%0期刊文章%@ 1438- 8871% I JMIR出版物%V 23卡塔尔世界杯8强波胆分析% N 11% P e26777% T描述非结构化患者报告结果的自然语言处理和机器学习方法:验证研究%A Lu，赵华%A Sim，珍阿%A Wang,Jade X %A Forrest,Christopher B %A Krull,Kevin R %A Srivastava,Deokumar %A Hudson,Melissa M %A Robison,Leslie L %A Baker,Justin N %A Huang,I-Chan %+流行病学和癌症控制科，St. Jude儿童研究医院，MS 735, 262I-Chan.Huang@STJUDE.ORG %K自然语言处理%K机器学习%K PROs %K儿科肿瘤学%D 2021 %7 3.11.2021 %9原始论文%J J医学互联网Res %G英语%X背景:在临床遭遇中通过访谈或对话评估患者报告的结果(PROs)提供了关于生存率的深刻信息。目的:本研究旨在测试自然语言处理(NLP)和机器学习(ML)算法在识别儿童和青少年癌症幸存者所经历的疼痛干扰和疲劳症状的不同属性方面的有效性，而不是作为验证NLP/ML算法的金标准的PRO内容专家的判断。方法:这项横断面研究的重点是8至17岁的儿童和青少年癌症幸存者以及护理人员，从中生成了391个疼痛干扰域的意义单位和423个疲劳域的意义单位用于分析。数据来自圣裘德儿童研究医院治疗完成后诊所。通过深度访谈报告经历过的疼痛干扰和疲劳症状。逐字转录后，可分析的句子(即意义单位)由2名内容专家对每个属性(物理、认知、社会或未分类)进行语义标记。两种NLP/ML方法用于提取和验证语义特征:来自变压器(BERT)和Word2vec的双向编码器表示，再加上一种ML方法，支持向量机或极端梯度增强。采用受试者工作特征和精密度-召回曲线来评价NLP/ML方法的准确性和有效性。 Results: Compared with Word2vec/support vector machine and Word2vec/extreme gradient boosting, BERT demonstrated higher accuracy in both symptom domains, with 0.931 (95% CI 0.905-0.957) and 0.916 (95% CI 0.887-0.941) for problems with cognitive and social attributes on pain interference, respectively, and 0.929 (95% CI 0.903-0.953) and 0.917 (95% CI 0.891-0.943) for problems with cognitive and social attributes on fatigue, respectively. In addition, BERT yielded superior areas under the receiver operating characteristic curve for cognitive attributes on pain interference and fatigue domains (0.923, 95% CI 0.879-0.997; 0.948, 95% CI 0.922-0.979) and superior areas under the precision-recall curve for cognitive attributes on pain interference and fatigue domains (0.818, 95% CI 0.735-0.917; 0.855, 95% CI 0.791-0.930). Conclusions: The BERT method performed better than the other methods. As an alternative to using standard PRO surveys, collecting unstructured PROs via interviews or conversations during clinical encounters and applying NLP/ML methods can facilitate PRO assessment in child and adolescent cancer survivors. %M 34730546 %R 10.2196/26777 %U //www.mybigtv.com/2021/11/e26777 %U https://doi.org/10.2196/26777 %U http://www.ncbi.nlm.nih.gov/pubmed/34730546