Clinical impact of AI in radiology department management: a systematic review

Abstract

Abstract Purpose Artificial intelligence (AI) has revolutionized medical diagnosis and treatment. Breakthroughs in diagnostic applications make headlines, but AI in department administration (admin AI) likely deserves more attention. With the present study we conducted a systematic review of the literature on clinical impacts of admin AI in radiology. Methods Three electronic databases were searched for studies published in the last 5 years. Three independent reviewers evaluated the records using a tailored version of the Critical Appraisal Skills Program. Results Of the 1486 records retrieved, only six met the inclusion criteria for further analysis, signaling the scarcity of evidence for research into admin AI. Conclusions Despite the scarcity of studies, current evidence supports our hypothesis that admin AI holds promise for administrative application in radiology departments. Admin AI can directly benefit patient care and treatment outcomes by improving healthcare access and optimizing clinical processes. Furthermore, admin AI can be applied in error-prone administrative processes, allowing medical professionals to spend more time on direct clinical care. The scientific community should broaden its attention to include admin AI, as more real-world data are needed to quantify its benefits. Limitations This exploratory study lacks extensive quantitative data backing administrative AI. Further studies are warranted to quantify the impacts.