ny of us likely believe that the work of emergency medical technicians is quite straightforward. They respond to an emergency and try to stabilize the patient according to a set medical model until arrival at a hospital.
In reality, EMTs and paramedics are sometimes confronted with a series of life-and-death questions and dilemmas at the scene: How many people have been injured? Who requires medical attention more urgently? What kind of treatment is needed urgently? How many more ambulances should be called?
In this kind of high-stress situation, the brain can become flooded with information and this can result in mistakes or even unnecessary deaths.
The research
“It’s a cognitively loaded environment,” Omer Peri, lecturer in the School of Industrial Engineering & Management at Afeka College of Engineering in Tel Aviv, tells ISRAEL21c.
Peri, an industrial engineer, is also a PhD candidate at Ben-Gurion University of the Negev (BGU), where for the past six years he has been researching how stressful mass-casualty events affect the cognitive abilities of paramedics.
“Using available data, we managed to create a behavior model of the incident commander; what we discovered is that in these situations people meet the edge of their human capacity,” explains Peri.
The research is conducted under the supervision of BGU’s Prof. Yuvan Bitan, an associate professor in the Department of Health Policy and Management, and Prof. Avishay Goldberg, chairperson of the Department of Health Systems Management, and in partnership with Eli Jaffe, deputy director general of community at Israel’s national emergency service, Magen David Adom.
AI to the rescue
The behavior model is intended to train first responders to react to mass-casualty events in the most efficient way possible, says Peri.
In addition, for the past year, Peri’s team has been working extensively to develop an AI-based algorithm that could help paramedics improve decision-making during high-stress casualty events.
“Through the recognition of human limitations, we are developing an algorithm that will help them function in these types of situations.”
Peri explains that a person’s working memory can retain up to seven pieces of information at a time. When an incident commander’s working memory has been maxed out and a new piece of data comes in, it can be uploaded into the algorithm for the AI to digest.
“In order to minimize the amount of information that a paramedic takes in, the algorithm scans the data on the casualties at the scene, and sets the order of their treatment and evacuation,” explains Perry.
“The ultimate goal is to minimize the deterioration of patients at the scene.”
Peri’s team is now training the algorithm to optimize its decision-making process.
Once it’s available for use, approximately in a year’s time, the algorithm could be executed by computers of emergency service operators, or even via smartphones of first responders at the scene.
“Nowadays, people want to paint every new technology as AI, even though sometimes it’s light years away from having anything to do with AI. Our algorithm can absolutely be called AI,” Peri says.
Impact of October 7 attacks
Following the October 7 attacks, Peri expanded the research to include the issue of allocation of patients to particular medical centers.
Peri’s team found that in the immediate aftermath of the attacks, most patients were brought to Soroka Medical Center in Beersheva and Barzilai Medical Center in Ashkelon. Being the two closest hospitals to the Gaza border, they reached maximum capacity almost instantly.
“All the while, emergency rooms at Ichilov Hospital [Tel Aviv Sourasky Medical Center] and Rabin Medical Center in Petah Tikva were practically empty; people lost limbs simply because they were not transferred to the right places,” notes Peri.
In the future, his research could lead to better methods of directing the transport of patients to the hospitals where they can best be cared for.
Produced in association with ISRAEL21c
