In order to give you a better service Deutsche Messe uses cookies. If you continue we assume that you consent to receive cookies on all Deutsche Messe websites. Legal Notes

Research & Innovation

AI application predicts the time of death

Russian researchers have created an artificial neural system that uses data provided by smartphones and wearables to estimate the life expectancy of the user.

18 Apr. 2018 Isabelle Reiff

Researchers from the small Russian biotechnology company Gero and the Moscow Institute of Physics and Technology (MIPT) have taken advantage of a multi-year National Health and Nutrition Survey ( NHANES ) done in the US. With the data collected, including both clinical and activity measurements, they trained a neural network to recognize patterns. The identified biomarkers now reveal relevant movement patterns and their relationship to health status and mortality risk (see also Scientific Reports ). The algorithm based on artificial intelligence (AI) surpasses all previously available models for calculating life expectancy in its accuracy. In addition, it can be easily coupled with the common activity sensors in smartphones and fitness trackers.

The beta version of a free app for iPhones has already been launched. Gero Lifespan estimates the lifespan of the user with the help of the integrated smartphone accelerometer. The researchers see many benefits for insurers, state pension planning, and anti-aging therapies.