{"title":"Artificial Intelligence and Decision Support in Orthopaedic Emergencies: A Review.","authors":"Raju Vaishya, Ayaz Saiyed, Abhishek Vaish, Rajesh Botchu, Kartikeyan P Iyengar, Albina Arjuman","doi":"10.1007/s43465-026-01760-y","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Orthopaedic emergencies require rapid, accurate clinical decisions to prevent limb- and life-threatening complications. Increasing patient volume, limited specialist availability, and diagnostic complexity contributes to delays and errors in emergency care. Artificial intelligence (AI) encompasses machine learning (ML) and deep learning (DL) and has emerged as a promising tool to augment decision-making in orthopaedic emergencies.</p><p><strong>Purpose: </strong>To review current applications, evidence, limitations, and future directions of AI-based decision-support systems in orthopaedic emergency care, with particular relevance to low- and middle-income healthcare settings.</p><p><strong>Methods: </strong>A comprehensive electronic search was performed in major academic databases. This included PubMed, Embase, Scopus, and Web of Science to identify all relevant literature published in English from inception up to February 2026, focusing on AI applications in orthopaedic trauma, emergency imaging, triage, clinical decision support, and outcome prediction. Emphasis was placed on clinically validated studies, systematic reviews, and emerging technologies relevant to emergency orthopaedics.</p><p><strong>Results: </strong>AI applications in orthopaedic emergencies primarily included automated fracture detection, injury classification, triage optimisation, clinical decision support, and complication prediction. DL models have demonstrated diagnostic accuracy comparable to or exceeding that of expert clinicians in fracture detection. AI-assisted triage systems reportedly outperformed conventional scoring tools in predicting severity and admission needed based on the studies included for this review. Clinical decision-support tools improved the completeness and consistency of emergency management recommendations based on hospital-based and practice-oriented studies. The reported challenges included those related to data bias, interpretability, clinical integration, ethical concerns, and regulatory oversight.</p><p><strong>Conclusions: </strong>AI has substantial potential to enhance diagnostic accuracy, efficiency, and decision-making in orthopaedic emergencies. Responsible adoption requires robust external validation, clinician education, explainable models, and integration into existing workflows. When appropriately implemented, AI can serve as a valuable adjunct to, rather than a replacement for, clinical expertise in orthopaedic emergency care.</p><p><strong>Graphical abstract: </strong></p>","PeriodicalId":13338,"journal":{"name":"Indian Journal of Orthopaedics","volume":"60 9","pages":"2494-2504"},"PeriodicalIF":1.3000,"publicationDate":"2026-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13627520/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Indian Journal of Orthopaedics","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s43465-026-01760-y","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2026/9/1 0:00:00","PubModel":"eCollection","JCR":"Q3","JCRName":"ORTHOPEDICS","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Orthopaedic emergencies require rapid, accurate clinical decisions to prevent limb- and life-threatening complications. Increasing patient volume, limited specialist availability, and diagnostic complexity contributes to delays and errors in emergency care. Artificial intelligence (AI) encompasses machine learning (ML) and deep learning (DL) and has emerged as a promising tool to augment decision-making in orthopaedic emergencies.
Purpose: To review current applications, evidence, limitations, and future directions of AI-based decision-support systems in orthopaedic emergency care, with particular relevance to low- and middle-income healthcare settings.
Methods: A comprehensive electronic search was performed in major academic databases. This included PubMed, Embase, Scopus, and Web of Science to identify all relevant literature published in English from inception up to February 2026, focusing on AI applications in orthopaedic trauma, emergency imaging, triage, clinical decision support, and outcome prediction. Emphasis was placed on clinically validated studies, systematic reviews, and emerging technologies relevant to emergency orthopaedics.
Results: AI applications in orthopaedic emergencies primarily included automated fracture detection, injury classification, triage optimisation, clinical decision support, and complication prediction. DL models have demonstrated diagnostic accuracy comparable to or exceeding that of expert clinicians in fracture detection. AI-assisted triage systems reportedly outperformed conventional scoring tools in predicting severity and admission needed based on the studies included for this review. Clinical decision-support tools improved the completeness and consistency of emergency management recommendations based on hospital-based and practice-oriented studies. The reported challenges included those related to data bias, interpretability, clinical integration, ethical concerns, and regulatory oversight.
Conclusions: AI has substantial potential to enhance diagnostic accuracy, efficiency, and decision-making in orthopaedic emergencies. Responsible adoption requires robust external validation, clinician education, explainable models, and integration into existing workflows. When appropriately implemented, AI can serve as a valuable adjunct to, rather than a replacement for, clinical expertise in orthopaedic emergency care.
背景：骨科急诊需要快速、准确的临床决策，以防止危及肢体和生命的并发症。患者数量增加、专家可用性有限以及诊断复杂性导致急诊护理的延误和错误。人工智能（AI）包括机器学习（ML）和深度学习（DL），并已成为增强骨科紧急情况决策的有前途的工具。目的：回顾目前基于人工智能的决策支持系统在骨科急诊护理中的应用、证据、局限性和未来方向，特别是与中低收入医疗保健机构相关。方法：对主要学术数据库进行全面的电子检索。这包括PubMed、Embase、Scopus和Web of Science，以确定从成立到2026年2月发表的所有相关英文文献，重点关注人工智能在骨科创伤、急诊成像、分流、临床决策支持和结果预测方面的应用。重点放在临床验证的研究、系统综述和与急诊骨科相关的新兴技术上。结果：人工智能在骨科急诊中的应用主要包括骨折自动检测、损伤分类、分诊优化、临床决策支持和并发症预测。DL模型已经证明了与骨折检测专家临床医生相当或超过的诊断准确性。据报道，基于本综述纳入的研究，人工智能辅助分诊系统在预测严重程度和需要入院方面优于传统评分工具。临床决策支持工具提高了基于医院和面向实践的研究的应急管理建议的完整性和一致性。报告的挑战包括与数据偏差、可解释性、临床整合、伦理问题和监管监督相关的挑战。结论：人工智能在提高骨科急诊诊断的准确性、效率和决策方面具有巨大的潜力。负责任的采用需要健壮的外部验证、临床医生教育、可解释的模型以及集成到现有工作流中。如果实施得当，人工智能可以作为骨科急诊临床专业知识的宝贵辅助，而不是替代。图形化的简介:
期刊介绍：
IJO welcomes articles that contribute to Orthopaedic knowledge from India and overseas. We publish articles dealing with clinical orthopaedics and basic research in orthopaedic surgery. Articles are accepted only for exclusive publication in the Indian Journal of Orthopaedics. Previously published articles, articles which are in peer-reviewed electronic publications in other journals, are not accepted by the Journal. Published articles and illustrations become the property of the Journal. The copyright remains with the journal. Studies must be carried out in accordance with World Medical Association Declaration of Helsinki.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
