基于特征融合的磁共振医疗设备故障自动检测研究  

Research on Automatic Fault Detection of Magnetic Resonance Medical Equipment Based on Feature Fusion

作　　者：李倩 LI Qian(Yunnan College of Business Management,Kunming,Yunnan 650000,China)

机构地区：[1]云南经济管理学院,云南昆明650000

出　　处：《自动化应用》2024年第13期226-229,共4页Automation Application

摘　　要：磁共振医疗设备故障检测通常需要在设备运行一段时间后才能开展,可能无法及时发现和处理一些潜在的故障,导致检测出的故障数目与实际不符,因此,提出基于特征融合的磁共振医疗设备故障自动检测研究方法。首先,深入分析磁共振医疗设备电路的故障特征,通过采集设备的运行数据,提取与故障相关的特征向量。其次,利用特征融合技术整合这些特征向量,形成一个综合的特征向量。最后,通过有效融合多个特征信息,提升状态判断的准确性,全面反映设备状态。为进一步优化特征融合的效果,引入特征加权策略,通过为每个特征赋予不同的权重,使故障检测更聚焦于关键特征,同时抑制不相关或冗余特征的影响,以实现磁共振医疗设备故障检测。结果表明,与基于阈值的故障检测方法和基于概率统计的故障检测方法相比,基于特征融合的磁共振医疗设备故障自动检测方法得出的故障检测结果与实际的检测结果比较吻合,检测精度高,实用性较好。The fault detection of magnetic resonance medical equipment usually needs to be carried out after the equipment has been running for a period of time,which may not be able to detect and handle some potential faults in a timely manner,resulting in the number of detected faults not matching the actual number.Therefore,a research method for automatic fault detection of magnetic resonance medical equipment based on feature fusion is proposed.Firstly,conduct a thorough analysis of the fault characteristics of magnetic resonance medical equipment circuits,and extract feature vectors related to the fault by collecting operational data of the equipment.Secondly,utilizing feature fusion technology to integrate these feature vectors and form a comprehensive feature vector.Finally,by effectively integrating multiple feature information,the accuracy of state judgment can be improved,comprehensively reflecting the equipment status.To further optimize the effectiveness of feature fusion,a feature weighting strategy is introduced,which assigns different weights to each feature,making fault detection more focused on key features while suppressing the influence of irrelevant or redundant features,in order to achieve magnetic resonance medical equipment fault detection.The results show that compared with threshold based fault detection methods and probability statistics based fault detection methods,the fault detection results obtained by the feature fusion based magnetic resonance medical equipment fault automatic detection method are more consistent with the actual detection results,with high detection accuracy and good practicality.

关 键 词：特征融合 磁共振医疗设备 设备故障检测 自动检测 

分 类 号：R318[医药卫生—生物医学工程]