Novel Retinal Imaging in Assessment of Cardiovascular Risk Factors and Systemic Vascular Diseases

Daniel S.W. Ting, Lily Peng, Avinash V. Varadarajan, Tin Yan Liu

Research output: Contribution to journalArticle

Abstract

The eye is the only organ that allows direct visualization of the systemic neuro-vasculature using non-invasive imaging modalities, providing invaluable microstructural changes that often precede the macrovascular diseases such as stroke and ischaemic heart diseases. Using computer-aided analysis, wider retinal venules and narrower arterioles were found to be associated with an increased risk of cardiovascular diseases (CVD). For chronic kidney disease, narrower retinal arteriolar calibre, smaller retinal vascular fractal dimensions, arteriovenous nicking and opacification were significantly associated with lower estimated glomerular filtration rate, microalbuminuria and higher albumin/creatinine ratio. More recently, the use of artificial intelligence (AI) using deep learning was shown to have promising results in detecting age, gender, systolic blood pressure and smoking status from the fundus photographs. The same AI system also had comparable area under curve to the conventional risk calculators in predicting 5-year cardiovascular risks. Lastly, optical coherence tomography (OCT), with and without angiography, is also an effective imaging modality to evaluate the CVDs. Diabetic patients had thinner retinal nerve fibre layer. OCT angiographic changes may serve as markers of diabetes even in the absence of typical retinal changes. Future research is important to explore incorporating some of these parameters into the calculation and prognostication of the long-term cardiovascular risk outcome.

Original languageEnglish (US)
Pages (from-to)106-118
Number of pages13
JournalFrontiers in Diabetes
Volume27
DOIs
StatePublished - Jan 1 2019

Fingerprint

Vascular Diseases
Artificial Intelligence
Optical Coherence Tomography
Blood Pressure
Retinal Vessels
Fractals
Venules
Arterioles
Glomerular Filtration Rate
Chronic Renal Insufficiency
Nerve Fibers
Area Under Curve
Myocardial Ischemia
Albumins
Creatinine
Angiography
Cardiovascular Diseases
Smoking
Stroke
Learning

ASJC Scopus subject areas

  • Internal Medicine
  • Endocrinology, Diabetes and Metabolism
  • Endocrinology

Cite this

Novel Retinal Imaging in Assessment of Cardiovascular Risk Factors and Systemic Vascular Diseases. / Ting, Daniel S.W.; Peng, Lily; Varadarajan, Avinash V.; Liu, Tin Yan.

In: Frontiers in Diabetes, Vol. 27, 01.01.2019, p. 106-118.

Research output: Contribution to journalArticle

Ting, Daniel S.W. ; Peng, Lily ; Varadarajan, Avinash V. ; Liu, Tin Yan. / Novel Retinal Imaging in Assessment of Cardiovascular Risk Factors and Systemic Vascular Diseases. In: Frontiers in Diabetes. 2019 ; Vol. 27. pp. 106-118.
@article{8a7dd611c7d54a58866811b121085587,
title = "Novel Retinal Imaging in Assessment of Cardiovascular Risk Factors and Systemic Vascular Diseases",
abstract = "The eye is the only organ that allows direct visualization of the systemic neuro-vasculature using non-invasive imaging modalities, providing invaluable microstructural changes that often precede the macrovascular diseases such as stroke and ischaemic heart diseases. Using computer-aided analysis, wider retinal venules and narrower arterioles were found to be associated with an increased risk of cardiovascular diseases (CVD). For chronic kidney disease, narrower retinal arteriolar calibre, smaller retinal vascular fractal dimensions, arteriovenous nicking and opacification were significantly associated with lower estimated glomerular filtration rate, microalbuminuria and higher albumin/creatinine ratio. More recently, the use of artificial intelligence (AI) using deep learning was shown to have promising results in detecting age, gender, systolic blood pressure and smoking status from the fundus photographs. The same AI system also had comparable area under curve to the conventional risk calculators in predicting 5-year cardiovascular risks. Lastly, optical coherence tomography (OCT), with and without angiography, is also an effective imaging modality to evaluate the CVDs. Diabetic patients had thinner retinal nerve fibre layer. OCT angiographic changes may serve as markers of diabetes even in the absence of typical retinal changes. Future research is important to explore incorporating some of these parameters into the calculation and prognostication of the long-term cardiovascular risk outcome.",
author = "Ting, {Daniel S.W.} and Lily Peng and Varadarajan, {Avinash V.} and Liu, {Tin Yan}",
year = "2019",
month = "1",
day = "1",
doi = "10.1159/000486269",
language = "English (US)",
volume = "27",
pages = "106--118",
journal = "Frontiers in Diabetes",
issn = "0251-5342",
publisher = "S. Karger AG",

