Department of Neurosurgery, Warren Alpert Medical School, Rhode Island Hospital, Brown University, Providence, RI, United States of America.
Department of Computer Science, Brown University, Providence, RI, United States of America.
PLoS One. 2024 May 9;19(5):e0303519. doi: 10.1371/journal.pone.0303519. eCollection 2024.
To establish whether or not a natural language processing technique could identify two common inpatient neurosurgical comorbidities using only text reports of inpatient head imaging.
A training and testing dataset of reports of 979 CT or MRI scans of the brain for patients admitted to the neurosurgery service of a single hospital in June 2021 or to the Emergency Department between July 1-8, 2021, was identified. A variety of machine learning and deep learning algorithms utilizing natural language processing were trained on the training set (84% of the total cohort) and tested on the remaining images. A subset comparison cohort (n = 76) was then assessed to compare output of the best algorithm against real-life inpatient documentation.
For "brain compression", a random forest classifier outperformed other candidate algorithms with an accuracy of 0.81 and area under the curve of 0.90 in the testing dataset. For "brain edema", a random forest classifier again outperformed other candidate algorithms with an accuracy of 0.92 and AUC of 0.94 in the testing dataset. In the provider comparison dataset, for "brain compression," the random forest algorithm demonstrated better accuracy (0.76 vs 0.70) and sensitivity (0.73 vs 0.43) than provider documentation. For "brain edema," the algorithm again demonstrated better accuracy (0.92 vs 0.84) and AUC (0.45 vs 0.09) than provider documentation.
A natural language processing-based machine learning algorithm can reliably and reproducibly identify selected common neurosurgical comorbidities from radiology reports.
This result may justify the use of machine learning-based decision support to augment provider documentation.
确定自然语言处理技术是否仅通过住院头部成像的文本报告即可识别两种常见的住院神经外科合并症。
确定了 2021 年 6 月入住单一医院神经外科或 2021 年 7 月 1 日至 8 日入住急诊部的患者的 979 例 CT 或 MRI 脑部扫描报告的训练和测试数据集。利用自然语言处理的各种机器学习和深度学习算法在训练集（总队列的 84%）上进行了训练，并在其余图像上进行了测试。然后评估了一个子集比较队列（n=76），以比较最佳算法的输出与实际住院文档的对比。
对于“脑压迫”，随机森林分类器在测试数据集中的准确率为 0.81，曲线下面积为 0.90，优于其他候选算法。对于“脑水肿”，随机森林分类器再次在测试数据集中优于其他候选算法，准确率为 0.92，曲线下面积为 0.94。在提供者比较数据集中，对于“脑压迫”，随机森林算法的准确率（0.76 对 0.70）和敏感度（0.73 对 0.43）均优于提供者文档。对于“脑水肿”，算法再次显示出优于提供者文档的准确率（0.92 对 0.84）和曲线下面积（0.45 对 0.09）。
基于自然语言处理的机器学习算法可以可靠且可重复地从放射学报告中识别出选定的常见神经外科合并症。
这一结果可能证明使用基于机器学习的决策支持来增强提供者文档是合理的。
