Opendata, web and dolomites


Artificial Intelligence for Emergency Medical Services: a smart digital assistant for faster and more accurate cardiac arrest recognition during emergency calls

Total Cost €


EC-Contrib. €






 AI4EMS project word cloud

Explore the words cloud of the AI4EMS project. It provides you a very rough idea of what is the project "AI4EMS" about.

dispatcher    natural    leveraging    faster    guidelines    ict    commercialization    communicating    recognise    accessed    presenting    disrupt    struggling    dispatchers    almost    million    dispatch    single    strategies    defibrillation    created    decision    manner    artificial    disruptive    decreasing    situations    ehealth    triage    historical    assistant    calls    total    accuracy    digital    device    causes    analytics    render    jobs    healthcare    saas    economy    world    medical    analysing    supporting    market    2020    sales    smart    prior    worldwide    plan    ohca    chances    320    2012    amounts    2003    out    95    minutes    hospital    dsm    first    127    recognition    unfeasible    commercialisation    efficiency    nvidia    society    forecasted    tx1    insights    accurate    speech    humans    activating    secure    arrest    recognising    solution    leaders    ai    86    73    2016    ems    10    revenues    supports    delay       upgrade    death    cardiac    services    action    intelligence    minute    critical    reducing    emergency    language    collapse    2024    survival    human    tools    time    data    ai4ems   

Project "AI4EMS" data sheet

The following table provides information about the project.


Organization address
address: BLEGDAMSVEJ 6
postcode: 2200
website: n.a.

contact info
title: n.a.
name: n.a.
surname: n.a.
function: n.a.
email: n.a.
telephone: n.a.
fax: n.a.

 Coordinator Country Denmark [DK]
 Project website
 Total cost 2˙055˙976 €
 EC max contribution 1˙439˙183 € (70%)
 Programme 1. H2020-EU.3. (PRIORITY 'Societal challenges)
2. H2020-EU.2.3. (INDUSTRIAL LEADERSHIP - Innovation In SMEs)
3. H2020-EU.2.1. (INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies)
 Code Call H2020-SMEInst-2018-2020-2
 Funding Scheme SME-2
 Starting year 2018
 Duration (year-month-day) from 2018-08-01   to  2020-01-31


Take a look of project's partnership.

# participants  country  role  EC contrib. [€] 
1    CORTI APS DK (KOBENHAVN) coordinator 1˙439˙183.00


 Project objective

Out-of-Hospital Cardiac Arrest (OHCA) is one of the leading causes of death worldwide. It is a time-critical condition with survival chances decreasing by 10% with every minute of delay from collapse to defibrillation. Currently, Emergency Medical Services (EMS) dispatchers use guidelines to recognise OHCA during emergency calls prior to activating the emergency response system. EMS are struggling as emergency calls have increased in Europe from 100 million calls in 2003 to 320 million in 2016. Thus, assistant decision tools will be necessary to help EMS to faster identify OHCA situations.

Our solution, AI4EMS, is the first and only smart digital assistant for EMS dispatchers that supports the triage decision-making by: 1) processing and analysing emergency calls in real-time; 2) recognising OHCA in an evidence-based process from large amounts of historical data (unfeasible to humans); and 3) presenting the most important insights to the EMS dispatcher in a user friendly manner. AI4EMS allows for faster (reducing almost 3 minutes on average) and more accurate (increase from 73.9% human accuracy to 95%) OHCA recognition by leveraging advanced speech analytics and AI. We offer a user-friendly and secure SaaS solution capable of communicating using Natural Language, accessed via a Nvidia TX1-based device. We are directly supporting the eHealth Action Plan 2012-2020 and Digital Single Market (DSM) strategies, by providing a disruptive ICT technology to improve EMS dispatch efficiency and triage accuracy – which will impact the economy and society at large.

With the upgrade and commercialisation of AI4EMS we will disrupt the Artificial Intelligence (AI) market for healthcare taking a step further on our goal to become world leaders in EMS artificial intelligence. Forecasted sales will render revenues of €86.7 million in the first five years of commercialization and a total of 127 new jobs will be created by 2024.


List of deliverables.
AI4EMS online content Websites, patent fillings, videos etc. 2019-08-01 13:01:00

Take a look to the deliverables list in detail:  detailed list of AI4EMS deliverables.


year authors and title journal last update
List of publications.
2020 Andreas Cleve, Dimitri Devillers, Matteo Palladini, Jerome Paris, Rose Michael, Etienne Faure, Rodolfo Bonora
Detecting Out-of-Hospital Cardiac Arrest Using Artificial Intelligence
published pages: , ISSN: , DOI:
2019 Valentin Liévin, Andrea Dittadi, Lars Maaløe, Ole Winther
Towards Hierarchical Discrete Variational Autoencoders
published pages: , ISSN: , DOI:
NeurIPS Workshop on Advances in Approximate Bayesian Inference 2020-02-13
2017 Marius Paraschiv, Lasse Borgholt, Tycho Max Sylvester Tax, Marco Singh, Lars Maaløe
Exploiting Nontrivial Connectivity for Automatic Speech Recognition
published pages: , ISSN: , DOI:
NIPS workshop on machine learning for audio 2019-08-05
2019 Lars Maaløe, Marco Fraccaro, Valentin Liévin, Ole Winther
BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling
published pages: , ISSN: , DOI:
arXiv preprint arXiv:1902.02102 2019-08-05
2017 Tycho Max Sylvester Tax, Jose Luis Diez Antich, Hendrik Purwins, Lars Maaløe
Utilizing Domain Knowledge in End-to-End Audio Processing
published pages: , ISSN: , DOI:
31st Conference on Neural Information Processing Systems (NIPS 2017) 2019-08-05
2019 Stig Nikolaj Blomberg, Fredrik Folke, Annette Kjær Ersbøll, Helle Collatz Christensen, Christian Torp-Pedersen, Michael R. Sayre, Catherine R. Counts, Freddy K. Lippert
Machine learning as a supportive tool to recognize cardiac arrest in emergency calls
published pages: 322-329, ISSN: 0300-9572, DOI: 10.1016/j.resuscitation.2019.01.015
Resuscitation 138 2019-08-06
2018 Jan Kremer, Corti, Copenhagen, Denmark, Lasse Borgholt, Corti, Copenhagen, Denmark, Lars Maaløe , Corti, Copenhagen, Denmark,
On the Inductive Bias of Word-Character-Level Multi-Task Learning for Speech Recognition
published pages: , ISSN: , DOI:
32nd Conference on Neural Information Processing Systems (NeurIPS 2018) 2019-08-05

Are you the coordinator (or a participant) of this project? Plaese send me more information about the "AI4EMS" project.

For instance: the website url (it has not provided by EU-opendata yet), the logo, a more detailed description of the project (in plain text as a rtf file or a word file), some pictures (as picture files, not embedded into any word file), twitter account, linkedin page, etc.

Send me an  email ( and I put them in your project's page as son as possible.

Thanks. And then put a link of this page into your project's website.

The information about "AI4EMS" are provided by the European Opendata Portal: CORDIS opendata.

More projects from the same programme (H2020-EU.3.;H2020-EU.2.3.;H2020-EU.2.1.)

PAROST (2019)

Disruptive technology for the identification, quantification and prediction of the evolution of damages in civil engineering structures that increases safety while reducing maintenance costs.

Read More  

HUD (2018)


Read More  

DeltaQon (2019)

IOT and cloud computing for online medical analysis service platform

Read More