医疗器械不良事件报告自动分析软件的设计  

Design of Automatic Analysis Software for Medical Device Adverse Event Reports

作　　者：凌庆庆 王浩文 夏景涛[1] 张博涵 李子好 孙遥[1] 陈宏文[1] LING Qingqing;WANG Haowen;XIA Jingtao;ZHANG Bohan;LI Zihao;SUN Yao;CHEN Hongwen(Department of Medical Engineering,Nanfang Hospital Southern Medical University,Guangzhou Guangdong 510515,China;Guangdong ADR Monitoring Center,Guangzhou Guangdong 510515,China)

机构地区：[1]南方医科大学南方医院医学工程科,广东广州510515 [2]广东省药品不良反应监测中心,广东广州510515

出　　处：《中国医疗设备》2024年第2期90-94,共5页China Medical Devices

基　　金：广东省重点领域研发计划(2019B111103001);南方医科大学南方医院院长基金项目(2020C031)。

摘　　要：目的为实现对不良事件报告的自动筛查和预警,提升不良事件分析的效率和准确率,开发一款医疗器械不良事件报告自动分析软件。方法软件基于Visual Studio 2013和QT 5.9.5开发,包括报告导入、报告显示、报告分析、分析结果、关键词库管理5个功能模块。利用以往的不良事件报告创建关键词库并持续更新,根据需求制定多项分析规则,选定规则后对导入的待查报告进行自动分析,导出分析结果,实现报告的筛查和风险预警。结果从5399份医用监护仪不良事件数据中提取并创建关键词库,以695份监护仪不良事件报告进行软件测试,自动分析的准确率达到85%左右,分析时间缩短至5 s以内,与手动分析相比差异有统计学意义(P<0.05)。结论医疗器械不良事件报告自动分析软件可同时筛选分析大量报告并导出分析结果,实现对不良事件报告的自动筛选及预警,极大地减轻了监测机构分析人员的负担。Objective In order to realize the automatic screening and early warning of adverse event reports and improve the efficiency and accuracy of adverse event analysis,to develop an automatic analysis software of medical device adverse event reports.Methods The software was developed based on Visual Studio 2013 and QT 5.9.5,including five functional modules:reports import,reports display,reports analysis,analysis results and keyword library management.The keyword library was created and continuously updated by using the previous adverse event report.A number of analysis rules were formulated according to the requirements,the imported report to be checked was automatically analyzed after the rules were selected,and the analysis results were exported to realize the screening and risk early warning of the report.Results A keyword library was extracted and created from 5399 medical monitor adverse event data,and the software was tested with 695 monitor adverse event reports.The results showed that the accuracy of automatic analysis was about 85%,and the analysis time was shortened to within 5 s,and the difference was statistically significant compared with manual analysis(P<0.05).Conclusion The software can screen and analyze a large number of reports and export the analysis results at the same time,realize the automatic screening and early warning of adverse event reports,and greatly reduce the burden of analysts in monitoring institutions.

关 键 词：医疗器械 不良事件报告 自动分析 关键词库 筛选预警 

分 类 号：R197.39[医药卫生—卫生事业管理]