基于关联挖掘的复杂医疗设备故障的检测技术研究

摘要

针对传统的医疗设备故障检测方法一直存在检测精度低,漏检、误检率高的问题,本文提出基于关联挖掘的复杂医疗设备故障检测方法,通过获取故障数据空间坐标,确定故障数据簇中心,采用模糊决策法计算复杂医疗设备故障数据依赖度.在此基础上,基于K-Means聚类算法,对医疗设备故障数据进行聚类,确定贝叶斯评分函数,采用OFWSC算法进行数据簇特征加权,引入关联挖掘法对复杂设备故障数据进行检测.实验对比结果表明,采用改进的故障检测方法,其检测精度、效率、时间均要优于传统方法,具有一定的实用性和优越性.%It is known that the detection precision of traditional medical equipment fault detection method is low, while the rate of leak detection and false drop rate are high. In the present study, a complex medical equipment fault detection method was put forward based on association mining. The failure data dependence of complex medical devices could be calculated by fuzzy decision method via obtaining the fault data space coordinates to determine the failure data cluster center. In addition, the failure data of medical equipment was clustered based on the K-Means clustering algorithm to determine the bayesian score function, and the fault data of complex equipment was tested via introduction of association mining method by using data OFWSC algorithm. The experimental comparison results showed that the detection accuracy, efficiency and time were superior to the traditional methods after applying the improved method of fault detection, which indicates that it has a certain practicality and superiority.