Stroke Prediction Boosted By Retinal Imaging


Stroke Prediction Boosted By Retinal Imaging
Credit: Inside Creative House/Getty Images

A score made up of a series of visual measurements taken from images of the retinal blood vessels at the back of the eye can significantly improve stroke prediction, shows new research.

The study, led by the Hong Kong Polytechnic University and the University of Melbourne, identified 29 retinal vascular features that significantly increased a person’s risk for stroke.

Each standard deviation increase in these features raised a person’s risk by 10–19% and when added to a model including other risk factors for stroke the addition of the retinal measures significantly improved the accuracy of the model.

Stroke, where poor blood flow to the brain causes cell death, is a leading cause of death and disability in the U.S. and around the world. Almost 800,000 people are predicted to have a stroke every year in the U.S. and the costs of treating those affected is very high, particularly as people who experience this type of cardiovascular event often need long-term rehabilitation care and assistance.

Although having a stroke can lead to significant health problems, such an event can be prevented in most cases by early treatment of risk factors such as high blood pressure and sustained lifestyle changes such as improvements in diet and exercise and reductions in smoking and drinking alcohol.

As reported in the journal Heart, the research team used data from 45,161 participants of the UK Biobank. During a median follow-up time of 12.5 years, 749 strokes occurred in the cohort.

The researchers analyzed 118 vascular parameters in retinal images from study participants. Of these, 29 were significantly associated with stroke risk after correcting for potential confounding factors. The measures included things like blood vessel density and how twisted and/or branched the retinal vessels are.

When the results of scoring the 29 retinal features were added to a wider stroke risk model including other clinical and lifestyle factors the overall accuracy was improved. Using a statistical test known as the area under the receiver operator curve, the score improved from 0.739 to 0.752, a significant difference, when comparing a standard stroke risk model with the standard model plus the retinal measures, respectively.

“Importantly, when combined with age and sex, the newly identified retinal parameters had comparable predictive power for stroke risk when compared with established traditional risk factors,” wrote co-lead author Mingguang He, a researcher at the School of Optometry Research Centre for SHARP Vision at Hong Kong Polytechnic University, and colleagues.

“Given that age and sex are readily available, and retinal parameters can be obtained through routine fundus photography, this model presents a practical and easily implementable approach for incident stroke risk assessment, particularly for primary healthcare and low-resource settings.”



Source link

Latest articles

Related articles

Discover more from Technology Tangle

Subscribe now to keep reading and get access to the full archive.

Continue reading

0