Artificial intelligence from Google will instantly detect a heart attack

Artificial intelligence from Google will instantly detect a heart attack

119
0
SHARE

It would seem that the new can come up in the field of methods of diagnosis of diseases of the heart, because all is already invented – know yourself improve existing methods! However, researchers from the biomedical company Verily (the owner of which, by the way, is Google) was able to develop a new method to identify cardiovascular diseases using neural networks and artificial intelligence. Moreover, this method does not require blood sampling and other invasive procedures, and the result is given in just a few seconds.

As, so to speak, of the diagnostic material from Verily algorithm uses an image of the retina. One procedure retina scan is enough to determine not only the presence of different cardiovascular pathologies, but also to identify a person’s age, blood pressure levels and find out if he smokes. In order to train the algorithm for the diagnosis, Verily used data from 300,000 patients. These data included information on the state of the retina, as well as General information about the patients and about their health using various indicators. Based on the data obtained, the neural network has identified a number of patterns that can speak about the presence of cardiovascular pathology.

The main advantage of the new method is that it provides extremely high speed diagnostics, practically not yielding to the accuracy of “traditional” methods. Verily the algorithm predicts the likelihood of a heart attack in the next 5 years with a probability of 70%, while the most accurate method now used SCORE (requiring, among other things, taking a blood sample) is right around 72% of cases. Despite the promising development right now to introduce it into medical practice is impossible, as it requires further testing, debugging, and monitoring patients. Do you think that in the future AI will replace the human at the doctor’s office? Speak out in our telegram chat.