Diagnostic performance of artificial intelligence in fracture detection with ultrasound: A systematic review and meta-analysis
Keywords:
Bone fractures, Artificial intelligence, point-of-care ultrasound, Musculoskeletal imaging, Deep learning modelsAbstract
Background: Bone fractures are common in acute care, and point-of-care ultrasound (POCUS) is an emerging diagnostic tool that can be complementary to or even in some cases an alternative to X-ray imaging. With the rise of artificial intelligence (AI) model incorporation in diagnostic interpretations, this systematic review and meta-analysis aimed to evaluate the diagnostic performance of deep learning models for detecting bone fractures on ultrasound images, using radiographic imaging with expert interpretation as the reference standard.
Methods: Comprehensive literature searches were conducted from inception to 15 August 2025, in databases including MEDLINE (Ovid interface), Embase (Ovid interface), CINAHL Plus with Full Text (EBSCOhost interface), Web of Science, ACM Digital Library, Scopus, and Google Scholar. We included studies evaluating deep learning models applied to ultrasound images or sweeps for fracture classification. The included studies predominantly evaluated pediatric cohorts presenting to emergency departments with suspected upper extremity fractures. Sensitivity and specificity were pooled using a bivariate random-effects model on the logit scale, and summary receiver operating characteristic (sROC) curve was constructed. Risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool.
Results: A total of 580 papers were identified in the preliminary literature search, and 333 studies were screened after duplicate removal. Screening 27 full-text articles resulted in eight studies that were included in the review. Across 21 reported model evaluations from five studies with extractable diagnostic data, pooled sensitivity was 0.77 (95% CI: 0.69-0.83) and pooled specificity was 0.90 (95% CI: 0.86-0.94), with an overall sROC AUC of 0.91. Patient-level evaluations yielded higher pooled sensitivity (0.89; 95% CI: 0.81-0.94) and specificity (0.94; 95% CI: 0.81-0.98) compared to video-level analyses.
Conclusions: AI-assisted ultrasound demonstrates promising diagnostic performance for fracture detection with generally high specificity and variable sensitivity across architectures and anatomic targets. Future work should prioritize prospective, externally validated studies with standardized acquisition protocols and clinically meaningful reporting at the patient level.
References
1. Miller AN. Evaluation of common musculoskeletal injuries in the urgent setting. MedEdPORTAL 2016 Dec 7. doi: 10.15766/mep_2374-8265.10514
2. Kuo RYL, Harrison C, Curran TA, Jones B, Freethy A, Cussons D, et al. Artificial intelligence in fracture detection: a systematic review and meta-analysis. Radiology 2022;304:50-62. doi: 10.1148/radiol.211785
3. Orces CH, Orces J. Trends in the U.S. childhood emergency department visits for fall-related fractures, 2001-2015. Cureus 2020;12:e11629. doi: 10.7759/cureus.11629
4. Al-Naser Y, Alshadeedi F. Bringing imaging to the people: enhancing access and equity in healthcare through mobile imaging. J Med Imaging Radiat Sci 2024 Dec 1. doi: 10.1016/j.jmir.2024.101715
5. Pinto A, Berritto D, Russo A, Riccitiello F, Caruso M, Belfiore MP, et al. Traumatic fractures in adults: missed diagnosis on plain radiographs in the emergency department. Acta Biomed 2018;89 Suppl 1:111-23. doi: 10.23750/abm.v89i1-s.7015
6. Al-Baghdadi R, Yu TJ, Ferro A, Roberts J. Radiography versus sonography in the detection of acute fractures: a case series. J Diagn Med Sonogr 2021;37:400-6. doi: 10.1177/8756479321996314
7. Lin EC. Radiation risk from medical imaging. Mayo Clin Proc 2010;85:1142-6. doi: 10.4065/mcp.2010.0260
8. Du MJ, Lin YH, Chen WT, Zhao H. Advances in the application of ultrasound for fracture diagnosis and treatment. Eur Rev Med Pharmacol Sci 2022;26:7949-54. doi: 10.26355/eurrev_202211_30146
