Artificial Intelligence as a Second Reader for Focused Point-of-Care Ul-trasound (POCUS) in the Emergency Department: A Scoping Review
Archives of Academic Emergency Medicine,
Vol. 14 No. 1 (2026),
1 October 2025
,
Page e60
https://doi.org/10.22037/aaem.v14i1.3030
Abstract
Introduction: Artificial intelligence (AI)/ machine learning (ML) as a second reader for focused point-of-care ultrasound (POCUS) may address two key emergency department (ED) challenges: operator variability and workflow delays. This review maps the existing literature on the application of AI/ML as a second reader for POCUS in the ED. Methods: We included studies on AI/ML for POCUS interpretation, acquisition guidance, or quality assessment in the ED/urgent care settings, with AI as a second reader supporting clinical decisions. We excluded fully autonomous AI, non-emergency settings, healthy volunteers, phantom data, and non-English publications. Two reviewers independently screened records from five databases (PubMed/MEDLINE, Scopus, Embase, Web of Science, IEEE Xplore; 2020– August 17 2026), extracting data on study characteristics, POCUS application, AI role, performance metrics, technology readiness level (TRL), and clinical implementation status. Results: Of 1,884 records identified, 44 studies met inclusion criteria. Lung ultrasound was the most frequent application (38.6%). AI’s primary role was as an interpretation assistant (77.3%). Models demonstrated high diagnostic performance (e.g., sensitivities ≥95% for hemoperitoneum and DVT detection), with one study achieving 100% sensitivity and specificity for pneumothorax detection, significantly outperforming 11 expert clinicians. However, most studies were at a preclinical validation stage (TRL: 4–5; 56.8%), with only 18 studies (40.9%) testing prototypes in live clinical workflows. Major limitations identified were small, single-center datasets, reliance on expert-acquired images, and lack of investigation into critical real-world challenges like domain shift and automation bias. Conclusions: AI for POCUS in the ED shows strong diagnostic performance in controlled research settings but remains distant from robust clinical integration. The field is characterized by a predominance of early-stage validation studies. Future research must prioritize multi-center, multi-device external validation, inclusion of novice-generated data, prospective workflow integration, and explicit safety evaluations to translate technical promise into reliable clinical impact.
- Artificial Intelligence
- Ultrasonography
- Point-of-Care Systems
- Emergency Service
- Hospital
- Triage
- Scoping Review
How to Cite
References
1. Chang H, Ko E, Lee JH, Kim M, Kim T, Shin TG, et al. Emer-gency department crowding: a national data report. Clin Exp Emerg Med. 2024;11(4):331-334. doi:10.15441/ceem.24.337.
2. Morley C, Unwin M, Peterson GM, Stankovich J, Kinsman L. Emergency department crowding: A systematic review of causes, consequences and solutions. PLoS One. 2018;13(8):e0203316. doi:10.1371/journal.pone.0203316
3. Alfoti BOO, Alfoti FOO, Alothman STH, Al-Dhafiri TMA, Al-Harbi NHM, Al-Khalidi AM, et al. Utilization of Point-of-Care Ultrasound (POCUS) in Emergency and Critical Care: Role of Nursing for Enhancing Diagnostic Accuracy and Efficiency-Systematic Review. Egyptian Journal of Chemistry 2024;67(13): 705-16.doi: 10.21608/ejchem.2024.297100.9851
4. Su J, Tie X, Wei Y, Zhou R, Zou T, Qin Y, et al. Critical care ultrasound: development, evolution, current and evolving clinical concepts in critical care medicine. Frontiers in medi-cine. 2025; 12:1622604.
5. Platt A, Peterson T, Peethumnongsin E, Kuchibhatla M, Theophanous RG. Prospective cohort study identifies barriers and facilitators to point-of-care ultrasound use in an academ-ic emergency department. BMJ Open Qual. 2026;15(3):e004210. doi:10.1136/bmjoq-2026-004210.
6. Gao Y, Zhang H, Wu C, Zhang Z, Pan L, He X, et al. Point of care ultrasound knowledge attitudes and practices among emergency and ICU healthcare providers. Sci Rep. 2025;15(1):33378.
