Introduction: Many point-of-care ultrasound (POCUS) machines now incorporate artificial intelligence (AI)-assisted acquisition features; however, their usefulness for the Focused Assessment with Sonography for Trauma (FAST) exam remains poorly defined. This repeated-measures study aimed to determine the immediate combined effect of AI auto-labeling and auto-grading on acquisition time and image quality of the right upper quadrant (RUQ) FAST exam window, as obtained by both novice and experienced emergency medicine (EM) physician trainees. Methods: Fourteen novices with limited POCUS training during medical school and 10 second- and third-year EM residents recorded RUQ windows with and without AI assistance on three standardized patients in randomized order. Acquisition time (in seconds) was compared using the Mann-Whitney U test. A chi-square analysis was used to compare the proportion of recordings meeting each of the three image quality criteria: visibility of all essential structures, correct image plane, and proper probe orientation. Results: The median (interquartile range) acquisition time was significantly longer with AI assistance for both novice trainees (85 (91) vs 53 (59) seconds; p < 0.01) and experienced trainees (44 (29) vs 28 (21) seconds; p < 0.01). All RUQ image quality criteria were significantly more likely to be met by experienced trainees than by novices. No significant differences in image quality were observed with versus without AI assistance within either group. Conclusion: The immediate effect of AI auto-labeling and auto-grading was an increase in acquisition time among both novice and experienced physician trainees obtaining RUQ FAST exam windows. AI assistance was not associated with an immediate improvement in image quality among either group. These findings suggest that such AI features are unlikely to aid physician trainees in acquiring RUQ FAST exam windows during isolated attempts.
Lenning et al. (Tue,) studied this question.