Ultra-compact AI X-ray unit slashes ICU bedside imaging time

Wednesday, December 2 | 9:30 a.m-10:30 a.m. | W3-SSER02-4 | Room E450A

An ultra-compact AI-powered x-ray system can cut bedside imaging workflow times in ICUS by more than 75% while preserving diagnostic performance, according to research in this session.

An ultra-compact AI-powered x-ray system can cut bedside imaging workflow times in intensive care units by more than three-quarters while preserving diagnostic performance, according to research in this session.

In a prospective study involving 121 ICU patients, Runqiu Huang, of Southern Medical University in Guangzhou, China, and colleagues compared the MiniAIR system (Poskom, South Korea), a 1.3-kilogram device roughly the size of a paperback book, with conventional mobile digital radiography for bedside imaging.

MiniAIR reduced total workflow time from 2,061 seconds to 475 seconds, a 76.9% decrease. Reductions were pronounced across all stages, particularly order initiation (87.5% faster) and post-acquisition processing (88.4% faster). AI image enhancement improved image quality, reducing noise scores and boosting spatial resolution from 34 to 40 line pairs per centimeter, according to the findings. Diagnostic agreement was high and non-inferior to conventional imaging, with intraclass correlation coefficients of at least 0.9987 and Cohen’s kappa values of 0.88 to 1.00 across all evaluated tasks.

Conventional mobile radiography has long constrained ICU workflows through bulky hardware and processing delays, the authors noted. Timely bedside imaging is critical in intensive care settings, where such inefficiencies can slow clinical decision-making, they added.

“This integrated hardware-AI approach addresses key limitations of conventional mobile radiography and supports rapid, point-of-care clinical decision-making,” the researchers suggested.

Grab a coffee and check out this session on AI research in emergency radiology to find out more details.

Page 1 of 1