Google learned to predict the date of death

The new neural network, developed by Google based on deep learning algorithms, is able to determine the approximate timing of death based on the patient's medical indicators. This is reported by Bloomberg.

The algorithm was tested on several hundred thousand patients, including a woman suffering from the terminal stage of breast cancer, whose lungs have already begun to fill with fluid. Computers analyzed the X-ray scan data and determined that there was a 9.3 percent chance of the patient dying while in hospital. However, the neural network, which accounted for 175 639 parameters, increased this figure to 19.9 percent. The woman died in ten days.

The neural network is able not only to predict death, but also other outcomes of diseases, as well as the length of stay in the hospital and the likelihood of re-hospitalization. The algorithm can take into account records in PDF-files and old medical cards. It also allows physicians to save up to 80 percent of the time that was normally spent on routine data preparation work.

According to the company's representatives, Google's next step will be the introduction of its prognostic system into clinics.

lenta.ru