论文部分内容阅读
目的:提取原发性肝癌CT诊断报告的术语,发现共性术语,探索特定术语,为报告文本数据结构化提供参考。方法:研究选取了实际临床工作中提交的在线CT报告,以标准案例库中的CT报告作为对照,比较两组152个标准术语的应用情况,以术语出现的频次作为主要研究指标。其中,对照组和在线组的报告例数各为326份,共652份。统计分析方法为方差分析、t检验或非参数检验方法。结果:152个术语中,有86个术语两组出现频次差异无统计学意义(p>0.05),说明这86个术语也是在线组的常用报告术语。326份CT报告按不同的治疗方式分为三组,162个术语中,有88个术语的平均出现频次受分组影响显著,说明这88个术语可能是某一组的特有术语。还有64个术语可能是三组报告的共性术语。结论:基于统计学方法可以帮助识别CT报告的共性术语和专用术语,便于报告文本的结构化处理。
OBJECTIVE: To extract the terminology of CT diagnosis report of primary liver cancer, to find common terminology, to explore specific terminology, and to provide a reference for the structuring of report text data. Methods: The study selected the online CT reports submitted in the actual clinical work and the CT reports in the standard case library as a control. The application of 152 standard terms in two groups was compared. The frequency of occurrence was taken as the main research index. Among them, the control group and the online group each reported 326 cases, a total of 652 copies. Statistical analysis methods for analysis of variance, t test or non-parametric test methods. Results: Of the 152 terms, there were 86 terms with no significant difference in frequency between the two groups (p> 0.05), indicating that these 86 terms are also common reporting terms of the online group. 326 CT reports were divided into three groups according to different treatment methods. Of the 162 terms, the average occurrence frequency of 88 terms was significantly affected by the grouping, indicating that the 88 terms may be specific terms of a group. A further 64 terms may be common terms for the three groups of reports. Conclusion: Based on statistical methods, common terminology and terminology of CT reports can be identified to facilitate the structured processing of report texts.