Deployment of Artificial Intelligence in Radiology: Strategies for Success.
Publication/Presentation Date
2-1-2025
Abstract
Radiology, as a highly technical and information-rich medical specialty, is well suited for artificial intelligence (AI) product development, and many U.S. FDA-cleared AI medical devices are authorized for uses within the specialty. In this Clinical Perspective, we discuss the deployment of AI tools in radiology, exploring regulatory processes, the need for transparency, and other practical challenges. We further highlight the importance of rigorous validation, real-world testing, seamless workflow integration, and end user education. We emphasize the role for continuous feedback and robust monitoring processes, to guide AI tools' adaptation and help ensure sustained performance. Traditional standalone and alternative platform-based approaches to radiology AI implementation are considered. The presented strategies will help achieve successful deployment and fully realize AI's potential benefits in radiology.
Volume
224
Issue
2
First Page
2431898
Last Page
2431898
ISSN
1546-3141
Published In/Presented At
Jiang, S., Bukhari, S. M. A., Krishnan, A., Bera, K., Sharma, A., Caovan, D., Rosipko, B., & Gupta, A. (2025). Deployment of Artificial Intelligence in Radiology: Strategies for Success. AJR. American journal of roentgenology, 224(2), e2431898. https://doi.org/10.2214/AJR.24.31898
Disciplines
Diagnosis | Medicine and Health Sciences | Other Analytical, Diagnostic and Therapeutic Techniques and Equipment | Radiology
PubMedID
39475198
Department(s)
Department of Radiology and Diagnostic Medical Imaging
Document Type
Article