Google teaches the neural network to predict the death of a person

Google teaches the neural network to predict the death of a person


Modern medicine aims to prevent and anticipate the development of serious life-threatening diseases. However, forces and human knowledge is often not enough to predict certain complications. To help doctors in the future may come AI DeepMind by Google, which at the moment is trained to predict the death of people.

As often happens in such a situation, to teach the AI used the already known clinical cases. The developers of Google “fed” DeepMind data almost 220 thousand adult patients who underwent treatment between 2009 and 2016. Patient information provided medical center University of California San Francisco and Medical University of Chicago. In addition, the researchers used information obtained from the Ministry of veterans Affairs of the United States, which gave access to data about 700,000 under the supervision of ex-military. Just a database made of more than 46 billion paragraphs about a variety of health indicators. At the moment the algorithm is aimed at searching for conditions such as acute renal failure and pneumonia. These States were chosen because, according to the authors of the work

“Acute renal failure or pneumonia can affect people of any age, and often they start after the ordinary procedures and operations. 11% of all deaths in the hospital due to the deterioration of health of the patient and changes in patient’s condition are not always able to notice at an early stage”.

Now the system is able to predict the likelihood of death of the patient within 24-48 hours, but the main goal of the project is to learn to predict death within the next 12 months. In addition, the planned extension of the data on the deadly pathological conditions. Such use of artificial intelligence in medical applications can be used not only in order to be ready for resuscitative measures, but also to prevent the development in patients of severe conditions.