基于BP神经网络的医疗设备精准预防性维护  

Precise preventive maintenance of medical equipment based on BP neural network

作　　者：罗林聪 钱雷鸣[1] 章莉[2] LUO Lin-cong;QIAN Lei-ming;ZHANG Li(Equipment Department, Hospital of Integrated Traditional Chinese and Western Medicine in Zhejiang Province, Hangzhou 310003, China;Hemodialysis Center, Hospital of Integrated Traditional Chinese and Western Medicine in Zhejiang Province, Hangzhou 310003, China)

机构地区：[1]浙江省中西医结合医院设备科,杭州310003 [2]浙江省中西医结合医院血透中心,杭州310003

出　　处：《医疗卫生装备》2019年第7期78-80,85,共4页Chinese Medical Equipment Journal

摘　　要：目的:针对当前医疗设备预防性维护(preventive maintenance,PM)中存在的问题,基于误差反向传播(back propagation,BP)神经网络实现医疗设备精准PM。方法:以医疗设备的运行数据、仪器使用率、工作时长、历史维护情况等作为输入数据,建立3层的BP神经网络,以BP神经网络经过训练得出的连接权值系数作为决策模型,预测医疗设备的维护期限、维护部件及维护项目。结果:可以预测出医疗设备的维护期限、维护部件及维护项目等,实现了医疗设备精准PM。结论:BP神经网络可以根据医疗设备运行参数及信息化平台上的数据学习决策,提高医疗设备PM的精准度,从而预防在诊断或治疗过程中因医疗设备故障带来的不良事件。Objective To execute medical equipment precise preventive maintenance (PM) based on error back propagation (BP) neural network to solve the existing problems.Methods A 3-layer BP neural network was established by using the operation parameter data of medical equipment, instrument utilization rate, working time and historical maintenance were as input data, and the weight coefficient obtained by BP neural network after training was regarded as decision model to predict the maintenance date, maintenance components and maintenance items, etc. Results Medical equipment maintenance could be predicted involving in the maintenance time limit, maintenance components and maintenance items so as to realize precise preventive maintenance.Conclusion BP neural network can learn and make decisions through the operation parameters of medical equipment and the data on the information platform, improving the accuracy of preventive maintenance of medical equipment, preventing adverse events caused by medical equipment in the process of diagnosis or treatment.

关 键 词：神经网络 BP神经网络 医疗设备 精准预防性维护 医疗设备故障 

分 类 号：R318[医药卫生—生物医学工程] R197.39[医药卫生—基础医学]