9. Tsou PY, Ma YK, Wang YH, Gillon JT, Rafael J, Deanehan JK. Diagnostic accuracy of ultrasound for upper extremity fractures in children: a systematic review and meta-analysis. Am J Emerg Med 2021;44:383-94. doi: 10.1016/j.ajem.2020.04.071
10. Atilla OD, Yesilaras M, Kilic TY, Tur FC, Reisoglu A, Sever M, et al. The accuracy of bedside ultrasonography as a diagnostic tool for fractures in the ankle and foot. Acad Emerg Med 2014;21:1058-61. doi: 10.1111/acem.12467
11. Bozorgi F, Shayesteh Azar M, Montazer SH, Chabra A, Heidari SF, Khalilian A. Ability of ultrasonography in detection of different extremity bone fractures; a case series study. Emergency 2017;5:e15. doi: 10.22037/emergency.v5i1.12821
12. Snelling PJ, Jones P, Bade D, Bindra R, Byrnes J, Davison M, et al. Ultrasonography or radiography for suspected pediatric distal forearm fractures. N Engl J Med 2023;388:2049-57. doi: 10.1056/nejmoa2213883
13. Ipsen S, Wulff D, Kuhlemann I, Schweikard A, Ernst F. Towards automated ultrasound imaging: robotic image acquisition in liver and prostate for long-term motion monitoring. Phys Med Biol 2021;66. doi: 10.1088/1361-6560/abf277
14. Li K, Xu Y, Meng MQH. An overview of systems and techniques for autonomous robotic ultrasound acquisitions. IEEE Trans Med Robot Bionics 2021;3:510-24. doi: 10.1109/tmrb.2021.3072190
15. Kojcev R, Khakzar A, Fuerst B, Zettinig O, Fahkry C, DeJong R, et al. On the reproducibility of expert-operated and robotic ultrasound acquisitions. Int J Comput Assist Radiol Surg 2017;12:1003-11. doi: 10.1007/s11548-017-1561-1
16. Gan K, Xu D, Lin Y, Shen Y, Zhang T, Hu K, et al. Artificial intelligence detection of distal radius fractures: a comparison between the convolutional neural network and professional assessments. Acta Orthop 2019;90:394-400. doi: 10.1080/17453674.2019.1600125
17. Nguyen T, Maarek R, Hermann AL, Kammoun A, Marchi A, Khelifi-Touhami MR, et al. Assessment of an artificial intelligence aid for the detection of appendicular skeletal fractures in children and young adults by senior and junior radiologists. Pediatr Radiol 2022;52:2215-26. doi: 10.1007/s00247-022-05496-3
18. Tenajas R, Miraut D, Illana CI, Alonso-Gonzalez R, Arias-Valcayo F, Herraiz JL. Recent advances in artificial intelligence-assisted ultrasound scanning. Appl Sci 2023;13:3693. doi: 10.3390/app13063693
19. Baloescu C, Bailitz J, Cheema B, Agarwala R, Jankowski M, Eke O, et al. Artificial intelligence-guided lung ultrasound by nonexperts. JAMA Cardiol 2025;10:245-53. doi: 10.1001/jamacardio.2024.4991
20. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. PLOS Med 2021;18:e1003583. doi: 10.1136/bmj.n71
21. Whiting PF, Rutjes AWS, Westwood ME, Mallett S, Deeks JJ, Reitsma JB, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med 2011;155:529-36. doi: 10.7326/0003-4819-155-8-201110180-00009
22. Knight J, Zhou Y, Keen C, Hareendranathan AR, Alves-Pereira F, Ghasseminia S, et al. 2D/3D ultrasound diagnosis of pediatric distal radius fractures by human readers vs artificial intelligence. Sci Rep 2023;13:14535. doi: 10.1038/s41598-023-41807-w
23. Großbröhmer C, Bartram L, Rheinbay C, Heinrich MP, Tüshaus L. Leveraging semantic information for sonographic wrist fracture assessment within children. In: Deserno TM, Handels H, Maier A, Maier-Hein K, Palm C, Tolxdorff T, editors. Bildverarbeitung für die Medizin 2023. Wiesbaden: Springer Fachmedien; 2023. p. 102-7. doi: 10.1007/978-3-658-41657-7_23
24. Hareendranathan AR, Tripathi A, Panicker MR, Zhang J, Boora N, Jaremko J. Deep learning approach for automatic wrist fracture detection using ultrasound bone probability maps. SN Compr Clin Med 2023;5:276. doi: 10.1007/s42399-023-01608-8
25. Zhou Y, Hareendranathan AR, Keen C, Knight J, Jaremko J. Wrist ultrasound segmentation by deep learning. Artif Intell Med 2022 Jul 9. doi: 10.1007/978-3-031-09342-5_22
26. Tripathi A, Panicker M, Zhang J, Boora N, Jaremko J, Hareendranathan AR. Domain specific transporter framework to detect fractures in ultrasound. Annu Int Conf IEEE Eng Med Biol Soc 2024;2024:1-4. doi: 10.1109/embc53108.2024.10782947
