Automated quantification of retinal vessel morphometry in the UK biobank cohort

Welikala, Roshan Alex; Fraz, Muhammad Moazam; Habib, M.M.; Daniel-Tong, S.; Yates, Max; Foster, Paul J.; Whincup, Peter H; Rudnicka, Alicja R.; Owen, Christopher G.; Strachan, David P.; Barman, Sarah

doi:10.1109/ipta.2017.8310108

Cited by 17 publications

(30 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Processing of retinal images was carried out using an automated computerized system (QUARTZ), as previously published in detail ( 23 ). In brief, the automated system distinguishes between right and left eyes (by optic disc localization) and venules and arterioles, identifies vessel segments, outputs centerline coordinates, and measures vessel width and angular change between vessel centerline coordinates, as well as provides further measures of tortuosity ( Figure 2 ) ( 23 ). These measures were summarized as a mean width per image, weighted by segment length, for arterioles and venules separately for each image.…”

Section: Methodsmentioning

confidence: 99%

“…These measures were summarized as a mean width per image, weighted by segment length, for arterioles and venules separately for each image. The following image processing modules were all validated on a subset of 4,692 retinal images from a random sample of 2,346 UK Biobank participants: vessel segmentation, image quality score, optic disc detection, vessel width measurement, tortuosity measurement, and arteriolar venular recognition ( 19 , 23 , 24 ). The performance of the arteriole/venule recognition had detection rates of up to 96% for arterioles and 98% for venules when the automated probability of artery or vein was set to a cutoff of 0.8.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Retinal Vascular Tortuosity and Diameter Associations with Adiposity and Components of Body Composition

Tapp

Owen

Barman

et al. 2020

Obesity

View full text Add to dashboard Cite

Objective The aim of this study was to assess whether adiposity or body composition relates to microvascular characteristics of the retina, indicative of cardiometabolic function. Methods A fully automated QUARTZ software processed retinal images from 68,550 UK Biobank participants (aged 40‐69 years). Differences in retinal vessel diameter and tortuosity with body composition measures from the Tanita analyzer were obtained by using multilevel regression analyses adjusted for age, sex, ethnicity, clinic, smoking, and Townsend deprivation index. Results Venular tortuosity and diameter increased by approximately 2% (P < 10−300) and 0.6 μm (P < 10−6), respectively, per SD increase in BMI, waist circumference index, waist‐hip ratio, total body fat mass index, and fat‐free mass index (FFMI). Venular associations with adiposity persisted after adjustment for FFMI, whereas associations with FFMI were weakened by FMI adjustment. Arteriolar diameter (not tortuosity) narrowing with FFMI was independent of adiposity (−0.6 μm; −0.7 to −0.4 μm per SD increment of FFMI), while adiposity associations with arteriolar diameter were largely nonsignificant after adjustment for FFMI. Conclusions This demonstrates, on an unprecedented scale, that venular tortuosity and diameter are more strongly associated with adiposity, whereas arteriolar diameter relates more strongly to fat‐free mass. Different attributes of the retinal microvasculature may reflect distinct roles of body composition and fatness on the cardiometabolic system.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Retinal Vascular Tortuosity and Diameter Associations with Adiposity and Components of Body Composition

Tapp

Owen

Barman

et al. 2020

Obesity

View full text Add to dashboard Cite

show abstract

“…A validated, fully automated AI-enabled system (QUARTZ) 18-20 extracted thousands of measures of retinal vessel width, tortuosity and area from the whole retinal image. Supervised machine learning techniques were used within QUARTZ; with a support vector machine used to create an image quality score 18 and deep learning was used to develop an algorithm to distinguish between arterioles and venules.…”

Section: Methodsmentioning

confidence: 99%

“…We developed a fully automated AI-enabled system (QUARTZ) for examining the retinal vascular tree, which overcomes many of the difficulties of earlier approaches, allowing detailed vasculometry quantification in large population studies. [18][19][20] In the subset of UK Biobank who underwent retinal imaging, 21 and in the EPIC-Norfolk 16 cohorts, we examine detailed characterisation of RV as a non-invasive maker of vascular health in relation to circulatory mortality prediction. In addition, we provide findings for Framingham risk scores for stroke, 22 and MI 23 in the same subset that underwent retinal imaging, and assess the incremental value of adding RV to Framingham risk scores for incident stroke and MI.…”

Section: Introductionmentioning

confidence: 99%

Artificial intelligence enabled retinal vasculometry for prediction of circulatory mortality, myocardial infarction and stroke

