The Accuracy of Artificial Intelligence Models in Hand/Wrist Fracture and Dislocation Diagnosis: A Systematic Review and Meta-Analysis.

IF 1.7 Q2 SURGERY

JBJS Reviews Pub Date : 2024-09-05 eCollection Date: 2024-09-01 DOI:10.2106/JBJS.RVW.24.00106

Chloe R Wong, Alice Zhu, Heather L Baltzer

{"title":"The Accuracy of Artificial Intelligence Models in Hand/Wrist Fracture and Dislocation Diagnosis: A Systematic Review and Meta-Analysis.","authors":"Chloe R Wong, Alice Zhu, Heather L Baltzer","doi":"10.2106/JBJS.RVW.24.00106","DOIUrl":null,"url":null,"abstract":"Background: Early and accurate diagnosis is critical to preserve function and reduce healthcare costs in patients with hand and wrist injury. As such, artificial intelligence (AI) models have been developed for the purpose of diagnosing fractures through imaging. The purpose of this systematic review and meta-analysis was to determine the accuracy of AI models in identifying hand and wrist fractures and dislocations.Methods: Adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Diagnostic Test Accuracy guidelines, Ovid MEDLINE, Embase, and Cochrane Central Register of Controlled Trials were searched from their inception to October 10, 2023. Studies were included if they utilized an AI model (index test) for detecting hand and wrist fractures and dislocations in pediatric (<18 years) or adult (>18 years) patients through any radiologic imaging, with the reference standard established through image review by a medical expert. Results were synthesized through bivariate analysis. Risk of bias was assessed using the QUADAS-2 tool. This study was registered with PROSPERO (CRD42023486475). Certainty of evidence was assessed using Grading of Recommendations Assessment, Development, and Evaluation.Results: A systematic review identified 36 studies. Most studies assessed wrist fractures (27.90%) through radiograph imaging (94.44%), with radiologists serving as the reference standard (66.67%). AI models demonstrated area under the curve (0.946), positive likelihood ratio (7.690; 95% confidence interval, 6.400-9.190), and negative likelihood ratio (0.112; 0.0848-0.145) in diagnosing hand and wrist fractures and dislocations. Examining only studies characterized by a low risk of bias, sensitivity analysis did not reveal any difference from the overall results. Overall certainty of evidence was moderate.Conclusion: In demonstrating the accuracy of AI models in hand and wrist fracture and dislocation diagnosis, we have demonstrated that the potential use of AI in diagnosing hand and wrist fractures is promising.Level of evidence: Level III. See Instructions for Authors for a complete description of levels of evidence.","PeriodicalId":47098,"journal":{"name":"JBJS Reviews","volume":"12 9","pages":""},"PeriodicalIF":1.7000,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"JBJS Reviews","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.2106/JBJS.RVW.24.00106","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/9/1 0:00:00","PubModel":"eCollection","JCR":"Q2","JCRName":"SURGERY","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Early and accurate diagnosis is critical to preserve function and reduce healthcare costs in patients with hand and wrist injury. As such, artificial intelligence (AI) models have been developed for the purpose of diagnosing fractures through imaging. The purpose of this systematic review and meta-analysis was to determine the accuracy of AI models in identifying hand and wrist fractures and dislocations.

Methods: Adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Diagnostic Test Accuracy guidelines, Ovid MEDLINE, Embase, and Cochrane Central Register of Controlled Trials were searched from their inception to October 10, 2023. Studies were included if they utilized an AI model (index test) for detecting hand and wrist fractures and dislocations in pediatric (<18 years) or adult (>18 years) patients through any radiologic imaging, with the reference standard established through image review by a medical expert. Results were synthesized through bivariate analysis. Risk of bias was assessed using the QUADAS-2 tool. This study was registered with PROSPERO (CRD42023486475). Certainty of evidence was assessed using Grading of Recommendations Assessment, Development, and Evaluation.

Results: A systematic review identified 36 studies. Most studies assessed wrist fractures (27.90%) through radiograph imaging (94.44%), with radiologists serving as the reference standard (66.67%). AI models demonstrated area under the curve (0.946), positive likelihood ratio (7.690; 95% confidence interval, 6.400-9.190), and negative likelihood ratio (0.112; 0.0848-0.145) in diagnosing hand and wrist fractures and dislocations. Examining only studies characterized by a low risk of bias, sensitivity analysis did not reveal any difference from the overall results. Overall certainty of evidence was moderate.

Conclusion: In demonstrating the accuracy of AI models in hand and wrist fracture and dislocation diagnosis, we have demonstrated that the potential use of AI in diagnosing hand and wrist fractures is promising.

Level of evidence: Level III. See Instructions for Authors for a complete description of levels of evidence.

查看原文本刊更多论文

人工智能模型在手部/手腕骨折和脱位诊断中的准确性：系统回顾与元分析》。

背景：早期准确的诊断对于保护手部和腕部损伤患者的功能和降低医疗成本至关重要。因此，人们开发了人工智能（AI）模型，用于通过成像诊断骨折。本系统综述和荟萃分析旨在确定人工智能模型在识别手部和腕部骨折及脱位方面的准确性：方法：根据《系统综述和荟萃分析诊断测试准确性首选报告项目》指南，检索了 Ovid MEDLINE、Embase 和 Cochrane Central Register of Controlled Trials 从开始到 2023 年 10 月 10 日的所有研究。如果研究采用了人工智能模型（指数测试），通过任何放射成像检测小儿（18 岁）患者的手部和腕部骨折及脱位，并通过医学专家的图像审查确定参考标准，则纳入该研究。通过双变量分析对结果进行综合。使用 QUADAS-2 工具评估偏倚风险。本研究已在 PROSPERO 注册（CRD42023486475）。证据的确定性采用建议评估、制定和评价分级法进行评估：系统综述确定了 36 项研究。大多数研究（27.90%）通过放射影像学成像（94.44%）对腕部骨折进行评估，放射科医生作为参考标准（66.67%）。在诊断手部和腕部骨折和脱位方面，人工智能模型显示了曲线下面积（0.946）、正似然比（7.690；95% 置信区间，6.400-9.190）和负似然比（0.112；0.0848-0.145）。敏感性分析仅考察了偏倚风险较低的研究，并未发现与总体结果有任何差异。总体证据的确定性为中等：通过证明人工智能模型在诊断手部和腕部骨折及脱位方面的准确性，我们证明了人工智能在诊断手部和腕部骨折方面的潜在应用前景广阔：证据等级：三级。有关证据等级的完整描述，请参阅 "作者须知"。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

JBJS Reviews SURGERY-

CiteScore

4.40

自引率

4.30%

发文量

132

期刊介绍： JBJS Reviews is an innovative review journal from the publishers of The Journal of Bone & Joint Surgery. This continuously published online journal provides comprehensive, objective, and authoritative review articles written by recognized experts in the field. Edited by Thomas A. Einhorn, MD, and a distinguished Editorial Board, each issue of JBJS Reviews, updates the orthopaedic community on important topics in a concise, time-saving manner, providing expert insights into orthopaedic research and clinical experience. Comprehensive reviews, special features, and integrated CME provide orthopaedic surgeons with valuable perspectives on surgical practice and the latest advances in the field within twelve subspecialty areas: Basic Science, Education & Training, Elbow, Ethics, Foot & Ankle, Hand & Wrist, Hip, Infection, Knee, Oncology, Pediatrics, Pain Management, Rehabilitation, Shoulder, Spine, Sports Medicine, Trauma.