Skip to content

Search the site


Pentagon asks if AI and drones could help identify “mass casualty” priorities

DARPA wants sensors on uncrewed aircraft vehicles or robots, and algorithms that analyze sensor data in real-time to...

For military or emergency response personnel, knowing whom to treat first in a mass casualty situation – whether that’s battlefield environment or an earthquake – can be nearly impossible and deeply overwhelming.

Now a new research project by the Pentagon’s DARPA hopes to pioneer the use of drones and robots equipped with sensors that can digitally identify the “signatures” of the most severe injuries to help medical responders.

DARPA hopes that the project will help it discover and deliver technology that lets first responders perform the most rapidly “scalable, timely, and accurate triage” in what it grimly calls  “mass casualty incidents (MCIs).”

DARPA Triage Challenge: Real-time data for mass casualty crises

The so-called DARPA Triage Challenge will run over three years.

Teams interested in applying for funding have until February 13, 2023 to put in proposals.

The long-term vision is “an initial, or primary stage of MCI triage supported by sensors on stand-off platforms, such as uncrewed aircraft vehicles or robots, and algorithms that analyze sensor data in real-time to allow medical personnel to identify casualties for urgent hands-on evaluation; followed by a secondary stage, after the most urgent casualties have been treated, supported by non-invasive contact sensors placed on casualties and algorithms that analyze sensor data in real-time to predict need for LSIs [life-saving interventions].”

(DARPA’s remit is performing “engineering alchemy” around research and development projects so far-fetched that they might not get explored elsewhere. It has a $4.1 billion budget for 2023. Its innovations have helped deliver the internet, as well as miniaturised GPS receivers of the kind that are now ubiquitous.)

“We are not prepared for a catastrophic mass casualty event”

Warfighters from the 129th Rescue Wing demonstrate casualty triage to students from a Hacking for Defense class at Stanford University. Credit: US Army

“We are not prepared for a catastrophic mass casualty event” said Thomas D. Kirsch, Professor and Director, National Center for Disaster Medicine and Public Health, Uniformed Services University

Accurate triage is difficult in any setting, but particularly after a mass casualty incident that involves dozens, hundreds, or even thousands of patients. Humans alone struggle under the chaos and stress - research has found that in mass casualty settings recommended triage methods are often not used or inaccurately applied. DARPA’s work to bring technology to triage to improve the capacity and accuracy of triage in both the battlefield and civilian settings is critical to improve our Nation’s preparedness.”

See also: MOD now has a dedicated AI centre operational

While any technology emerging from the research project may indeed prove life-saving – the advent of image recognition (e.g. here potentially used of blood loss)and democratisation of cheap sensors (temperature, etc) has meant such remote monitoring and anomaly detection tools are not unthinkable – early research shows broader readiness is also critical to minimising loss of life in such scenarios.

The project prepares to launch after earlier research found that even at moderate rather than mass casualty scale, nearly a quarter of Americans killed in action over 10 years (nearly 1,000 people) died of wounds that they could potentially have survived. In nine out of 10 cases, troops bled to death from wounds that might have been stanched. In 8%, soldiers died of airway damage that better care might have controlled.

Those findings were reported in the Journal of Trauma and Acute Care Surgery in 2012 – which found that more than 90% of aid stations in Afghanistan didn't put anti-bleeding agent tranexamic acid in medics' kits, only two-thirds possessed junctional tourniquets ssed to stop haemorrhaging in injuries too close to the victim's trunk for normal tourniquets, while just 12% of medics carried ketamine; recommended instead of morphine because it doesn’t trigger a drop in blood pressure or breathing.

Follow The Stack on LinkedIn