Assessing the Efficiency and Ergonomic Benefits of Artificial Intelligence–Assisted Acquisition
Objective: The aim of this study was to evaluate the effect of artificial intelligence (AI) assistance on examination efficiency and ergonomics during cardiac sonography. Materials and Methods: Randomized, cross-over study design was used with an ultrasound equipment system, cardiovascular software, and integrated AI features. A cohort of experienced sonographers (N = 30) performed standardized examinations with and without AI features enabled. All participating sonographers were new equipment users. Metrics of timing with and without AI assistance were acquired. Movements of the non-scanning hand were recorded during a subset of studies to assess time spent in a predefined ergonomic zone. Results: Artificial intelligence assistance reduced mean examination time by 10.16 ± 0.95 minutes (95% CI, 12.10-8.22; P < .001) independent of experience, age, scanning sequence, or prior AI familiarity. A sub-analysis of hand motion during sonography (n = 9) found AI assistance reduced average distance traveled by 50% (47.9 m) compared with unassisted examinations (SD, 10.6; 95% CI, 39.74-55.97, P < .01). Conclusion: In this cohort of users, AI-assisted cardiac sonography significantly reduced acquisition time and increased time spent within their optimal hand reach. This kind of upgrade may help lower the risk of work-related musculoskeletal injuries in cardiac sonographers. Further studies, based on actual clinical settings, would help to further clarify these benefits.
Authors
- LS. Lissa Sugeng (ORCID: https://orcid.org/0000-0001-6406-3690)
- Neil J. Weissman
- Steve Walling (ORCID: https://orcid.org/0009-0008-5716-3616)
- Alicia Armour
Institutions
- Duke University (US)
- Yale University (US)
- Saint Francis Hospital & Medical Center (US)
- MedStar Health (US)
Publication Details
- Journal
- Journal of diagnostic medical sonography
- Published
- 2026-10-09
- DOI
- https://doi.org/10.1177/87564793261489172
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00