A novel method of adverse event detection can accurately identify venous thromboembolisms (VTEs) from narrative electronic health record data
A novel method of adverse event detection can accurately identify venous thromboembolisms (VTEs) from narrative electronic health record data
复制标题
DOI：
10.1136/amiajnl-2014-002768
复制
发表时间：
2015-01-01
影响因子：
6.4
通讯作者：
Buckeridge, David L.
中科院分区：
文献类型：
--
作者：
Rochefort, Christian M.;Verma, Aman D.;Buckeridge, David L.
背景静脉血栓栓塞（vte），包括深静脉血栓形成（DVT）和肺栓塞（PE），与住院患者的死亡率、发病率和成本相关。为了评估预防措施的成功与否，需要准确有效的静脉血栓栓塞率监测方法。因此，我们试图确定统计自然语言处理（NLP）从电子健康记录数据中识别DVT和PE的准确性。方法：我们随机抽取2008年至2012年加拿大蒙特利尔疑似DVT/PE患者的2000份叙事放射学报告。我们在每个报告中手动识别DVT/PE，作为我们的参考标准。使用词袋方法，我们训练了10个预测DVT和10个预测PE的替代支持向量机（SVM）模型。采用嵌套的10倍交叉验证对SVM进行训练和测试，并对各模型的平均准确率进行测量和比较。结果经人工复查，324例（16.2%）dvt阳性，154例（7.7%）pe阳性。最佳DVT模型的平均灵敏度为0.80 (95% CI 0.76 ~ 0.85)，特异性为0.98 (98% CI 0.97 ~ 0.99)，阳性预测值（PPV）为0.89 (95% CI 0.85 ~ 0.93)，曲线下面积（AUC）为0.98 （95% CI 0.97 ~ 0.99）。最佳PE模型的灵敏度为0.79 (95% CI 0.73 ~ 0.85)，特异性为0.99 (95% CI 0.98 ~ 0.99)， PPV为0.84 (95% CI 0.75 ~ 0.92)， AUC为0.99 （95% CI 0.98 ~ 1.00）。结论统计学NLP可以准确地从叙事放射学报告中识别静脉血栓栓塞。
Background Venous thromboembolisms (VTEs), which include deep vein thrombosis (DVT) and pulmonary embolism (PE), are associated with significant mortality, morbidity, and cost in hospitalized patients. To evaluate the success of preventive measures, accurate and efficient methods for monitoring VTE rates are needed. Therefore, we sought to determine the accuracy of statistical natural language processing (NLP) for identifying DVT and PE from electronic health record data.Methods We randomly sampled 2000 narrative radiology reports from patients with a suspected DVT/PE in Montreal (Canada) between 2008 and 2012. We manually identified DVT/PE within each report, which served as our reference standard. Using a bag-of-words approach, we trained 10 alternative support vector machine (SVM) models predicting DVT, and 10 predicting PE. SVM training and testing was performed with nested 10-fold cross-validation, and the average accuracy of each model was measured and compared.Results On manual review, 324 (16.2%) reports were DVT-positive and 154 (7.7%) were PE-positive. The best DVT model achieved an average sensitivity of 0.80 (95% CI 0.76 to 0.85), specificity of 0.98 (98% CI 0.97 to 0.99), positive predictive value (PPV) of 0.89 (95% CI 0.85 to 0.93), and an area under the curve (AUC) of 0.98 (95% CI 0.97 to 0.99). The best PE model achieved sensitivity of 0.79 (95% CI 0.73 to 0.85), specificity of 0.99 (95% CI 0.98 to 0.99), PPV of 0.84 (95% CI 0.75 to 0.92), and AUC of 0.99 (95% CI 0.98 to 1.00).Conclusions Statistical NLP can accurately identify VTE from narrative radiology reports.