文献速递胸部技术与安全DR

深度学习用于急诊科胸部X线诊断

Deep Learning for Chest Radiograph Diagnosis in the Emergency Department.
PubMed 文献 · Radiology · 2019-10-22

该单中心回顾性研究纳入2017年1月1日至3月31日急诊科连续就诊并首次行胸部X线检查的1135例患者,评估商用深度学习算法识别临床相关异常的表现。算法识别异常胸片的AUC为0.95,在高敏感度阈值下敏感度88.7%、特异度69.6%,在高特异度阈值下敏感度81.6%、特异度90.3%;值班放射科住院医师敏感度65.6%、特异度98.1%,使用算法输出后住院医师敏感度提高至73.4%、特异度降至94.3%。

要点

  • 深度学习算法识别急诊异常胸片的AUC为0.95(95%CI:0.93,0.96)。
  • 高敏感度阈值下算法敏感度88.7%、特异度69.6%;高特异度阈值下敏感度81.6%、特异度90.3%。
  • 住院医师使用算法输出后敏感度由65.6%升至73.4%,特异度由98.1%降至94.3%。

核心结论

  • 深度学习算法用于急诊科胸片时对识别临床相关异常显示出诊断性能,并帮助提高了放射科住院医师评估的敏感度。

研究设计与样本

  • 单中心回顾性研究,纳入2017年1月1日至3月31日急诊科连续就诊并首次行胸部X线检查的患者1135例(平均年龄53岁±18;582名男性)。
  • 胸片使用商用深度学习算法分析,以预设的高敏感度和高特异度操作阈值评估算法性能。
  • 算法敏感度和特异度与在实际工作中解读胸片的值班放射科住院医师比较,采用McNemar检验。

诊断标准与关键数值

  • 识别异常胸片时算法AUC为0.95(95%CI:0.93,0.96)。
  • 高敏感度阈值下算法敏感度88.7%(227/256张胸片;95%CI:84.1%,92.3%),特异度69.6%(612/879张胸片;95%CI:66.5%,72.7%)。
  • 高特异度阈值下算法敏感度81.6%(209/256张胸片;95%CI:76.3%,86.2%),特异度90.3%(794/879张胸片;95%CI:88.2%,92.2%)。
  • 放射科住院医师敏感度65.6%(168/256张胸片;95%CI:59.5%,71.4%,P < .001)低于算法,特异度98.1%(862/879张胸片;95%CI:96.9%,98.9%,P < .001)高于算法。
  • 使用算法输出重新解读胸片后,住院医师敏感度提高至73.4%(188/256张胸片;95%CI:68.0%,78.8%,P = .003),特异度降至94.3%(829/879张胸片;95%CI:92.8%,95.8%,P < .001)。

临床应用与阅片要点

  • 算法与住院医师对同一批胸片存在不一致发现时,住院医师使用算法输出重新解读了胸片。
  • 使用算法输出后住院医师敏感度提高而特异度降低,提示算法辅助可影响阅片者的敏感度与特异度权衡。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

BackgroundThe performance of a deep learning (DL) algorithm should be validated in actual clinical situations, before its clinical implementation.PurposeTo evaluate the performance of a DL algorithm for identifying chest radiographs with clinically relevant abnormalities in the emergency department (ED) setting.Materials and MethodsThis single-center retrospective study included consecutive patients who visited the ED and underwent initial chest radiography between January 1 and March 31, 2017. Chest radiographs were analyzed with a commercially available DL algorithm. The performance of the algorithm was evaluated by determining the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity at predefined operating cutoffs (high-sensitivity and high-specificity cutoffs). The sensitivities and specificities of the algorithm were compared with those of the on-call radiology residents who interpreted the chest radiographs in the actual practice by using McNemar tests. If there were discordant findings between the algorithm and resident, the residents reinterpreted the chest radiographs by using the algorithm's output.ResultsA total of 1135 patients (mean age, 53 years ± 18; 582 men) were evaluated. In the identification of abnormal chest radiographs, the algorithm showed an AUC of 0.95 (95% confidence interval [CI]: 0.93, 0.96), a sensitivity of 88.7% (227 of 256 radiographs; 95% CI: 84.1%, 92.3%), and a specificity of 69.6% (612 of 879 radiographs; 95% CI: 66.5%, 72.7%) at the high-sensitivity cutoff and a sensitivity of 81.6% (209 of 256 radiographs; 95% CI: 76.3%, 86.2%) and specificity of 90.3% (794 of 879 radiographs; 95% CI: 88.2%, 92.2%) at the high-specificity cutoff. Radiology residents showed lower sensitivity (65.6% [168 of 256 radiographs; 95% CI: 59.5%, 71.4%], P < .001) and higher specificity (98.1% [862 of 879 radiographs; 95% CI: 96.9%, 98.9%], P < .001) compared with the algorithm. After reinterpretation of chest radiographs with use of the algorithm's outputs, the sensitivity of the residents improved (73.4% [188 of 256 radiographs; 95% CI: 68.0%, 78.8%], P = .003), whereas specificity was reduced (94.3% [829 of 879 radiographs; 95% CI: 92.8%, 95.8%], P < .001).ConclusionA deep learning algorithm used with emergency department chest radiographs showed diagnostic performance for identifying clinically relevant abnormalities and helped improve the sensitivity of radiology residents' evaluation.Published under a CC BY 4.0 license.Online supplemental material is available for this article.See also the editorial by Munera and Infante in this issue.

来源:Radiology,查看原文。版权归原作者所有。

引用

  1. Hwang EJ, Nam JG, Lim WH, et al. Deep Learning for Chest Radiograph Diagnosis in the Emergency Department. Radiology. 2019. PMID: 31638490. doi:10.1148/radiol.2019191225

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。