7. Akkus Z, Cai J, Boonrod A, Zeinoddini A, Weston AD, Phil-brick KA, et al. A survey of deep-learning applications in ul-trasound: Artificial intelligence–powered ultrasound for im-proving clinical workflow. J Am Coll Radiol. 2019;16(9 Pt B):1318–1328. doi:10.1016/j.jacr.2019.06.004
8. Liu X, Faes L, Kale AU, Wagner SK, Fu DJ, Bruynseels A, et al. A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis. Lancet Digit Health. 2019;1(6):e271–e297. doi:10.1016/S2589-7500(19)30123-2.
9. Wistrom, R., Khait, L. & Nelson, G. Artificial Intelligence in Point-of-care Ultrasound. Curr Emerg Hosp Med Rep 12, 89–94 (2024). https://doi.org/10.1007/s40138-024-00293-8
10. Toporek GA, Wang H, inventors. US Patent 11,712,220 B2. 2023.
11. Kayarian F, Patel D, O'Brien JR, Schraft EK, Gottlieb M. Artificial intelligence and point-of-care ultrasound: Benefits, limitations, and implications for the future. Am J Emerg Med. 2024;80:119-22.
12. Kim S, Fischetti C, Guy M, Hsu E, Fox J, Young SD. Artifi-cial Intelligence (AI) Applications for Point of Care Ultra-sound (POCUS) in Low-Resource Settings: A Scoping Review. Diagnostics. 2024;14(15):1669.
13. Tricco AC, Lillie E, Zarin W, et al. PRISMA Extension for Scoping Reviews (PRISMA-ScR): Checklist and Explanation. Ann Intern Med. 2018;169(7):467-73.
14. Mankins JC. Technology readiness levels: a white paper. Washington, DC: NASA; 1995.
15. Leo MM, Potter IY, Zahiri M, Vaziri A, Jung CF, Feldman JA. Using Deep Learning to Detect the Presence and Location of Hemoperitoneum on the Focused Assessment with Sonog-raphy in Trauma (FAST) Examination in Adults. J Digit Imag-ing. 2023;36(5):2035-50. doi: 10.1007/s10278-023-00845-6
16. Yıldız Potter İ, Leo MM, Vaziri A, Feldman JA. Automated detection and localization of pericardial effusion from point-of-care cardiac ultrasound examination. Med Biol Eng Com-put. 2023;61(8):1947–1959. doi: 10.1007/s11517-023-02855-6
17. Kornblith AE, Addo N, Dong R, Rogers R, Grupp-Phelan J, Butte A, et al.. Development and Validation of a Deep Learn-ing Strategy for Automated View Classification of Pediatric FAST. J Ultrasound Med. 2022;41 (8):1915-24.
18. Cull J, Morrow D, Manasco C, Vaughan A, Eicken J, Smith H, et al. A quality assessment tool for focused abdominal so-nography for trauma examinations using artificial intelli-gence. J Trauma Acute Care Surg. 2025;98(1):111-6.
19. Ienghong K, Cheung LW, Gaysonsiri D, Apiratwarakul K. The diagnostic performance of automatic B-lines detection for evaluating pulmonary edema in the emergency depart-ment among novice point-of-care ultrasound practitioners. Emerg Radiol. 2025;32(2):241–246. doi:10.1007/s10140-025-02319-4.
20. Chaudhary R, Ho J, Smith D, Hossain S, Hargun J, VanBer-lo B, et al. Diagnostic accuracy of an automated classifier for the detection of pleural effusions in patients undergoing lung ultrasound.AmJEmergMed.2025;90:142-150.doi:10.1016/j.ajem.2025.01.041.
21. Clausdorff Fiedler H, Prager R, Smith D, Wu D, Dave C, Tschirhart J, et al. Automated real-time detection of lung slid-ing using artificial intelligence: a prospective diagnostic accu-racystudy.Chest.2024;166(2):362-370.doi:10.1016/j.chest.2024.02.011.
22. VanBerlo B, Derek Wu, Brian Li, Marwan A Rahman, Gregory Hogg, VanBerlo, ., et al. Accurate assessment of the lung sliding artefact on lung ultrasonography using a deep learning approach. Comput Biol Med.2022;148:105953.doi:10.1016/j.compbiomed.2022.105953.
23. Gottlieb M, Patel D, Viars M, Tsintolas J, Peksa GD, Bailitz J. Comparison of artificial intelligence versus real-time physi-cian assessment of pulmonary edema with lung ultrasound. Am J EmergMed.2023;70:109-112.doi:10.1016/j.ajem.2023.05.029.
