基于反向传播神经网络算法建立识别定量测量程序随机误差的方法及性能评价  

Establishment and evaluation of a method for identifying the random error in the quantitative measurement procedure based on back propagation neural network

作　　者：梁玉芳[1] 郑华荣[2] 王哲 冯祥 韩泽文 宋彪 程华丽 王清涛[1] 周睿[1] Liang Yufang;Zheng Huarong;Wang Zhe;Feng Xiang;Han Zewen;Song Biao;Cheng Huali;Wang Qingtao;Zhou Rui(Department of Clinical Laboratory,Beijing Chaoyang Hospital Affiliated to Capital Medical University,Beijing 100020,China;National Institute of Metrology,China,Beijing 100029,China;Inner Mongolia Wesure Date Technology Co.,Ltd,Inner Mongolia 010000,China;Department of Laboratory Medicine,Aviation General Hospital,Beijing 100012,China)

机构地区：[1]首都医科大学附属北京朝阳医院检验科,北京100020 [2]中国计量科学研究院,北京100029 [3]内蒙古卫数数据科技有限公司,呼和浩特010000 [4]航空总医院检验科,北京100012

出　　处：《中华检验医学杂志》2022年第5期543-548,共6页Chinese Journal of Laboratory Medicine

基　　金：北京市临床重点专科卓越项目(检验科)。

摘　　要：目的利用反向传播神经网络(BPNN)算法,建立一种可以识别血糖项目随机误差的实时质控新方法并评价模型效能。方法通过北京朝阳医院实验室信息系统导出2019年1月至2020年7月在西门子advia2400分析系统上报告的全部患者血糖信息,共计219000条,作为本研究的无偏数据。人为引入6个偏差生成相应的有偏数据,每种偏差下用2种算法测试。进行计量学及临床评价。结果BPNN步长设置为10,全部偏差下假阳性率均在0.1%以内;MovSD的最佳步长为150,拦截限为10%,全部偏差下假阳性率为0.38%,比BPNN高0.28%。MovSD在0.5与1误差因子下全部未检出,误差因子>1之后,可检出,但MNPed偏高;而BPNN在全部偏差下MNPed均低于MovSD,两者相差最高达91.67倍。计量学溯源过程生成460000条参考数据,采用参考数据评定BPNN模型的不确定度仅为0.078%。结论成功建立了基于BPNN算法识别检测过程随机误差的实时质控方法,模型准确度高,临床效能显著优于MovSD方法。Objective To establish and evaluate a new real-time quality control method that can identify the random errors by using the backpropagation neural network(BPNN)algorithm and taking blood glucose test as an example.Methods A total of 219000 blood glucose results measured by Siemens advia 2400 analytical system from January 2019 to July 2020 and derived from Laboratory Information System of Beijing Chaoyang Hospital Laboratory Department was regarded as the unbiased data of our study.Six deviations with different sizes were introduced to generate the corresponding biased data.With each biased data,BPNN and MovSD algorithms were used and tested,and then evaluated by traceability method and clinical method.Results For BPNN algorithm,the block size was pre-set to 10 and the false-positive rate in all biases was within 0.1%.For MovSD,however,the optimal block size and exclusive limit were 150 and 10%separately and its false-positive rate in all biases was 0.38%,which was 0.28%higher than BPNN.Especially,for the least two error factors of 0.5 and 1,all the random errors were not detected by MovSD;for the error factor larger than 1,random errors could be detected by MovSD but the MNPed was higher than that of BPNN under all deviations.The difference was up to 91.67 times.460000 reference data were produced by traceability procedure.The uncertainty of BPNN algorithm evaluated by these reference data was only 0.078%.Conclusion A real-time quality control method based on BPNN algorithm was successfully established to identify random errors in analytical phase,which was more efficient than MovSD method and provided a new idea and method for the identification of random errors in clinical practice.

关 键 词：人工智能 反向传播神经网络 随机误差 实时质控 患者数据 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] R446[自动化与计算机技术—控制科学与工程]