Use of Natural Language Processing Algorithms to Identify Common Data Elements in Operative Notes for Knee Arthroplasty.
Use of Natural Language Processing Algorithms to Identify Common Data Elements in Operative Notes for Knee Arthroplasty.
复制标题
DOI：
10.1016/j.arth.2020.09.029
复制
发表时间：
2021-03
期刊：
影响因子：
--
通讯作者：
Maradit Kremers H
中科院分区：
文献类型：
--
作者：
Sagheb E;Ramazanian T;Tafti AP;Fu S;Kremers WK;Berry DJ;Lewallen DG;Sohn S;Maradit Kremers H
自然语言处理（NLP）方法能够处理电子健康记录中的临床自由文本，减少了昂贵的手动图表审查的需求，并提高了数据质量。我们开发了基于规则的NLP算法，以自动从膝关节置换术手术记录中提取手术特定的数据元素。在2000年至2017年一家大型高等教育机构的20，000份膝关节置换术手术记录中，我们随机选择了独立的训练集和测试集，以开发和评估NLP算法来检测五个主要数据元素。训练和测试数据集的大小相似，范围在420到1592次手术之间。使用手术记录中的关键词的专家规则来实现NLP算法，捕获：（1）手术类别（全膝关节置换术、单间室膝关节置换术、髌股关节置换术），（2）手术偏侧性，（3）限制类型，（4）髌骨表面置换的存在，以及（5）植入物型号（目录号）。我们使用机构注册数据作为我们的黄金标准来评估NLP算法。检测手术类别、偏侧性、限制和髌骨表面置换的NLP算法在测试数据集上的准确率分别为98.3%、99.5%、99.2%和99.4%。植入模型算法实现了99.9%的F1分数（精确度和召回率的调和平均值）。NLP算法是一种有前途的替代昂贵的手动图表审查，以自动提取膝关节置换术手术记录中的嵌入信息。在其他医院环境中的进一步验证将提高用于研究和临床目的的数据采集的广泛实施和效率。三级
Natural language processing (NLP) methods have the capability to process clinical free text in electronic health records, decreasing the need for costly manual chart review and improving data quality. We developed rule-base NLP algorithms to automatically extract surgery specific data elements from knee arthroplasty operative notes. Within a cohort of 20,000 knee arthroplasty operative notes from 2000 to 2017 at a large tertiary institution, we randomly selected independent pairs of training and test sets to develop and evaluate NLP algorithms to detect five major data elements. The size of the training and test datasets were similar and ranged between 420 to 1592 surgeries. Expert rules using keywords in operative notes were used to implement NLP algorithms capturing: (1) category of surgery (total knee arthroplasty, unicompartmental knee arthroplasty, patellofemoral arthroplasty), (2) laterality of surgery, (3) constraint type, (4) presence of patellar resurfacing, and (5) implant model (catalog numbers). We used institutional registry data as our gold standard to evaluate the NLP algorithms. NLP algorithms to detect category of surgery, laterality, constraint and patellar resurfacing achieved 98.3%, 99.5%, 99.2% and 99.4% accuracy on test datasets, respectively. The implant model algorithm achieved an F1-score (harmonic mean of precision and recall) of 99.9%. NLP algorithms are a promising alternative to costly manual chart review to automate extraction of embedded information within knee arthroplasty operative notes. Further validation in other hospital settings will enhance widespread implementation and efficiency in data capture for research and clinical purposes. Level III
登录
查看更多内容
影响因子：
5.3
作者：
Kurtz, Steven M.;Ong, Kevin L.;Bozic, Kevin J.
通讯作者：
Bozic, Kevin J.
影响因子：
5.3
作者：
Bozic, Kevin J.;Grosso, Laura M.;Drye, Elizabeth E.
通讯作者：
Drye, Elizabeth E.
影响因子：
120.7
作者：
Murff, Harvey J.;FitzHenry, Fern;Speroff, Theodore
通讯作者：
Speroff, Theodore
影响因子：
3.5
作者：
Tibbo, Meagan E.;Wyles, Cody C.;Kremers, Hilal Maradit
通讯作者：
Kremers, Hilal Maradit
影响因子：
3.5
作者：
Ramkumar, Prem N.;Muschler, George F.;Mont, Michael A.
通讯作者：
Mont, Michael A.