index AMD lesion components and for the AREDS severity scale was tested at the AREDS2 reading center.Digital color stereoscopic fundus photographs from 4203 AREDS2 subjects collected at baseline and annual follow-up visits were optimized for tonal balance and graded according to a standard protocol slightly modified from AREDS. The reproducibility of digital grading of AREDS2 images was assessed by reproducibility exercises, temporal drift (regrading a subset of baseline annually, n = 88), and contemporaneous masked regrading (ongoing, monthly regrade on 5% of submissions, n = 1335 eyes).In AREDS2, 91% and 96% of images received replicate grades within two steps of the baseline value on the AREDS severity scale for temporal drift and contemporaneous assessment, respectively (weighted Kappa of 0.73 and 0.76). Historical data for temporal drift in replicate gradings on the AREDS film-based images were 88% within two steps (weighted Kappa = 0.88). There was no difference in AREDS2-AREDS concordance for temporal drift (exact P = 0.57).Digital color grading has nearly the same reproducibility as historical film grading. There is substantial agreement for testing the predictive utility of the AREDS severity scale in AREDS2 as a clinical trial outcome. (ClinicalTrials.gov number, NCT00345176.)

SUBMITTER: Danis RP

PROVIDER: S-EPMC3706107 | biostudies-literature | 2013 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Methods and reproducibility of grading optimized digital color fundus photographs in the Age-Related Eye Disease Study 2 (AREDS2 Report Number 2).

Danis Ronald P RP Domalpally Amitha A Chew Emily Y EY Clemons Traci E TE Armstrong Jane J SanGiovanni John Paul JP Ferris Frederick L FL

Investigative ophthalmology & visual science 20130708 7

<h4>Purpose</h4>To establish continuity with the grading procedures and outcomes from the historical data of the Age-Related Eye Disease Study (AREDS), color photographic imaging and evaluation procedures for the assessment of age-related macular degeneration (AMD) were modified for digital imaging in the AREDS2. The reproducibility of the grading of index AMD lesion components and for the AREDS severity scale was tested at the AREDS2 reading center.<h4>Methods</h4>Digital color stereoscopic fun ...[more]

PMID: 23620429

Similar Datasets

Project description:PurposeThe Age-Related Eye Disease Study (AREDS) demonstrated beneficial effects of oral supplementation with antioxidant vitamins and minerals on the development of advanced age-related macular degeneration (AMD) in persons with at least intermediate AMD (bilateral large drusen with or without pigment changes). Observational data suggest that other oral nutrient supplements might further reduce the risk of progression to advanced AMD. The primary purpose of the Age-Related Eye Disease Study 2 (AREDS2) is to evaluate the efficacy and safety of lutein plus zeaxanthin (L+Z) and/or ω-3 long-chain polyunsaturated fatty acid (LCPUFA) supplementation in reducing the risk of developing advanced AMD. The study also assesses the reduction in zinc and the omission of β-carotene from original AREDS formulation.DesignMulticenter, phase III, randomized, controlled clinical trial.ParticipantsPersons aged 50 to 85 with bilateral intermediate AMD or advanced AMD in 1 eye.MethodsAll participants were randomly assigned to placebo (n = 1012), L+Z (10 mg/2 mg; n = 1044), ω-3 LCPUFAs (eicosapentaenoic acid + docosahexaenoic acid [650 mg/350 mg]; n = 1069), or the combination of L+Z and ω-3 LCPUFAs (n = 1078). All participants were offered a secondary randomization to 1 of 4 variations of the original AREDS formulation keeping vitamins C (500 mg) and E (400 IU) and copper (2 mg) unchanged while varying zinc and β-carotene as follows: Zinc remains at the original level (80 mg), lower only zinc to 25 mg, omit β-carotene only, or lower zinc to 25 mg and omit β-carotene.Main outcome measuresProgression to advanced AMD determined by centralized grading of annual fundus photographs.ResultsWe enrolled 4203 participants at 82 clinical centers located in the United States. Population characteristics at baseline were as follows: Mean age, 74 years; 57% female; 97% white; 7% current smokers; 19% with prior cardiovascular disease; and 44% and 50% taking statin-class cholesterol-lowering drugs and aspirin, respectively. Ocular characteristics include 59% with bilateral large drusen, 32% with advanced AMD in 1 eye and mean visual acuity of 20/32 in eyes without advanced AMD.ConclusionsThis report presents the AREDS2 study design and the participants' baseline demographic and ocular characteristics.

Project description:PURPOSE:To assess the utility of deep learning in the detection of geographic atrophy (GA) from color fundus photographs and to explore potential utility in detecting central GA (CGA). DESIGN:A deep learning model was developed to detect the presence of GA in color fundus photographs, and 2 additional models were developed to detect CGA in different scenarios. PARTICIPANTS:A total of 59?812 color fundus photographs from longitudinal follow-up of 4582 participants in the Age-Related Eye Disease Study (AREDS) dataset. Gold standard labels were from human expert reading center graders using a standardized protocol. METHODS:A deep learning model was trained to use color fundus photographs to predict GA presence from a population of eyes with no AMD to advanced AMD. A second model was trained to predict CGA presence from the same population. A third model was trained to predict CGA presence from the subset of eyes with GA. For training and testing, 5-fold cross-validation was used. For comparison with human clinician performance, model performance was compared with that of 88 retinal specialists. MAIN OUTCOME MEASURES:Area under the curve (AUC), accuracy, sensitivity, specificity, and precision. RESULTS:The deep learning models (GA detection, CGA detection from all eyes, and centrality detection from GA eyes) had AUCs of 0.933-0.976, 0.939-0.976, and 0.827-0.888, respectively. The GA detection model had accuracy, sensitivity, specificity, and precision of 0.965 (95% confidence interval [CI], 0.959-0.971), 0.692 (0.560-0.825), 0.978 (0.970-0.985), and 0.584 (0.491-0.676), respectively, compared with 0.975 (0.971-0.980), 0.588 (0.468-0.707), 0.982 (0.978-0.985), and 0.368 (0.230-0.505) for the retinal specialists. The CGA detection model had values of 0.966 (0.957-0.975), 0.763 (0.641-0.885), 0.971 (0.960-0.982), and 0.394 (0.341-0.448). The centrality detection model had values of 0.762 (0.725-0.799), 0.782 (0.618-0.945), 0.729 (0.543-0.916), and 0.799 (0.710-0.888). CONCLUSIONS:A deep learning model demonstrated high accuracy for the automated detection of GA. The AUC was noninferior to that of human retinal specialists. Deep learning approaches may also be applied to the identification of CGA. The code and pretrained models are publicly available at https://github.com/ncbi-nlp/DeepSeeNet.

Dataset Information

Methods and reproducibility of grading optimized digital color fundus photographs in the Age-Related Eye Disease Study 2 (AREDS2 Report Number 2).

Publications

Methods and reproducibility of grading optimized digital color fundus photographs in the Age-Related Eye Disease Study 2 (AREDS2 Report Number 2).

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets