Raytheon, Kitware, Parallax, CACI International and the University of Maryland have been selected as performers for the US. Defense Advanced Research Projects Agency (DARPA) In the Moment (ITM) program, a program to develop artificial intelligence (AI) to make difficult triage medical decisions in austere environments and mass casualty events.
The selection of these performers was announced in a release from DARPA on June 6, which details the two-phase approach to addressing challenges in medical triage using an AI algorithm.
Difficult decisions in medical triage are those where trusted decision makers disagree, according to Dr. Matt Turek, DARPA’s ITM program manager and deputy director of the Information Innovation Office. Turk describes these decisions as decisions where there is no right answer and reaching a course of action is complicated by uncertainty, time pressure, and conflicting values.
The first phase of the ITM program will examine medical triage for small military units in austere environments using artificial intelligence to make treatment decisions, while the second phase will examine approaches to using eh technology in mass casualty events.
Raytheon BBN Technologies and Soar Technology, Inc. will collaborate to create decision-maker characterization approaches that will identify and quantify critical characteristics of human decision-makers in challenging domains.
Kitware, Inc. and Parallax Inc. they will create algorithmic decision makers that match key characteristics of reliable human decision makers.
“Key characteristics may include how an algorithm assesses a situation, how it relies on domain knowledge, how it responds to time pressures, and what principles or values it uses to prioritize care,” said Turek.
CACI International Inc. will develop and implement program evaluation, focusing on how key human characteristics can contribute to trusted decision-making delegation.
“From a technical perspective, the difficult decisions made in medical triage will likely require approaches that are not primarily based on training data to implement, as these approaches can be notoriously fragile,” Tarek said.
The University of Maryland Applied Research Laboratory for Intelligence and Security and the Institute for Defense Analyzes will be responsible for policy/practice integration and outreach, with ELSI experts advising throughout the study process.
Turek predicts that ITM improvements will eventually enable fully automated and semi-automated decision-making, with humans having the option to veto the algorithm.
In March, researchers at Edge Hill University launched a new AI-powered drone project for battlefield triage. Project ATRACT, which stands for A Trustworthy Robotic Autonomous system to support Casualty Triage, aims to create an aerial drone that can assist and accelerate triage in the critical minutes after trauma that shape the chances of survival on the battlefield.
Read the original at Defence247.gr