27. Teng Y, Pan D, Zhao W. Application of deep learning ultrasound imaging in monitoring bone healing after fracture surgery. J Radiat Res Appl Sci 2023;16:100493. doi: 10.1016/j.jrras.2022.100493
28. Zhang J, Boora N, Melendez S, Hareendranathan AR, Jaremko J. Diagnostic accuracy of 3D ultrasound and artificial intelligence for detection of pediatric wrist injuries. Children 2021;8:431. doi: 10.3390/children8060431
29. Zhou Y, Knight J, Alves-Pereira F, Keen C, Hareendranathan AR, Jaremko JL. Wrist and elbow fracture detection and segmentation by artificial intelligence using point-of-care ultrasound. J Ultrasound 2025;28:465-73. doi: 10.1007/s40477-025-01019-6
30. Sivrikaya S, Aksay E, Bayram B, Oray NC, Karakasli A, Altintas E. Emergency physicians performed point-of-care ultrasonography for detecting distal forearm fracture. Turk J Emerg Med 2016;16:98-101. doi: 10.1016/j.tjem.2016.04.002
31. Snelling PJ, Jones P, Bade D, Bindra R, Davison M, Gillespie A, et al. Diagnostic accuracy of point-of-care ultrasound versus radiographic imaging for pediatric distal forearm fractures: a randomized controlled trial. Ann Emerg Med 2024;83:198-207. doi: 10.1016/j.annemergmed.2023.10.008
32. Yousefifard M, Baikpour M, Ghelichkhani P, Asady H, Darafarin A, Amini Esfahani MR, et al. Comparison of ultrasonography and radiography in detection of thoracic bone fractures; a systematic review and meta-analysis. Emergency 2016;4:55-64.
33. Haak SL, Vos H, Borgstede MG, Boendermaker AE, Rietveld V, Kroon T, et al. Diagnostic accuracy of point-of-care ultrasound in detecting clavicle fractures. Am J Emerg Med 2025;88:156-61. doi: 10.1016/j.ajem.2024.11.008
34. Ozturk P, Aksay E, Oray NC, Bayram B, Basci O, Tokgoz D. Emergency physician accuracy using ultrasonography to diagnose lateral malleolar fracture. Am J Emerg Med 2018;36:362-5. doi: 10.1016/j.ajem.2017.08.020
35. Kuo RYL, Harrison C, Curran TA, Jones B, Freethy A, Cussons D, et al. Artificial intelligence in fracture detection: a systematic review and meta-analysis. Radiology 2022;304:50-62. doi: 10.1148/radiol.211785
36. Thian YL, Li Y, Jagmohan P, Sia D, Chan VEY, Tan RT. Convolutional neural networks for automated fracture detection and localization on wrist radiographs. Radiol Artif Intell 2019;1:e180001. doi: 10.1148/ryai.2019180001
37. Brattain LJ, Telfer BA, Dhyani M, Grajo JR, Samir AE. Machine learning for medical ultrasound: status, methods, and future opportunities. Abdom Radiol 2018;43:786-99. doi: 10.1007/s00261-018-1517-0
38. Troxler D, Sanchez C, de Trey T, Mayr J, Walther M. Non-inferiority of point-of-care ultrasound compared to radiography to diagnose upper extremity fractures in children. Children 2022;9:1496. doi: 10.3390/children9101496
39. Morello R, Mariani F, Snelling PJ, Buonsenso D. Point-of-care ultrasound for the diagnosis of distal forearm fractures in children and adolescents: a scoping review. Eur J Pediatr 2024;184:19. doi: 10.1007/s00431-024-05877-w
40. Snelling PJ, Keijzers G, Byrnes J, Bade D, George S, Moore M, et al. Bedside ultrasound conducted in kids with distal upper limb fractures in the emergency department (BUCKLED): a protocol for an open-label non-inferiority diagnostic randomised controlled trial. Trials 2021;22:282. doi: 10.1186/s13063-021-05239-z
41. Epema AC, Spanjer MJB, Ras L, Kelder JC, Sanders M. Point-of-care ultrasound compared with conventional radiographic evaluation in children with suspected distal forearm fractures in the Netherlands: a diagnostic accuracy study. Emerg Med J 2019;36:613-6. doi: 10.1136/emermed-2018-208380
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Michael Xie, Justin JY Kim, James HB Im, Liz Dennett, Abhilash Hareendranathan, Jacob L Jaremko (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Authors retain the copyright for their published work. No formal permission will be required to reproduce parts (tables or illustrations) of published papers, provided the source is quoted appropriately and reproduction has no commercial intent.
How to Cite
