文献速递创伤与烧伤消化、代谢与肾脏急症

自然语言处理与机器学习识别急诊出院创伤患者影像偶然发现

A Natural Language Processing and Machine Learning Approach to Identification of Incidental Radiology Findings in Trauma Patients Discharged from the Emergency Department.
PubMed 文献 · Ann Emerg Med · 2022-10-31
一句话结论自然语言处理联合随机森林模型可从急诊出院创伤患者CT报告中高敏感度(92.2%)、高阴性预测值(97.2%)识别偶然发现。

本研究回顾性分析2019年某整合医疗系统内急诊出院创伤患者的CT报告,由两名独立标注者人工标注报告是否存在偶然发现作为参考标准,使用正则表达式提取特征并构建随机森林模型,按推导集(690份报告,偶然发现患病率22.3%)和验证集(282份报告,患病率22.7%)评估。模型推导集AUC为0.88(95%CI 0.84-0.92),验证集AUC为0.92(95%CI 0.88-0.96),敏感度92.2%、特异度79.4%、阴性预测值97.2%,在创伤中心、高容量和低容量社区急诊中均表现良好。作者认为自然语言处理有助于自动化审查自由文本报告以识别偶然发现,并可能促进及时随访的干预。

要点

  • 随机森林模型在验证集识别CT报告偶然发现的AUC为0.92(95%CI 0.88-0.96),敏感度92.2%、特异度79.4%、阴性预测值97.2%。
  • 推导集偶然发现患病率22.3%,验证集22.7%,提示急诊出院创伤患者CT报告中偶然发现并不少见。
  • 模型在创伤中心、高容量和低容量社区急诊分层分析中均表现良好。
  • 作者认为自然语言处理可自动化审查自由文本报告以识别偶然发现,并可能促进及时随访的干预。

适用人群与场景

  • 2019年某整合医疗系统内急诊就诊后出院的创伤患者,其CT报告纳入分析。

研究设计与关键结果

  • 研究类型:回顾性分析2019年某整合医疗系统内急诊出院创伤患者的CT报告,由两名独立标注者人工标注偶然发现作为参考标准。
  • 数据集:推导集690份报告(偶然发现患病率22.3%),验证集282份报告(患病率22.7%)。
  • 模型构建:使用正则表达式提取特征并构建随机森林模型,采用开源软件和机器学习软件。
  • 主要结果:随机森林模型推导集AUC 0.88(95%CI 0.84-0.92),验证集AUC 0.92(95%CI 0.88-0.96);最终模型敏感度92.2%、特异度79.4%、阴性预测值97.2%。
  • 分层分析:在创伤中心、高容量社区急诊和低容量社区急诊中均表现良好。

证据类型、等级与局限

  • 证据类型:回顾性分析。
  • 局限:原文未报告局限性。

对床旁实践的改变

  • 作者认为自然语言处理可自动化审查自由文本报告以识别偶然发现,并可能促进及时随访的干预。

提炼说明

  • 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开

STUDY OBJECTIVE: Patients undergoing diagnostic imaging studies in the emergency department (ED) commonly have incidental findings, which may represent unrecognized serious medical conditions, including cancer. Recognition of incidental findings frequently relies on manual review of textual radiology reports and can be overlooked in a busy clinical environment. Our study aimed to develop and validate a supervised machine learning model using natural language processing to automate the recognition of incidental findings in radiology reports of patients discharged from the ED.

METHODS: We performed a retrospective analysis of computed tomography (CT) reports from trauma patients discharged home across an integrated health system in 2019. Two independent annotators manually labeled CT reports for the presence of an incidental finding as a reference standard. We used regular expressions to derive and validate a random forest model using open-source and machine learning software. Final model performance was assessed across different ED types.

RESULTS: The study CT reports were divided into derivation (690 reports) and validation (282 reports) sets, with a prevalence of incidental findings of 22.3%, and 22.7%, respectively. The random forest model had an area under the curve of 0.88 (95% confidence interval [CI], 0.84 to 0.92) on the derivation set and 0.92 (95% CI, 0.88 to 0.96) on the validation set. The final model was found to have a sensitivity of 92.2%, a specificity of 79.4%, and a negative predictive value of 97.2%. Similarly, strong model performance was found when stratified to a dedicated trauma center, high-volume, and low-volume community EDs.

CONCLUSION: Machine learning and natural language processing can classify incidental findings in CT reports of ED patients with high sensitivity and high negative predictive value across a broad range of ED settings. These findings suggest the utility of natural language processing in automating the review of free-text reports to identify incidental findings and may facilitate interventions to improve timely follow-up.

来源:Ann Emerg Med,查看原文。版权归原作者所有。

参考来源

  1. Evans CS, Dorris HD, Kane MT, et al. A Natural Language Processing and Machine Learning Approach to Identification of Incidental Radiology Findings in Trauma Patients Discharged from the Emergency Department. Ann Emerg Med. 2022. PMID: 36328850. doi:10.1016/j.annemergmed.2022.08.450

供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。