Rudnicka

Welikala

Barman

et al. 2022

Preprint

View full text Add to dashboard Cite

AimsWe examine whether inclusion of Artificial Intelligence (AI)-enabled retinal vasculometry (RV) improves existing risk algorithms for incident stroke, myocardial infarction (MI) and circulatory mortality.MethodsAI-enabled retinal vessel image analysis processed images from 88,052 UK Biobank (UKB) participants (aged 40-69 years at image capture) and 7,411 EPIC-Norfolk participants (aged 48-92). Retinal arteriolar and venular width, tortuosity and area were extracted. Prediction models were developed in UKB using multivariable Cox proportional hazards regression for circulatory mortality, incident stroke and MI, and externally validated in EPIC-Norfolk. Model performance was assessed using optimism adjusted calibration, C- and R2 statistics. Performance of Framingham risk scores (FRS) for incident stroke and incident MI, with addition of RV to FRS, were compared with a simpler model based on RV, age, smoking status and medical history (antihypertensive/cholesterol lowering medication, diabetes, prevalent stroke/MI).ResultsUKB prognostic models were developed on 65,144 participants (mean age 56.8; median follow-up 7.7 years) and validated in 5,862 EPIC-Norfolk participants (67.6, 9.1 years respectively). Prediction models for circulatory mortality in men and women had optimism adjusted C- and R2 statistics between 0.75-0.77 and 0.33-0.44 respectively. For incident stroke and MI, addition of RV to FRS did not improve model performance in either cohort. However, the simpler RV model performed equally or better than FRS.ConclusionRV offers an alternative predictive biomarker to traditional risk-scores for vascular health, without the need for blood sampling or blood pressure measurement. Further work is needed to examine RV in population screening to triage individuals at high-risk. (250 words)HIGHLIGHTSWhat is already known on this topicPopulation screening for MI and stroke using risk prediction tools exist but have limited uptake; risk scores for circulator mortality do not exist.What this study addsRisk models developed in UK Biobank (validated in EPIC-Norfolk) using Artificial Intelligence (AI)-enabled retinal vasculometry (RV), age, history of cardiovascular disease, use of hypertensive medication and smoking yielded high predictive test performance for circulatory mortality.Risk scores for MI and stroke performed similarly to established risk scores.How this study might affect research, practice or policyAI-enabled RV extraction offers a non-invasive prognostic biomarker of vascular health that does not require blood sampling or blood pressure measurement, and potentially has greater community reach to identify individuals at medium-high risk requiring further clinical assessment.SYNOPSIS/PRECISRisk models developed in UK Biobank (validated in EPIC-Norfolk) using Artificial Intelligence enabled retinal vasculometry indices, age, history of cardiovascular disease, use of hypertensive medication and smoking yielded high predictive test performance for circulatory mortality. Risk scores for MI and stroke performed similarly to established risk scores.

show abstract

“…Artificial intelligence (AI) deep learning (DL) algorithms developed for retinal photographs have shown highly accurate detection and diagnosis of major eye diseases (e.g., diabetic retinopathy, 1 – 6 age-related macular degeneration, 7 – 9 glaucoma 10 – 12 ), measurement of retinal vessel caliber, 13 retinal vessel segmentation, 14 , 15 and even estimation of cardiovascular risk factors. 16 – 18 As such, integrating DL algorithms into real-time clinical workflow is a priority to realize the significant potential of AI for clinical diagnosis and disease risk stratification.…”

Section: Introductionmentioning

confidence: 99%

Deep-Learning–Based Pre-Diagnosis Assessment Module for Retinal Photographs: A Multicenter Study

Yuen

Ran

Shi

et al. 2021

Trans. Vis. Sci. Tech.

View full text Add to dashboard Cite

Purpose Artificial intelligence (AI) deep learning (DL) has been shown to have significant potential for eye disease detection and screening on retinal photographs in different clinical settings, particular in primary care. However, an automated pre-diagnosis image assessment is essential to streamline the application of the developed AI-DL algorithms. In this study, we developed and validated a DL-based pre-diagnosis assessment module for retinal photographs, targeting image quality (gradable vs. ungradable), field of view (macula-centered vs. optic-disc-centered), and laterality of the eye (right vs. left). Methods A total of 21,348 retinal photographs from 1914 subjects from various clinical settings in Hong Kong, Singapore, and the United Kingdom were used for training, internal validation, and external testing for the DL module, developed by two DL-based algorithms (EfficientNet-B0 and MobileNet-V2). Results For image-quality assessment, the pre-diagnosis module achieved area under the receiver operating characteristic curve (AUROC) values of 0.975, 0.999, and 0.987 in the internal validation dataset and the two external testing datasets, respectively. For field-of-view assessment, the module had an AUROC value of 1.000 in all of the datasets. For laterality-of-the-eye assessment, the module had AUROC values of 1.000, 0.999, and 0.985 in the internal validation dataset and the two external testing datasets, respectively. Conclusions Our study showed that this three-in-one DL module for assessing image quality, field of view, and laterality of the eye of retinal photographs achieved excellent performance and generalizability across different centers and ethnicities. Translational Relevance The proposed DL-based pre-diagnosis module realized accurate and automated assessments of image quality, field of view, and laterality of the eye of retinal photographs, which could be further integrated into AI-based models to improve operational flow for enhancing disease screening and diagnosis.

show abstract

Automated quantification of retinal vessel morphometry in the UK biobank cohort

Cited by 17 publications

References 33 publications

Retinal Vascular Tortuosity and Diameter Associations with Adiposity and Components of Body Composition

Retinal Vascular Tortuosity and Diameter Associations with Adiposity and Components of Body Composition

Artificial intelligence enabled retinal vasculometry for prediction of circulatory mortality, myocardial infarction and stroke

Deep-Learning–Based Pre-Diagnosis Assessment Module for Retinal Photographs: A Multicenter Study

Contact Info

Product

Resources

About