}

TY - JOUR

T1 - Novel Retinal Imaging in Assessment of Cardiovascular Risk Factors and Systemic Vascular Diseases

AU - Ting, Daniel S.W.

AU - Peng, Lily

AU - Varadarajan, Avinash V.

AU - Liu, Tin Yan

PY - 2019/1/1

Y1 - 2019/1/1

N2 - The eye is the only organ that allows direct visualization of the systemic neuro-vasculature using non-invasive imaging modalities, providing invaluable microstructural changes that often precede the macrovascular diseases such as stroke and ischaemic heart diseases. Using computer-aided analysis, wider retinal venules and narrower arterioles were found to be associated with an increased risk of cardiovascular diseases (CVD). For chronic kidney disease, narrower retinal arteriolar calibre, smaller retinal vascular fractal dimensions, arteriovenous nicking and opacification were significantly associated with lower estimated glomerular filtration rate, microalbuminuria and higher albumin/creatinine ratio. More recently, the use of artificial intelligence (AI) using deep learning was shown to have promising results in detecting age, gender, systolic blood pressure and smoking status from the fundus photographs. The same AI system also had comparable area under curve to the conventional risk calculators in predicting 5-year cardiovascular risks. Lastly, optical coherence tomography (OCT), with and without angiography, is also an effective imaging modality to evaluate the CVDs. Diabetic patients had thinner retinal nerve fibre layer. OCT angiographic changes may serve as markers of diabetes even in the absence of typical retinal changes. Future research is important to explore incorporating some of these parameters into the calculation and prognostication of the long-term cardiovascular risk outcome.

AB - The eye is the only organ that allows direct visualization of the systemic neuro-vasculature using non-invasive imaging modalities, providing invaluable microstructural changes that often precede the macrovascular diseases such as stroke and ischaemic heart diseases. Using computer-aided analysis, wider retinal venules and narrower arterioles were found to be associated with an increased risk of cardiovascular diseases (CVD). For chronic kidney disease, narrower retinal arteriolar calibre, smaller retinal vascular fractal dimensions, arteriovenous nicking and opacification were significantly associated with lower estimated glomerular filtration rate, microalbuminuria and higher albumin/creatinine ratio. More recently, the use of artificial intelligence (AI) using deep learning was shown to have promising results in detecting age, gender, systolic blood pressure and smoking status from the fundus photographs. The same AI system also had comparable area under curve to the conventional risk calculators in predicting 5-year cardiovascular risks. Lastly, optical coherence tomography (OCT), with and without angiography, is also an effective imaging modality to evaluate the CVDs. Diabetic patients had thinner retinal nerve fibre layer. OCT angiographic changes may serve as markers of diabetes even in the absence of typical retinal changes. Future research is important to explore incorporating some of these parameters into the calculation and prognostication of the long-term cardiovascular risk outcome.

UR - http://www.scopus.com/inward/record.url?scp=85066932862&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066932862&partnerID=8YFLogxK

U2 - 10.1159/000486269

DO - 10.1159/000486269

M3 - Article

AN - SCOPUS:85066932862

VL - 27

SP - 106

EP - 118

JO - Frontiers in Diabetes

JF - Frontiers in Diabetes

SN - 0251-5342

ER -