基于Apriori算法分析2021年山东省医疗器械不良事件的关联性

doi:10.6040/j.issn.1671-7554.0.2022.0574

摘要/Abstract

摘要： 目的医疗器械不良事件监测是医疗器械上市后风险管理的重要手段。本研究旨在基于Apriori算法分析2021年山东省医疗器械不良事件的关联性。方法对2021年山东省各监测机构上报的63 041起不良伤害事件,按广义医疗器械分类划分为三类(无源医疗器械、有源医疗器械以及体外诊断试剂)医疗器械不良事件进行描述分析。采用关联规则挖掘中的Apriori模型,挖掘出与不良事件相关的器械类别、使用科室、医院类别、是否超期使用以及上报单位所属地区,探索医疗器械不良事件关联风险。结果不良事件中包含有源医疗器械20 564起、无源医疗器械42 181起及体外诊断试剂296起。其中,无源医疗器械不良事件发生最多的地级市为烟台市(5 711起)、科室为手术室(835起)、医院类别为二级综合医院(5 320起);有源医疗器械不良事件发生最多的地级市为济南市(2 271起)、科室为手术室(196起)、医院类别为三级综合医院(1 108起);体外诊断试剂不良事件发生最多的地级市为烟台市(42起)、科室为儿科(6起)、医院类别为一级医院(42起)。根据关联规则可知,一级医院中卫生院使用未超期无源器械关联规则支持度最高,而在超期使用产品中,日照市的三级综合医院重症监护室使用有源器械发生不良事件支持度最高。结论各不良事件发生与各级别医院及使用科室存在强关联,而在超期产品使用中也存在类似问题,这可为各监测单位及医疗机构深化管理医疗器械提供指导。

关键词: 数据挖掘, 关联性分析, Apriori关联规则, 医疗器械不良事件, 关联规则挖掘

Abstract: Objective Monitoring of adverse events related to medical devices is an important means of post-marketing risk management of medical devices. The purpose of this study is to analyze the correlation of adverse events related to medical devices in Shandong Province in 2012 based on the Apriori algorithm. Methods A descriptive analysis was conducted on 63 041 adverse injury events reported by monitoring institutions in Shandong Province in 2021, which were classified into three categories based on the classification of generalized medical devices, including passive medical devices, active medical devices and in vitro diagnostic reagents. The Apriori model for association rule mining was used to investigate the risks associated with adverse events related to medical devices by mining the device category, department of use, hospital category, use of expired devices, and the region to which the reported hospital belonged. Results Of the 63 041 adverse events, 20 564 were related to active medical devices, 42 181 to passive medical devices, and 296 to in vitro diagnostic reagents. As for the adverse events related to passive medical devices, the city, department and hospital which had the most adverse events were Yantai(n=5 711), operating room(n=835), and secondary general hospitals(n=5 320). As for the adverse events related to active medical devices, the city, department and hospital which had the most adverse events were Jinan(n=2 271), operating room(n=196), and tertiary general hospitals(n=1 108). As for the adverse events related to in vitro diagnostic reagents, the city, department and hospital which had the most adverse events were Yantai(n=42), the pediatric department(n=6), and primary hospitals(n=42). According to the association rule, the use of unexpired passive devices in health centers among primary hospitals received the most support, while for the use of expired devices, the most support was found in the use of active devices in the intensive care units of tertiary general hospitals in Rizhao. Conclusion The occurrence of adverse events is strongly associated with hospitals and departments at all levels, and similar problems also exist in the use of expired products. Our findings provide references for monitoring units and medical institutions to impprove the management of medical devices.

Key words: Data mining, Association analysis, Apriori rule, Medical device adverse events, Association rule mining

中图分类号:

R197.39

吴雨桐,吴思佳,杨建卫,何依娜,李洪凯,黄琳,刘云霞. 基于Apriori算法分析2021年山东省医疗器械不良事件的关联性[J]. 山东大学学报 (医学版), 2022, 60(12): 111-118.

