AMBUSH

Automated Burn Diagnostic System for Healthcare

Approximately 1.25 million people in the United States are treated each year for burn injuries. About 40,000 are hospitalized for the treatment of these injuries resulting in approximately $7.9 billion in costs per year. In the military setting, burn casualties correspond to approximately 8% of combat-related injuries. For this reason, precise burn injury analysis is an important aspect of the medical AI field.

The AMBUSH project seeks to discover new ways of diagnosing burn wounds by developing Artificial Intelligence (AI) systems that interact with doctors for the good of the patient. Specifically, AMBUSH will develop and deliver an automated assessment system for burn injuries in austere and normal environments — a task requiring novel developments in the medical and computer science domains.

On the computer science side, AMBUSH will develop algorithms and methodologies for the automated analysis of burn data. This includes work in classification of burn depth, segmentation of burn images, prediction of developing burn characteristics, and explainability of burn AI models. This will be done using modern machine learning techniques as well as novel algorithms carefully designed with the burn injury domain in mind.

AMBUSH is a collaboration between Purdue University, Indiana University School of Medicine, and the US Army Institute of Surgical Research. AMBUSH is funded by the Department of Defense.

This work was supported by the Office of the Assistant Secretary of Defense for Health Affairs and the Defense Health Agency J9, Research and Development Directorate, or the U.S. Army Medical Research Acquisition Activity at the U.S. Army Medical Research and Development Command, in the amount of $2M through the Combat Readiness Medical Research Program under Award No. W81XWH-21-2-0030. Opinions, interpretations, conclusions and recommendations are those of the author and are not necessarily endorsed by the Department of Defense.