人工智能医疗器械企业质量管理体系构建关键指标筛选研究  

Research on Key Indicators Screening for Quality Management System Construction of Artificial Intelligence Medical Device Enterprises

作　　者：刘毅 王浩[5] 李澍[5] 任海萍 樊瑜波[2,3] LIU Yi;WANG Hao;LI Shu;REN Haiping;FAN Yubo(School of Biological Science and Medical Engineering,Beihang University,Beijing 100191,China;Beijing Advanced Innovation Centre for Biomedical Engineering,Key Laboratory for Biomechanics and Mechanobiology of Chinese Education Ministry,School of Biological Science and Medical Engineering,Beihang University,Beijing 100191,China;School of Engineering Medicine,Beihang University,Beijing 100191,China;Beijing Beiling Special Purpose Vehicle Co.,Ltd.,Beijing 101500,China;Institute for Medical Devices Control,National Institutes for Food and Drug Control,Beijing 102629,China;Medical Technology Academy of Sinopharm Group Co.,Ltd.,Beijing 100028,China)

机构地区：[1]北京航空航天大学生物与医学工程学院,北京100191 [2]北京航空航天大学生物医学工程高精尖创新中心,生物力学与力学生物学教育部重点实验室,生物与医学工程学院,北京100191 [3]北京航空航天大学医学科学与工程学院,北京100191 [4]北京北铃专用汽车有限公司,北京101500 [5]中国食品药品检定研究院医疗器械检定所,北京102629 [6]国药集团医疗器械研究院,北京100028

出　　处：《中国医疗设备》2021年第3期24-27,43,共5页China Medical Devices

基　　金：国家重点研发计划(2020YFC2007104)。

摘　　要：目的筛选出与人工智能医疗器械(Artificial Intelligence Medical Device,AIMD)企业质量管理体系相关的关键指标,从而为进一步构建生产研发过程中AIMD质量管理体系提供参考依据。方法基于YY/T 0287-2017《医疗器械质量管理体系用于法规的要求》,通过两轮专家咨询法,筛选出与AIMD质量管理体系相关的指标,并对这些指标采用Likert 5点法进行权重打分。打分结果经过数据格式化处理后,应用IBM SPSS 21.0进行统计学分析,计算各指标的权重均值、标准差、变异系数(Coefficient of Variation,CV值)和满分率。结果一共发放15份调研问卷,回收15份问卷,回收率为100%,专家的权威程度Cr值为0.91。共筛选得到5个一级指标、12个二级指标,36个三级指标,所有指标的平均分值介于3.40~4.93之间,标准差均小于1.00,CV值均小于25%。其中“设计和开发验证”“设计和开发确认”和“设计和开发更改的控制”三个指标的得分均值分别为4.93、4.87和4.87,标准差分别为0.26、0.35和0.35,CV值分别为5.23%、7.23%和7.23%,满分率分别为93.33%、86.67%和86.67%。结论通过两轮专家咨询法,基本确定了构建AIMD企业生产研发过程中的产品质量管理体系的关键指标,将为后续进一步构建产品质量管理体系提供参考依据。Objective To screen out the key indicators related to the quality management system of artificial intelligence medical device(AIMD),so as to provide reference for the further construction of AIMD quality management system in the process of production and development.Methods Based on YY/T 0287-2017 Medical Device Quality Management System Requirements for Regulatory Purposes,two rounds of expert consultation were conducted to select the indicators related to AIMD quality management system,and Likert 5-point method was used for weight scoring of these indicators.After data formatting,IBM SPSS 21.0 was used for statistical analysis to calculate the weight mean,standard deviation(SD),coefficient of variation(CV)and full score rate of each index.Results A total of 15 questionnaires were issued and 15 questionnaires were collected,with the response rate of 100%,and the authority coefficient(Cr)value of experts of 0.91.A total of 5 first-level indicators,12 second-level indicators and 36 third-level indicators were screened.The average scores of all indicators ranged from 3.40 to 4.93,with SD less than 1.00 and CV less than 25%.The average scores of“design and development verification”,“design and development confirmation”and“design and development change control”were 4.93,4.87 and 4.87 respectively,the SD were 0.26,0.35 and 0.35 respectively,the CV were 5.23%,7.23%and 7.23%respectively,and the full score rates were 93.33%,86.67%and 86.67%respectively.Conclusion Through two rounds of expert consultation,the key indicators for the construction of product quality management system in the production and development process of AIMD enterprises are basically determined,which will provide reference for the further construction of product quality management system.

关 键 词：人工智能 人工智能医疗器械 产品质量管理体系 专家咨询法 关键指标 

分 类 号：R197.39[医药卫生—卫生事业管理] TP18[医药卫生—公共卫生与预防医学]