Artificial Intelligence/Machine Learning Education in Radiology: Multi-institutional Survey of Radiology Residents in the United States.

RATIONALE AND OBJECTIVES: To evaluate radiology residents' perspectives regarding inclusion of artificial intelligence/ machine learning (AI/ML) education in the residency curriculum. MATERIALS AND METHODS: An online anonymous survey was sent to 759 residents at 21 US radiology residency programs. Resident demographics, sub-specialty interests, educational background and research experiences, as well as the awareness, availability, and usefulness of various resources for AI/ML education were collected. RESULTS: The survey response rate was 27% (209/759). A total of 74% of respondents were male, 80% were training at large university programs, and only a minority (<20) had formal education or research experience in AI/ML. All four years of training were represented (range: 20%-38%). The majority of the residents agreed or strongly agreed (83%) that AI/ML education should be a part of the radiology residency curriculum and that such education should equip them with the knowledge to troubleshoot an AI tool in practice / determine whether a tool is working as intended (82%). Among the residency programs that offer AI/ML education, the most common resources were lecture series (43%), national informatics courses (28%), and in-house/institutional courses (26%). About 24% of the residents reported no AI/ML educational offerings in their residency curriculum. Hands on AI/ML laboratory (67%) and lecture series (61%) were reported as the most beneficial or effective. The majority of the residents preferred AI/ML education offered as a continuous course spanning the radiology residency (R1 to R4) (76%), followed by mini fellowship during R4 (32%) and as a course during PGY1 (21%). CONCLUSION: Residents largely favor the inclusion of formal AI/ML education in the radiology residency curriculum, prefer hands-on learning and lectures as learning tools, and prefer a continuous AI/ML course spanning R1-R4.

Utilization of Chest and Abdominopelvic CT for Traumatic Injury From 2011 to 2018: Evaluation Using a National Commercial Database.

BACKGROUND. Increases in the use of CT to evaluate patients presenting with trauma have raised concern about inappropriate imaging. The evolving utilization of CT for trauma evaluation may be impacted by injury severity. OBJECTIVE. The purpose of this study was to explore patterns in utilization of chest and abdominopelvic CT among trauma-related emergency department (ED) visits across the United States. METHODS. This retrospective study was conducted with national commercial claims information extracted from the MarketScan Commercial Database. Trauma-related ED encounters were identified from the 2011-2018 MarketScan database files and classified by injury severity score (minor, intermediate, and major injuries) on the basis of International Classification of Diseases codes. ED encounters were also assessed for chest CT, abdominopelvic CT, and single-encounter chest and abdominopelvic CT examinations. Utilization per 1000 trauma-related ED encounters was determined. Multivariable Poisson regression models were used to determine incidence rate ratios (IRRs) as a measure of temporal changes in utilization. RESULTS. From 2011 to 2018, 8,369,092 trauma-related ED encounters were identified (5,685,295 for minor, 2,624,944 for intermediate, and 58,853 for major injuries). Utilization of chest CT per 1000 trauma-related ED encounters increased from 4.9 to 13.5 examinations (adjusted IRR, 1.15 per year; minor injuries, from 2.2 to 7.7 [adjusted IRR, 1.17]; intermediate injuries, from 8.5 to 21.5 [adjusted IRR, 1.16]; major injuries, from 117.8 to 200.1 [adjusted IRR, 1.08]). Utilization of abdominopelvic CT per 1000 trauma-related ED encounters increased from 7.5 to 16.4 (adjusted IRR, 1.12; minor injuries, 4.8 to 12.2 [adjusted IRR, 1.13]; intermediate injuries, 10.6 to 21.7 [adjusted IRR, 1.13]; major injuries, 134.8 to 192.6 [adjusted IRR, 1.07]). Utilization of single-encounter chest and abdominopelvic CT per 1000 trauma-related ED encounters increased from 3.4 to 8.9 [adjusted IRR, 1.16; minor injuries, 1.1 to 4.6 [adjusted IRR, 1.18]; intermediate injuries, 6.4 to 16.4 [adjusted IRR, 1.16]; major injuries, 99.6 to 179.9 [adjusted IRR, 1.08]). CONCLUSION. National utilization of chest and abdominopelvic CT for trauma-related ED encounters increased among commercially insured patients from 2011 to 2018, particularly for single-encounter chest and abdominopelvic CT examinations and for minor injuries. CLINICAL IMPACT. Given concerns about increased cost and detection of incidental findings, further investigation is warranted to explore the potential benefit of single-encounter chest and abdominopelvic CT examinations of patients with minor injuries and to develop strategies for optimizing appropriateness of imaging orders.