Google has developed a system using MachineLearning, which allows detecting possible cardiovascular problems before they happen and which has a 70% success rate.

One of the companies that is betting the strongest on Artificial Intelligence and Macine Leraning is Google, which is developing solutions that contribute a lot. Some time ago, they presented software that had the characteristic of predicting people's age, if they were a smoker and even measuring blood pressure. These data are obtained, thanks to the fact that the information from a subject's eye is read, something that we find truly interesting and that is not at all invasive.

This utility, which gives us a lot of information just by looking at our eye, through Machine Learning techniques, can give us new important data, including the probability of suffering a heart problem. This can be achieved instantly and the most interesting thing is that it does not require any type of blood test, or any other invasive method.

This utility, which is in development, has an accuracy rating of approximately 70%. If we compare it with traditional methods, which are more expensive, slow and invasive, the outlook improves, since a blood test, the most common method, is 72% accurate. The research has been carried out with a total of 500.000 patients in the United Kingdom, who have volunteered to carry out this kind of experiment.

The interesting thing is that the eye has a series of patterns in its blood vessels, which are indicators of a possible cardiovascular problem. The neural networks of the system have the ability to detect patterns and allow an assessment of risks.

Detecting possible cardiovascular accidents clearly allows saving lives. In addition to saving lives, this method can reduce the direct health costs of the early detection of this type of pathology and also those after suffering the disease. By knowing in advance, at minimal cost, potential future problems, millions can be saved and the death rate reduced - a medical advance.

Source: Eteknix