A deep-learning system for the assessment of cardiovascular disease risk via the measurement of retinal-vessel calibre

Autor: Paul Mitchell, Clement C Y Tham, Louise M Burrell, Avshalom Caspi, Charumathi Sabanayagam, Tien Yin Wong, Jost B. Jonas, Gavin Tan, Jason C. S. Yam, Bamini Gopinath, Mong Li Lee, Tyler Hyungtaek Rim, Chew Yian Chai, Marco Yu, Ling-Jun Li, Carol Y. Cheung, Daniel S W Ting, Omar Farouque, Yih Chung Tham, Terrie E. Moffitt, Richie Poulton, Dejiang Xu, Wynne Hsu, Su Jeong Song, Ya Xing Wang, Ching-Yu Cheng
Rok vydání: 2020
Předmět:
Male
0301 basic medicine
Intraclass correlation
Myocardial Infarction
Datasets as Topic
Medicine (miscellaneous)
Blood Pressure
Coronary Disease
Body Mass Index
chemistry.chemical_compound
0302 clinical medicine
Risk Factors
Photography
Aged
80 and over
education.field_of_study
Diabetic retinopathy
Middle Aged
Computer Science Applications
Stroke
Cholesterol
Female
Risk assessment
Biotechnology
Adult
medicine.medical_specialty
Population
Biomedical Engineering
Bioengineering
Hypertensive Retinopathy
Risk Assessment
Retina
03 medical and health sciences
Deep Learning
Hypertensive retinopathy
Ophthalmology
Image Interpretation
Computer-Assisted
medicine
Humans
education
Aged
Retrospective Studies
Glycated Hemoglobin
business.industry
Retinal Vessels
Retrospective cohort study
Retinal
medicine.disease
030104 developmental biology
Blood pressure
chemistry
business
030217 neurology & neurosurgery
Zdroj: Nature Biomedical Engineering. 5:498-508
ISSN: 2157-846X
Popis: Retinal blood vessels provide information on the risk of cardiovascular disease (CVD). Here, we report the development and validation of deep-learning models for the automated measurement of retinal-vessel calibre in retinal photographs, using diverse multiethnic multicountry datasets that comprise more than 70,000 images. Retinal-vessel calibre measured by the models and by expert human graders showed high agreement, with overall intraclass correlation coefficients of between 0.82 and 0.95. The models performed comparably to or better than expert graders in associations between measurements of retinal-vessel calibre and CVD risk factors, including blood pressure, body-mass index, total cholesterol and glycated-haemoglobin levels. In retrospectively measured prospective datasets from a population-based study, baseline measurements performed by the deep-learning system were associated with incident CVD. Our findings motivate the development of clinically applicable explainable end-to-end deep-learning systems for the prediction of CVD on the basis of the features of retinal vessels in retinal photographs. Deep-learning models for the automated measurement of retinal-vessel calibre in retinal photographs perform comparably to or better than expert graders in associations of measurements of retinal-vessel calibre with cardiovascular risk factors.
Databáze: OpenAIRE
Externí odkaz: