Vogon Today

Selected News from the Galaxy

StartMag

Will an algorithm predict heart attacks or heart disease?

Will an algorithm predict heart attacks or heart disease?

According to science, the eyes are the mirror of the heart. A new artificial intelligence algorithm called Quartz could be the breakthrough for predicting heart attacks or heart disease just by looking at our retina. All the details

Having a crystal ball to predict disease may not be feasible but artificial intelligence is working to do its part . In fact, an algorithm called Quartz is being studied that will be able to predict whether we will have heart attacks or heart disease, simply by observing a patient's retina.

The research was conducted by St George's Medical School in London and published in the British Journal of Ophthalmology .

THE TARGET

Predicting and preventing disease is one of the most ambitious goals of science and medicine but thanks to the help of artificial intelligence it always seems a little more possible.

This is the case of the Quartz algorithm which, by examining the shape and thickness of the circulatory system in the eye, aims to identify and predict possible serious future problems. In particular, cardiovascular diseases and heart attacks.

The algorithm, based on artificial intelligence, in fact aims to understand the onset of potential diseases through the analysis of the blood vessels of the retina.

THE POWER OF THE EYES

Starting from the eyes to understand the heart has potential that science had already intuited. The retina, in fact, can send some signals in advance because it is sensitive to variations in the risk factors of cardiovascular diseases, such as blood sugar and blood pressure, and therefore it is important to pay attention to the signals sent by the changes in its small blood vessels.

I STUDY

The team of researchers then trained the algorithm using a database of over 88,000 British patients aged 40 to 69 to recreate a model for predicting stroke, heart attack and death from circulatory disease using the analysis of the thickness and structure of the ocular blood vessels. In particular, the width, the area of ​​the vessels and the degree of curvature of the arterioles and venules in the retina were examined.

Then on a smaller group, made up of about 7,400 people between 48 and 92 years old and whose information came from a different research institute, he applied the same model to investigate the possible link between cancer and nutrition.

In order to study the actual development of diseases, the patients were followed for a period of 7 to 9 years.

THE RESULTS

By combining retinal imaging analysis and other information such as age, medical history, and patient habits, the algorithm has proven to be able to predict between half and two-thirds of deaths from severe cardiovascular disease occurring at that time. period of time.

The analysis of the results also showed that if in men, the arteriolar and venular width, the tortuosity and the variation in width were predictors of death from circulatory diseases, in women, on the other hand, they contributed to the prediction of the risk.

THE ADVANTAGES OF THIS NEW TECHNIQUE

This technique, already used in the UK and the US, has many advantages. It is non-invasive, does not require blood draws, is fully automated and low-cost.

Furthermore, compared to its predecessor, Framingham Risk Scores has been shown to be more effective because it takes into account much more information about patients (for example if a person is a smoker or if they take particular medications), allowing them to be more precise in forecasting.

DATA ON CARDIOVASCULAR DISEASES

Having more and more tools to prevent circulatory, cardiovascular, coronary artery disease, stroke and heart failure is also essential because they are the leading causes of disease and death worldwide.

In Europe, reminds Repubblica , every year 11 million people are diagnosed with cardiovascular disease, which becomes the cause of death for 4.3 million, and reaches the rate of 48% of deaths worldwide, of which 54% are women and 43% men.


This is a machine translation from Italian language of a post published on Start Magazine at the URL https://www.startmag.it/sanita/un-algoritmo-prevedra-infarti-o-malattie-cardiache/ on Sun, 30 Oct 2022 07:58:59 +0000.