Detecting out-of-hospital cardiac arrest using artificial intelligence:
Report on results of EENA/Corti project now available.
Out-of-hospital cardiac arrest is one of the leading causes of death both in Europe and worldwide. When suffering a cardiac arrest, chances of survival decrease up to 10% per minute. The work of emergency call-takers and emergency respondents is thus key to ensure early recognition and life-saving intervention. As it is so crucial, how to better assist them in their job?
When faced with potential cases of cardiac arrest, time and accuracy are key, which is why Danish company Corti looked into how AI could provide real-time decision support in medical dispatch – and developed a technology that acts as a virtual assistant for call-takers.
In 2018, EENA & Corti partnered to pilot this technology in emergency response centers in France and Italy.
In the report prepared by EENA, Corti and the pilot sites, you will learn about the challenges faced during the project, including data privacy issues and the difficulties of acquiring the necessary datasets.
“The EENA-Corti project was an important learning experience for the use of AI in emergency services, demonstrating not only the potential of the technology, but also how to overcome significant challenges to pave the way for the future of emergency response”, Jerome Paris, EENA Managing Director.
Recommendations & Conclusions from the report:
• Artificial Intelligence does have the potential to assist decision-making of emergency call-takers, by increasing the accuracy of out-of-hospital cardiac arrest detections.
• The pilot project ran in France demonstrated that the AI can also speed up the detections of cardiac arrest over the phone.
• Further training of the AI is needed to keep improving the performance and optimise the models. Wider and good quality datasets play a crucial role to further improving accuracy.
• To ensure maximum efficacy, the AI should be run alongside effective protocols .
• Additional data should be considered an aid to emergency call-takers and emergency response professionals in order to save lives.
• Such data should be presented in a user-friendly manner in order to be effective.
“On top of developing preliminary AI models for Out-of-Hospital Cardiac Arrest detection in French and Italian, the results are important because they confirm – despite the challenges – the potential of AI in augmenting call-takers and dispatchers. We look forward to moving ahead and beyond the pilot phases.”, Andreas Cleve, CEO, Corti.