AI heart attack detection system Corti to be deployed across Europe

Copenhagen-based startup Corti will deploy systems across EENA member countries from early spring in 2019

15Oct

Corti, an AI system used to detect heart attacks over the phone, has received the green light to begin trials across European Emergency Number Association (EENA) member countries from early 2019.

The Corti AI system has already been deployed in the Danish capital Copenhagen where the startup has been very successful. After analyzing thousands of emergency calls made in Copenhagen in 2014, the AI system was on average 30 seconds faster at detecting cardiac arrests in callers than human operators, functioning on a 93% accuracy rate in comparison to the humans 73% accuracy rate.

Ischemic heart disease results in more deaths than any other disease or ailment worldwide and also one of the most time-sensitive ailments a person can suffer from. The American Heart Association states that a person's odds of survival dip by 7–10% every minute that passes before CPR is administered. Only 2–11% of people who have a heart attack outside of a hospital end up surviving, meaning the extra seconds the Corti system provides could have a huge impact on survival rates.


Visit Innovation Enterprise's Big Data & Analytics for Pharma Summit, part of DATAx New York on December 12–13, 2018


The EENA, which is based in Brussels, has more than 80-member countries, most of whom will begin trialing the AI system in spring of 2019. While Corti wants the system to be part of all emergency call operations, with no distinct global standards for organizations trying to save lives with AI, it will largely come down to how the trials proceed and how each nation navigates privacy laws around AI.

The Corti AI system works by using deep neural networks to identify heart attacks through patients' voices, with operators able to flag and review calls. While the system has only currently been trained to identify heart attacks, Corti hope to expand to its identification capabilities to other cardiac events and even drug overdoses.

How to take the risk out of saas development small

Read next:

How to take the risk out of SaaS development

i