Google systems can replace real cardiologists

Google has long been interested in medicine. For example, earlier developers of the company suggested using learning algorithms so that the network could recognize from the photos of people's eyes the presence of diabetic retinopathy - damage to the vessels in the eyes.

Now experts believe that artificial intelligence can determine whether a person is at risk of experiencing a heart attack or other cardiovascular problems, Izvestia writes. So, thanks to special algorithms, Google's neural network can process the image of the blood vessels of the patient's eye, comparing the information obtained with the data of medical research put into the program.

To speed up the processing of information even more, the neural network can also use the results of treatment and diagnostics recorded in the patient's medical record. For example, with a control "reconciliation" of the state of vessels and the prediction of diseases it is possible to use information about previously transmitted diseases, bad habits and other factors negatively affecting the cardiovascular system. According to the developers, the neural network Google predicts deviations in health for the next five years with an accuracy of 70%.

meddaily.ru