WU Yutong, WU Sijia, YANG Jianwei, HE Yina, LI Hongkai, HUANG Lin, LIU Yunxia. Association analysis of medical device adverse events in Shandong Province in 2021: Apriori algorithm[J]. Journal of Shandong University (Health Sciences), 2022, 60(12): 111-118.

参考文献

[1] 王兰明. 医疗器械不良事件监测-现状与展望[J]. 中国医疗器械杂志, 2004, 28(4): 282-289. WANG Lanming. Medical devices adverse events monitoring-the present and the future [J]. Chinese Journal of Medical Instrumentation, 2004, 28(4): 282-289.
[2] 鄂恒, 刘禹, 李丹. 医疗器械不良事件案例分析与监测管理改进探讨[J]. 中国医疗器械信息, 2019, 25(9): 179-180.
[3] 侯凤. 医疗器械不良事件案例分析及监测管理措施[J].中国药物与临床, 2021, 21(22): 3767-3768.
[4] 王志. 630例医疗器械不良事件分析报告[J]. 中国医药科学, 2011, 1(16): 136-137.
[5] 程艳敏, 刘岩, 刘亚民. 我国医疗不良事件报告系统研究与应用现状述评[J]. 中国医院管理, 2012, 32(10): 40-42. CHENG Yanmin, LIU Yan, LIU Yamin. Review on research and application of medical adverse events reporting system in China [J]. Chinese Journal of Hospital Administration, 2012, 32(10): 40-42.
[6] 徐军, 蒋鹏程, 王鑫扬, 等. 医疗不良事件报告体系研究[J].中国卫生质量管理, 2009, 16(1): 23-25.
[7] 蒋亚萍, 胡志强. 270例医疗器械不良事件报告分析[J]. 医疗装备, 2021, 34(17): 43-44. JIANG Yaping, HU Zhiqiang. Report analysis about 270 cases of adverse events of medical devices [J]. Chinese Journal of Medical Device, 2021, 34(17): 43-44.
[8] 余永强, 郑玉建. 我国医疗器械不良事件监测现状研究[J]. 中国药物警戒, 2008,5(1): 24-27. YU Yongqiang, ZHENG Yujian. Study on the current situation of medical device adverse events monitoring in China [J]. Chinese Journal of Pharmacovigilance, 2008, 5(1): 24-27.
[9] Rakesh Agrawal, Tomasz Imieliński, Arun Swami. Mining association rules between sets of items in large databases [J]. SIGMOD Rec, 1993, 22(2): 207-216.
[10] 刘华婷, 郭仁祥, 姜浩. 关联规则挖掘Apriori算法的研究与改进[J]. 计算机应用与软件, 2009, 26(1): 146-149. LIU Huating, GUO Renxiang, JIANG Hao. Research and improvement of Apriori algorithm for mining association rules [J]. Computer Applications and Software, 2009, 26(1): 146-149.
[11] Han JW, Kamber M. Data mining concepts and techniques [M]. Beijing: Higher Education Press, 2001.
[12] Niu Zhendong, Nie Yaoxin, Zhou Qian, et al. A brain-region-based meta-analysis method utilizing the Apriori algorithm [J]. BMC Neurosci, 2016, 17(1): 23.
[13] Nahar J, Tickle KS, Ali AB, et al. Significant cancer prevention factor extraction: an association rule discovery approach[J]. J Med Syst, 2011, 35(3): 353-367.
[14] Zhang QR, Long Q, Ott J. Apriori GWAS, a new pattern mining strategy for detecting genetic variants associated with disease through interaction effects [J]. PLoS Comput Biol, 2014, 10(6): e1003627.
[15] 吴小东, 曾玉珠. 基于Apriori算法的高校学生成绩数据挖掘[J]. 廊坊师范学院学报(自然科学版), 2019, 19(1): 31-36. WU Xiaodong, ZENG Yuzhu. Data mining of university student achievement based on apriori algorithm [J]. Journal of Langfang Teachers University(Natural Science Edition), 2019, 19(1): 31-36.
