Diagnostic imaging is essential for diagnosing and managing a variety of patient presentations. Frequently in the emergency department (ED), imaging is required and a rapid interpretation is demanded to aid in the acute management of a critically ill patient. As the utility of imaging has increased, and technology advances, it places a larger burden on emergency radiologists to produce accurate reports in a timely manner, especially in urgent situations [1,2]. The deployment of artificial intelligence (AI) has begun to assist radiologists in meeting these demands. AI is gaining momentum and enthusiasm for its utilization in a variety of industries and is reaching many areas of healthcare [3,4].

There have been copious examples reported of potential AI applications that can be used in the ED to aid in both interpretive (e.g., recognizing findings on imaging) and non-interpretive (e.g., workflow prioritization) uses, providing great assistance to ED physicians and radiologists [5,6]. For emergency radiology, there has been a sharp increase in recent literature and suggestions on the potential applications and usage of AI technology [7,8]. As radiology continues to be one of the most innovative medical specialties, AI infrastructure shows incredible promise towards improving radiological practice and patient outcomes [9].

Societal rumors suggest that AI and its applications may replace radiologists entirely, but this has been largely invalidated in the near-medium term. It is now held that AI may provide assistance and relief to the growing case load and increasing demands placed on radiologists. In a recent survey, a majority of radiologist respondents see AI as an opportunity to improve their practice all-round, including holding expectations for a lower error rate and interpretation time [10]. Although there are drawbacks to the deployment of AI, there is great potential for patient care to be improved and providing more rapid solutions to urgent problems.

The purpose of this review was to sum up the current advances of AI in emergency radiology, including current diagnostic applications (interpretive) and applications that stretch beyond imaging (non-interpretive), and analyze current drawbacks of AI in emergency radiology.