Work

Context matters: evaluating AI-assisted chest radiograph interpretation in military field hospitals

Year: 2025

Type: article

Abstract: Májovský et al investigate the utility of a Food and Drug Administration-cleared, commercially available artificial intelligence (AI)-assisted chest radiograph (CXR) interpretation platform in Role 2 military field hospital settings.1 The AI product has the capability to detect multiple findings on a CXR but this work focuses specifically on the trauma-related indications. Non-radiologist clinicians used the program to evaluate 159 CXRs, demonstrating a mean 9.4% absolute improvement in diagnostic accuracy.1 However, the degree of improvement ranged widely among the users ranging from <1% to 17%.1 Overall accuracy remained low: non-radiologist readers achieved 40% accuracy for all 10 elements evaluated, which improved to about 50% with AI assistance.1

Source: Trauma Surgery & Acute Care Open

DOI: 10.1136/tsaco-2025-002112

Author: Rachael A Callcut

Institutions: University of California Davis Medical Center, University of California, Davis

Language: English

Topic: Artificial Intelligence in Healthcare and Education

Subfield: Health Informatics

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $2,100