AI system to predict out-of-hospital cardiac arrests

Japanese scientists have recently discovered a way to use artificial intelligence (AI) to predict accurately the risk of an out-of-hospital cardiac arrest by using time and weather data to identify patterns.

Indeed, it was found out that information such as temperature, relative humidity, and rainfall, alongside the different times of the day, could help providers identify when cardiac arrest would most likely occur, and thus, be able to prevent it.

Hence, the study revealed that the risk of cardiac arrests was higher during Sundays, Mondays, and bank holidays, as well as when the temperature would fall sharply. In order to reach that conclusion, the scientists studied more than 525,000 cases that happened between 2005 and 2013 before comparing them to another 135,000 cases from two years later to test the algorithms’ accuracy.

Though it is important to note that the researchers only have detailed information on the location of cardiac arrests in Japan’s Kobe city, where the study had taken place. However, they believe that it can be standardized somewhere else thanks to the large sample size and comprehensive meteorological information.

Therefore, the AI model aims to prevent out-of-hospital cardiac arrest and improve the prognosis of patients with a warning system for citizens and emergency medical services on high-risk days in the future.

Related Posts