24. Gazzoni M, La Salvia M, Torti E, Secco G, Perlini S, Le-porati F. Perceptive SARS-CoV-2 End-To-End Ultrasound Video Classification through X3D and Key-Frames Selection. Bioengineering(Basel).2023;10(3):282. doi:10.3390/bioengineering10030282
25. Goldsmith AJ, Jin M, Lucassen R, Duggan NM, Harrison NE, Wells W, et al. Comparison of pulmonary congestion se-verity using artificial intelligence-assisted scoring versus clin-ical experts: A secondary analysis of BLUSHED-AHF. Eur J Heart Fail. 2023;25(7):1166-1169. doi:10.1002/ejhf.2881.
26. Orosz G, Szabó RZ, Szabó M, Gyombolai P, Tóth JT, Ruttkay T, et al. Explainable transfer learning ensemble AI model for lung ultrasound pneumothorax detection with ex-pert benchmark. Scand J Trauma Resusc Emerg Med. 2026;34(1):95. doi:10.1186/s13049-026-01614-4.
27. Lin YC, Fang YH, Chen TY, Wang JT, Lin CM, Chen YS, et al. A Hierarchical CNN-Geometric Framework for Pleural Ef-fusion Severity Classification in Lung Ultrasound. IEEE Sens J. 2026;1-1. doi:10.1109/JSEN.2026.3721141.
28. Liu T, Zhou H, Wang JH, Zhang Y, Chen C, Chen Y, et al. HSD: Hough-Based Structure-Aware Detection of B-Lines in Lung Ultrasound. IEEE J Biomed Health Inform. 2026; Epub ahead of print. doi:10.1109/JBHI.2026.3684930.
29. Cote M, Smith D, Orozco N, Huggard B, Wu B, Tran K, et al. Diagnostic Accuracy of Automated Pneumothorax Detec-tion Via Novice-Acquired Ultrasound After Chest Tube Re-moval: Comparison With Expert Interpretation and Chest Radiography. CHEST Pulmonary. 2026;4(2):100246. doi: 10.1016/j.chpulm.2026.100246
30. Gottlieb M, Schraft E, O'Brien J, Patel D. Diagnostic accu-racy of artificial intelligence for identifying systolic and dias-tolic cardiac dysfunction in the emergency department. Am J Emerg Med. 2024;86:115-119. doi:10.1016/j.ajem.2024.10.019.
31. He B, Dash D, Duanmu Y, Tan TX, Ouyang D, Zou J. AI-enabled assessment of cardiac function and video quality in emergency department point-of-care echocardiograms. J EmergMed.2024;66(2):184-191.doi:10.1016/j.jemermed.2023.02.005.
32. Oikonomou EK, Vaid A, Holste G, Coppi A, McNamara RL, Baloescu C, et al. Artificial intelligence-guided detection of under-recognised cardiomyopathies on point-of-care cardiac ultrasonography: a multicentre study. Lancet Digit Health. 2025;7(2):e113-e123. doi:10.1016/S2589-7500(24)00249-8.
33. Lin X, Yang F, Chen Y, Chen X, Wang W, Chen X, et al. Echocardiography-based AI detection of regional wall motion abnormalities and quantification of cardiac function in myo-cardial infarction. Front Cardiovasc Med. 2022;9:903660. doi:10.3389/fcvm.2022.903660
34. Chou HH, Chang YC, Lien WC, Lin LC, Lin XZ, Hsu TE, et al. Efficient Deep Learning Models Revolutionize Doctor’s Training for Point-of-Care Ultrasound. IEEE Access. 2025;13:76038-76046. doi:10.1109/ACCESS.2025.3562674.
35. Shan C, Guo G, Shi Z, Qi Z, Chen F, Cheng H, et al. ESPNet: an automated left ventricular ejection fraction measurement via multi-frame temporal aggregation. IEEE J Biomed Health Inform. 2026; In press. doi:10.1109/JBHI.2026.3715789.
36. Blaivas M, Blaivas LN, Tsung JW. Deep learning algorithm performance compared to experts in visual evaluation of in-ferior vena cava collapse on ultrasound to determine intra-venous fluid need in dehydration management. Signa Vitae. 2021;17(5):34.
37. Wshah S, Xu B, Steinharter J, Reilly C, Morrissette K. Classification of clinically relevant intravascular volume sta-tus using point of care ultrasound and machine learning. J Med Imaging (Bellingham). 2022;9(5):054502.
38. Crockett D, Kelly C, Brundage J, Jones J, Ockerse P. A stress test of artificial intelligence: can deep learning models trained from formal echocardiography accurately interpret point-of-care ultrasound? J Ultrasound Med. 2022;41(12):3003–3012. doi:10.1002/jum.16003.
