Saving Lives With AI in an Emergency Dispatch Center

A unique technological system developed with AudioCodes has been installed in MDA's 101 Emergency Dispatch Center and is already helping EMTs and paramedics in the dispatch center to improve the effectiveness of the call-taking process.

The system, based on artificial intelligence (AI) in Hebrew, and the first of its kind in the world specifically intended for use in an emergency dispatch center, transcribes in real time the call on the 101 line, and helps the EMT in the Dispatch Center to provide a faster, more professional response—in cases where there is a disturbance in the quality of the call from the patient or the caller.

Over the past two years, MDA, along with AudioCodes, have developed the unique system that works based on AI and that has been fully installed on MDA's command and control (C&C) system. At the time of receiving a call, the EMT in the Dispatch Center must be at maximum concentration levels, even when surrounded by a great deal of noise, in order to receive information from the caller.

The information required includes the location and building an initial understanding of the medical situation in order to give effective advice and to dispatch the relevant ambulance or mobile intensive care unit (MICU). The new system transcribes the entire conversation and allows the crew to concentrate on the critical information, and makes understanding the caller easier—for example confirming the address—without struggling to understand the caller and wasting critical time by asking the caller to repeat themselves. The system has learned key phrases connected to medicine and emergencies. This also helps the EMT better understand what is being said and provides them with an overall picture that allows for rapid, professional decision-making while undertaking multiple simultaneous tasks.  

The system continues to undergo improvements and development that will improve the phone response. The system will identify key words for medical treatment and emphasize them in order to bring them to the attention of the EMT who can then give the relevant online advice to calls such as chest pain, resuscitation, or a serious vehicle accident, can send messages to the relevant managers in order to ensure correct scene and personnel management, and can set in motion relevant automated actions, such as sending the correct teams to significant events. The system will be able in the future to transcribe other languages and save even more lives.

"We are proud to have the honor," said Shabtai Adlersberg, president and CEO of AudioCodes, "of developing and providing a world-first call-transcribing system that will allow for ultimate management of the call-taking system, reducing response times and saving lives. MDA is an early adopter of AI tools as part of its preparations for its future needs, and that is an honor for us, as a global company used to marketing and selling its products overseas, to work to benefit MDA and the Israeli public. The accuracy of the AI transcribing solutions have significantly improved and can sometimes prove to be better than human ability, for example the ability to respond to and analyze speech that is whispered or of a low sound quality, and presenting them to the call takers in MDA accurately and quickly. We thank MDA for this cooperation, and see this latest development as an opportunity for us and MDA to present this unique innovation to other emergency services around the world."   

"Magen David Adom, Israel's National Rescue Service, stands at the forefront of technology with world-leading command and control systems that were developed within the organization," said Deputy Director General for Community Dr. Eli Jaffe. "Today we have taken another significant step by advancing new life-saving technology along with AudioCodes. Integrating AI that allows us to improve the response to a patient in distress, to quickly and effectively understand the content of the call, and to reduce the time needed to obtain the required details and dispatching of teams to the field during an emergency call."