基于可视化物联网的大型医疗设备运行可靠性预测平台研究

Research on Operation Reliability Prediction Platform of Large-Scale Medical Equipment Based on Visual Internet of Things

针对医疗设备可靠性在临床实践中的问题,开发了一个基于物联网和可视化技术的大型医疗设备运行可靠性预测平台。首先引入可视化物联网仿真建模技术,利用传感器集群实时获取大型医疗设备运行多维状态数据,形成大型医疗设备运行可靠性核心要素样本集;然后对大型医疗设备运行可靠性核心要素样本集进行池化处理,构建融合前置训练集和后置测试集的可靠性核心要素样本特征数据池;最后利用深度卷积神经网络DCNN对特征数据池进行特征辨识,构建时间正序下的大型医疗设备运行可靠性核心要素样本集精准预测机制。选取某三甲医院核磁共振成像设备为案例分析载体,对平台进行了临床应用实践验证,结果表明,该平台有效地满足了大型医疗设备运行可靠性预测的智慧化改造需求,显著优化了预测的智慧可控感知机制,且平台的核心参数符合临床实践的标准要求。

Addressing medical device reliability in clinical practice,several shortcomings that have become increasingly prominent in clinical practice,developed the large-scale medical equipment operation reliability prediction platform based on the visual Internet of Things.First,introduce the visualization of Internet of Things simulation modeling technology,use sensor clusters to obtain real-time multi-dimensional status data of large-scale medical equipment operation,and form a sample set of core elements of large-scale medical equipment operation reliability;Then pool the sample set of the core elements of the reliability of the operation of large medical equipment,and construct the data pool of the core elements of reliability that combines the pre-training set and the posttest set;Finally,the deep convolutional neural network DCNN i s used to identify the characteristics of the feature data pool,and build the accurate prediction mechanism for the sample set of the core elements of the reliability of large-scale medical equipment in positive time sequence.The nuclear magnetic resonance imaging equipment of the tertiary first-class hospital was selected as the case analysis carrier,and the platform was verified in clinical application practice,the results show that the platform satisfies the needs of intelligent transformation of large-scale medical equipment operation reliability prediction,and greatly optimizes the intelligent controllable perception mechanism of large-scale medical equipment operation reliability prediction,and the core parameters of the large-scale medical equipment operation reliability prediction platform meet the requirements of clinical practice.