39. Avgerinos E, Spiliopoulos S, Psachoulia F, Yfantis A, Plakas G, Grigoriadis S, et al. Novel artificial intelligence guid-ed non-expert compression ultrasound deep vein thrombosis diagnostic pathway may reduce vascular laboratory venous testing. Eur J Vasc Endovasc Surg. 2025;70(4):517-522. doi:10.1016/j.ejvs.2025.04.070.
40. Oppenheimer J, Mandegaran R, Staabs F, Adler A, Sin-göhl S, Kainz B, et al. Remote expert DVT triaging of novice-user compression sonography with AI-guidance. Ann Vasc Surg. 2024;99:272-279. doi:10.1016/j.avsg.2023.08.022.
41. Speranza G, Mischkewitz S, Al-Noor F, Kainz B. Value of clinical review for AI-guided deep vein thrombosis diagnosis with ultrasound imaging by non-expert operators. NPJ Digit Med. 2025;8(1):135. doi:10.1038/s41746-025-01518-0.
42. Nothnagel K, Aslam MF. Evaluating the benefits of ma-chine learning for diagnosing deep vein thrombosis com-pared with gold standard ultrasound: a feasibility study. BJGP Open. 2025;8(4): BJGPO.2024.0057. doi: 10.3399/BJGPO.2024.0057.
43. Park, S., Yoon, H., Yeon Kang, S., Joon Jo, I., Heo, S., Chang, H., Eun Park, J et al. Artificial intelligence-based eval-uation of carotid artery compressibility via point-of-care ul-trasound in determining the return of spontaneous circula-tion during cardiopulmonary resuscita-tion.Resuscitation.2024;202:110302. doi:10.1016/j.resuscitation.2024.110302.
44. Blaivas M, Blaivas LN, Tsung JW. Deep Learning Pitfall: Impact of Novel Ultrasound Equipment Introduction on Algo-rithm Performance and the Realities of Domain Adaptation. J UltrasoundMed.2022;41(4):855-863.doi:10.1002/jum.15765.
45. Zhou Y, Knight J, Alves-Pereira F, Keen C, Hareendrana-than AR, Jaremko JL. Wrist and elbow fracture detection and segmentation by artificial intelligence using point-of-care ul-trasound.JUltrasound.2025;28(2):465-473.Doi:10.1007/s40477-025-01019-6.
46. Zhang J, Boora N, Melendez S, Rakkunedeth Ha-reendranathan A, Jaremko J. Diagnostic Accuracy of 3D Ul-trasound and Artificial Intelligence for Detection of Pediatric Wrist Injuries. Chil-dren(Basel).2021;8(6):431.doi:10.3390/children8060431.
47. Madhusoodanan M, Panicker MR, Gopalakrishnan PH, Hareendranathan AR. Beam-net: a deep learning framework with bone enhancement attention mechanism for high reso-lution high frame rate ultrasound beamforming. arXiv pre-print arXiv:2507.15306. 2025. doi:10.48550/arXiv.2507.15306.
48. Song F, Liu H, Ma H, Chen X, Wang S, Qin T, et al. AI model based on diaphragm ultrasound to improve the pre-dictive performance of invasive mechanical ventilation wean-ing: prospective cohort study. JMIR Form Res. 2025;9:e72482. doi: 10.2196/72482
49. Wu C-Y, Li J-D, Shih P-Y, Huang C-C, Cheng H-L, Wu C-Y, et al. Real-Time Identification of Cricothyrotomy Landmarks in Emergency Care and Obstetric Patients Using Wireless Handheld Ultrasound and Edge-Computing Artificial Intelli-gence: A Prospective Observational Study. J Med Syst. 2025;49(1):131.
50. Wang Y, Ye R, Cai Q, Wu Y, Gong L, Li Z, et al. Deep Learning for Ultrasound-Based Auxiliary Diagnosis of Emer-gency Ascites. Ultrasound Med Biol. 2026;52(4):782-8.
51. Komatsu M, Teraya N, Natsume T, Harada N, Takeda K, Hamamoto R. Clinical Application of Artificial Intelligence in Ultrasound Imaging for Oncology. Jma j. 2025;8(1):18-25.
52. Zeng H, Zou K, Chen Z, Gao Y, Chen H, Zhang H, et al. Training‑Free Image Style Alignment for Domain Shift on Handheld Ultrasound Devices. IEEE Trans Med Imaging. 2025;44(4):1942-52.