[16] Han EH, Karypis G, Kumar V.Scalable parallel data mining for associ-ation rules [C]. ACM SIGMOD International Conference on Management of Data, 1997.
[17] 陆丽娜, 陈亚萍, 魏恒义, 等. 挖掘关联规则中Apriori算法的研究[J]. 小型微型计算机系统, 2000, 21(9): 940-943. LU Lina, CHEN Yaping, WEI Hengyi, et al. Research on the algorithm Apriori of mining association rules [J]. Journal of Chinese Computer Systems, 2000, 21(9): 940-943.
[18] 崔贯勋, 李梁, 王柯柯, 等. 关联规则挖掘中Apriori算法的研究与改进[J]. 计算机应用, 2010, 30(11): 2952-2955. CUI Guanxun, LI Liang. WANG Keke, et al. Research and improvement on Apriori algorithm of association rule mining [J]. Computers and Applied Chemistry, 2010, 30(11): 2952-2955.
[19] 汤骏. 基于频集的Apriori关联规则算法在招标采购活动中的应用[J]. 南通职业大学学报, 2021, 35(2): 94-99. TANG Jun. Application of Apriori association rule algorithm based on frequency itemset in bidding and purchasing activities [J]. Journal of Nantong Vocational College, 2021, 35(2): 94-99.
[20] 降惠, 尹振保, 武丽娟, 等. Apriori算法在手足口病与气象因素关系分析中的应用[J]. 智能计算机与应用, 2021, 11(7): 124-128. JIANG Hui, YIN Zhenbao, WU Lijuan, et al. Application of Apriori algorithm in the analysis of the relationship between hand-foot-mouth disease and meteorological factors [J]. Intelligent Computer and Applications, 2021, 11(7): 124-128.
[21] 颜雪松, 蔡之华. 一种基于Apriori的高效关联规则挖掘算法的研究[J]. 计算机工程与应用, 2002, 38(10): 209-211. YAN Xuesong, CAI Zhihua. Research on a high-efficient algorithm of mining association rule based on Apriori [J]. Computer Engineering and Applications, 2002, 38(10): 209-211.
[22] 王明艳. 基于Apriori算法的零售商品购买关联分析[J]. 科技创新与生产力, 2021, 341(8): 50-52. WANG Mingyan. Association analysis of retail goods purchase based on Apriori algorithm [J]. Sci-tech Innovation and Productivity, 2021, 341(8): 50-52.
[23] Li Z, Li X, Tang RH, et al. Apriori algorithm for the data mining of global cyberspace security issues for human participatory based on association rules [J]. Front Psychol, 2021, 11: 582480. doi:10.3389/fpsyg.2020.582480.
[24] 黄进, 尹治本. 关联规则挖掘的Apriori算法的改进[J]. 电子科技大学学报, 2003, 32(1): 76-79. HUANG Jin, YIN Zhiben. Improvement of Apriori Algorithm for Mining Association Rules [J]. Journal of University of Electronic Science and Technology of China, 2003, 32(1): 76-79.
[25] 钱光超, 贾瑞玉, 张然, 等. Apriori算法的一种优化方法[J].计算机工程, 2008, 34(23): 196-198. QIAN Guangchao, JIA Ruiyu, ZHANG Ran, et al. One optimized method of Apriori algorithm [J]. Computer Engineering, 2008, 34(23): 196-198.
[26] 曾令贵, 龙丽萍, 姚金成, 等. 3 123例可疑医疗器械不良事件统计分析[J]. 药物流行病学杂志, 2008,17(6): 391-394. ZENG Linggui, LONG Liping, YAO Jincheng, et al. Analysis of 3,123 reports of adverse events caused by medical devices [J]. Chinese Journal of Pharmacoepidemiology, 2008, 17(6): 391-394.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed