The Relationship between the Use of Artificial Intelligence and Waiting Time for Emergency Chest X-Ray Interpretation: A Systematic Review of Randomized Controlled Trials and Primary Studies
Keywords:
Artificial Intelligence, Emergency Radiology, Chest Radiography, Report Turnaround Time, Worklist PrioritizationAbstract
Introduction: Delayed interpretation of emergency chest X-rays (CXRs) represents a critical bottleneck in emergency department (ED) workflow, potentially compromising outcomes in time-sensitive conditions. Artificial intelligence (AI) algorithms have demonstrated capacity for automated detection and prioritization of radiographic findings, offering a potential solution to interpretation delays.
Methods: A systematic search adhered to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) 2020 guidelines for studies evaluating AI-assisted CXR interpretation in emergency settings reporting time-related outcomes. Risk of bias was assessed using QUADAS-2 and RoB 2.0 tools. Seventeen primary studies encompassing 16 distinct AI algorithms and over 1.2 million CXR images were included.
Results: AI implementation was consistently associated with significant reductions in mean report turnaround time (RTAT; range: 2.7–29.7 minutes reduction) and radiologist reading time (average reduction 13–27%). Active worklist prioritization reduced RTAT for critical findings from 80.1 to 35.6 minutes (p<0.0001) and reduced average reporting delay from 11.2 to 2.7 days. AI-assisted interpretation improved diagnostic sensitivity for pneumothorax (AUC 0.97), pleural effusion (AUC 0.94–0.97), and consolidation (AUC 0.83–0.88). Non-radiology emergency clinicians derived greatest benefit, with pneumothorax AUC improving from 0.846 to 0.974 with AI assistance. Generative AI models produced reports of equivalent clinical accuracy to radiologists while improving documentation efficiency by 15.5%.
Discussion: AI integration significantly reduces RTAT and optimizes worklist prioritization, particularly for critical findings. Active worklist re-ordering algorithms produced greater RTAT reductions (43.7%) than passive notification systems (7.6%). Algorithmic bias against underserved populations and false-negative safety risks mandate mandatory equity auditing and time-cap mechanisms prior to clinical deployment.
Conclusion: AI implementation for emergency CXR interpretation is clinically effective in reducing waiting time and RTAT, improving diagnostic sensitivity, and enhancing workflow efficiency. Active worklist prioritization systems yield greatest benefit. Future prospective multicenter randomized controlled trials are necessary to establish standardized implementation protocols and evaluate long-term clinical outcomes.
References
Annarumma M, Withey SJ, Bakewell RJ, Salamati P, Rees G, Montana G. Automated Triaging of Adult Chest Radiographs with Deep Artificial Neural Networks. Radiology. 2019;291(1):196–202. https://doi.org/10.1148/radiol.2019182178
Ram S, Elias J, Shulman JG, Desai SR. Implementation of Artificial Intelligence–Assisted Chest X-ray Interpretation: It Is About Time. Annals of the American Thoracic Society. 2023;20(5):657–659. https://doi.org/10.1513/AnnalsATS.202212-1046ED
Baltruschat I, Steinmeister L, Nickisch H, Saalbach A, Grass M, Adam G, Knopp T. Smart chest X-ray worklist prioritization using artificial intelligence: a clinical workflow simulation. European Radiology. 2021;31(6):3837–3845. https://doi.org/10.1007/s00330-020-07480-7
Momin E, Chagnon M, Li M, et al. Systematic review on the impact of deep learning-driven worklist triage on radiology workflow and clinical outcomes. European Radiology. 2025. https://doi.org/10.1007/s00330-025-11395-7
Shin HJ, Kim E, Kim M. The impact of artificial intelligence on the reading times of radiologists for chest radiographs. NPJ Digital Medicine. 2023;6(1):82. https://doi.org/10.1038/s41746-023-00829-4
Ahn JS, Ebrahimian S, McDermott S, Lee SI, Naccarato L, Wu M, Digumarthy SR, Kalra MK, Dreyer KJ. Association of Artificial Intelligence–Aided Chest Radiograph Interpretation With Reader Performance and Efficiency. JAMA Network Open. 2022;5(8):e2229289. https://doi.org/10.1001/jamanetworkopen.2022.29289
