Corti is an AI system that can analyze background noises and diagnose whether someone is suffering cardiac arrest over the phone. Emergency dispatchers could soon get some assistance thanks to AI (Photo via Washington Post)

 

It’s not easy being an emergency dispatcher. You not only have to keep the person on the other line calm, but you need to ask them questions at the same time to help them as much as possible.  It’s a lot to take on, but soon these dispatchers may get their own AI assistants. In 2016, dispatchers in Copenhagen worked with an AI named Corti, who understands words and sounds during a call to recognize cardiac arrest. The AI then prompts the emergency team with the right questions to get a more accurate diagnosis.

 

Rather than searching for signals, Corti trains itself by listening to sounds from a large volume of calls to identify important factors and continue improving its model as it continues working. The most important sounds are non-verbal, and Corti needs the ability to sort through background noise like sirens or yelling to pick out clues.

 

Giving dispatchers more accurate diagnosis allows them to help patients even more. They may be able to coach someone on the phone through CPR or even better prepare first responders. Some cities can even use the technology to send out drones with automatic defibrillators, which can be faster than an ambulance.

 

In one instance, the dispatcher concluded a man who’d fallen off the roof had broken his back. Corti analyzed the background noises and hear a faint rattling, which fit the pattern of someone trying to breathe despite a stopped heart. Unfortunately, since the AI was in the testing phase, it didn’t send any alerts to the dispatcher.

 

Corti helps out in other ways, like reminding the dispatcher to ask for the address of the incident and whether the ambulance is going to the right address. It all sounds promising and shows that AI can actually help human workers rather than outright replace them. Though Corti’s tests are running smoothly, the company is still working on the tech. The AI system still needs to refine its diagnosing abilities as it takes in more information. The company also hopes to announce plans to bring Corti to the US soon.

 

While most people are suspicious of AI technology, this is an instance where it can actually make a big difference. Even the most trained people are prone to error, and in such a panic-inducing situation, it’s easy to miss the small clues. Having technology that can pick up on this can help out a lot of people in the long run. 

 

Below is an example of the AI in use:

 

Have a story tip? Message me at: cabe(at)element14(dot)com

http://twitter.com/Cabe_Atwell