53. Shaddock L, Smith T. Potential for Use of Portable Ultra-sound Devices in Rural and Remote Settings in Australia and Other Developed Countries: A Systematic Review. J Multidis-cip Healthc. 2022;15:605-25.
54. Sorensen B, Hunskaar S. Point-of-care ultrasound in pri-mary care: a systematic review of generalist performed point-of-care ultrasound in unselected populations. Ultra-sound J. 2019;11(1):31.
55. Berkhout WEM, van Wijngaarden JJ, Workum JD, van de Sande D, Hilling DE, Jung C, et al. Operationalization of Artifi-cial Intelligence Applications in the Intensive Care Unit: A Systematic Review. JAMA Netw Open. 2025;8(7):e2522866.
56. Mehta A, Motavaf M, Raza D, Jairath N, Pulavarty A, Xu Z, et al. Deep Learning Image Processing Models in Dermato-pathology. Diagnostics (Basel). 2025;15(19):2517. doi:10.3390/diagnostics15192517.
57. John A, Alhajj R, Rokne J. A systematic review of AI as a digital twin for prostate cancer care. Comput Methods Pro-grams Biomed. 2025;268:108804.
58. Goddard K, Roudsari A, Wyatt J. Decision support and automation bias: methodology and preliminary results of a systematic review. Stud Health Technol Inform. 2011;164:3-7.
59. Lyell D, Coiera E. Automation bias and verification com-plexity: a systematic review. Journal of the American Medical Informatics Association: JAMIA. 2017;24(2):423-31.
60. Goddard K, Roudsari A, Wyatt JC. Automation bias: a sys-tematic review of frequency, effect mediators, and mitiga-tors. Journal of the American Medical Informatics Associa-tion: JAMIA. 2012;19(1):121-7.
61. Thyagachandran A, Lefchak B, Murthy HA, Bergmann KR, Madhok M. Development of a deep learning model for in-tussusception using point-of-care ultrasound. Frontiers in radiology. 2026;6:1812587.
62. Marini TJ, Nayaruparambil AA, Baran TM, Murali T, Dru-ry S, Zhao YT, et al. New lung ultrasound system for rapid triage of pulmonary disease without a radiologist or sonog-rapher. BMC Pulm Med. 2025;25(1):499.
63. Russell FM, Ehrman RR, Barton A, Sarmiento E, Ottenhoff JE, Nti BK. B-line quantification: comparing learners novice to lung ultrasound assisted by machine artificial intelligence technology to expert review. Ultrasound J. 2021;13(1):33.
64. Pare JR, Gjesteby LA, Tonelli M, Leo MM, Muruganandan KM, Choudhary G, et al. Transfer Learning-Based B-Line As-sessment of Lung Ultrasound for Acute Heart Failure. Ultra-sound Med Biol. 2024;50(6):825-832. doi: 10.1016/j.ultrasmedbio.2024.02.004
65. Wang E, Wu J, Gao S, Zhou P, Chen Y, Zhou G. SM-FSL: Similarity Guided Multi-Source Few-Shot Learning for Lung Ultrasound Diagnosis. IEEE J Biomed Health Inform. 2026; Epub ahead of print. doi:10.1109/JBHI.2026.3721499.
66. Holste G, Oikonomou EK, Tokodi M, Kovács A, Wang Z, Khera R. Complete AI-Enabled Echocardiography Interpreta-tion With Multitask Deep Learning. JAMA. 2025;334(4):306-318. doi:10.1001/jama.2025.8731.
67. Saadat A, Hashemi N, Khodabakhshian B, Tsang MY, Lu-ong C, Tsang TSM, et al. Optimizing point-of-care ultrasound video acquisition for probabilistic multi-task heart failure de-tection. Int J Comput Assist Radiol Surg. 2026;21(5):1151-1159. doi:10.1007/s11548-026-03680-6.
68. Maeng SJ, Park S, Jo IJ, Lee G, Hwang SY, Chung MJ, et al. AI-driven carotid artery compressibility assessment via point-of-care ultrasound for blood pressure estimation in critically ill and post-resuscitation patients: a prospective ob-servational study. Crit Care. 2026;30:2. doi:10.1186/s13054-025-05772-2 .
- Abstract Viewed: 25 times
- pdf Downloaded: 17 times
Download Statastics
Make a Submission
SJR
COPE