Chen M, Wei C, Cai R, et al. Impact of human and artificial intelligence collaboration on workload reduction in medical image interpretation. NPJ Digital Medicine. 2024;7(1):77. https://doi.org/10.1038/s41746-024-01072-1
Seah JCY, Tang CHM, Buchlak QD, Holt XG, Wardman JB, Aimoldin A, Esmaili N, Ahmad H, Pham H, Lambert JF, Hachey B, Hogg SJ, Johnston BP, Bennett C, Oakden-Rayner L, Brotchie P, Jones CM. Effect of a comprehensive deep-learning model on the accuracy of chest x-ray interpretation by radiologists: a retrospective, multireader multicase study. The Lancet Digital Health. 2021;3(8):e496–e506. https://doi.org/10.1016/S2589-7500(21)00106-0
Selvam S, Giry M, Taleb C, et al. Efficacy of a deep learning-based software for chest X-ray analysis in an emergency department. Diagnostic and Interventional Imaging. 2025. https://doi.org/10.1016/j.diii.2025.01.007
Plesner LL, Bering T, Henriksen OM, et al. Commercially Available Chest Radiograph AI Tools for Detecting Airspace Disease, Pneumothorax, and Pleural Effusion. Radiology. 2023;308(2):e230908. https://doi.org/10.1148/radiol.230908
Batra K, Bhargava M, Sumer S, Fields BKK, Bhatt S. Radiologist Worklist Reprioritization Using Artificial Intelligence: Impact on Report Turnaround Times for CTPA Examinations Positive for Acute Pulmonary Embolism. AJR American Journal of Roentgenology. 2023;220(4):512–520. https://doi.org/10.2214/AJR.22.28210
Rudolph J, Lisson C, Becker A, et al. Nonradiology Health Care Professionals Significantly Benefit From AI Assistance in Emergency-Related Chest Radiography Interpretation. Chest. 2024;165(4):891–904. https://doi.org/10.1016/j.chest.2023.09.033
Tyler PD, Du R, Feng M, et al. Deployment of Artificial Intelligence for Radiographic Diagnosis of COVID-19 Pneumonia in the Emergency Department. Annals of Emergency Medicine. 2020;76(5):S144–S145. https://doi.org/10.1016/j.annemergmed.2020.09.285
Huang J, Neill L, Wittbrodt M, Melnick D, Klug M, Thompson M, Criner G, Avutu V, Mohebi A, Fisco M, Jordan J, Dilling T, Loftus T, Shah A. Generative Artificial Intelligence for Chest Radiograph Interpretation in the Emergency Department. JAMA Network Open. 2023;6(10):e2336100. https://doi.org/10.1001/jamanetworkopen.2023.36100
Huang J, Neill L, Radionova I, Rao V, Loftus T. Efficiency and Quality of Generative AI–Assisted Radiograph Reporting. JAMA Network Open. 2025;8(2):e2460165. https://doi.org/10.1001/jamanetworkopen.2024.60165
Seyyed-Kalantari L, Zhang H, McDermott MBA, Chen IY, Ghassemi M. Underdiagnosis bias of artificial intelligence algorithms applied to chest radiographs in under-served patient populations. Nature Medicine. 2021;27(12):2176–2182. https://doi.org/10.1038/s41591-021-01595-0
Sugibayashi T, Omura A, Yamasaki Y, et al. Deep learning for pneumothorax diagnosis: a systematic review and meta-analysis. European Respiratory Review. 2023;32(167):220213. https://doi.org/10.1183/16000617.0213-2022
Strohm L, Hehakaya C, Ranschaert ER, Boon WPC, Moors EHM. Implementation of artificial intelligence (AI) applications in radiology: hindering and facilitating factors. European Radiology. 2020;30(10):5525–5532. https://doi.org/10.1007/s00330-020-06946-y
Wenderott K, Wernhart S, Hannemann A, Hutter T, Karlen W, Tschoellitsch T, Meier J, Saller T. Effects of artificial intelligence implementation on efficiency in medical imaging—a systematic literature review and meta-analysis. NPJ Digital Medicine. 2024;7(1):3. https://doi.org/10.1038/s41746-023-00988-4
Seyam M, Weikert T, Sauter A, Brehm A, Psychogios MN, Blackham KA, et al. Utilization of Artificial Intelligence-based Intracranial Hemorrhage Detection on Emergent Noncontrast CT Images in Clinical Workflow. Radiology: Artificial Intelligence. 2022;4(3):e210168. https://doi.org/10.1148/ryai.210168
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Novita Lusiana (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Authors publishing here keep copyright but grant the journal first publication rights, under a Creative Commons Attribution-NonCommercial 4.0 License. They can distribute their work non-exclusively elsewhere with an acknowledgment of its first publication in this journal. Posting the work online before and during submission for earlier and greater citation is encouraged, reflecting Open Access benefits.
