基于改进粒子滤波算法的医用超声设备寿命预测研究  

Research on Life Prediction of Medical Ultrasonic Equipment Based on Improved Particle Filter Algorithm

作　　者：吴一未 张晔 王文杰 金伟 WU Yiwei;ZHANG Ye;WANG Wenjie;JIN Wei(Department of Medical Engineering,The Affiliated Wuxi People’s Hospital of Nanjing Medical University,Wuxi Jiangsu 214000,China;Center of Information,The Affiliated Wuxi People’s Hospital of Nanjing Medical University,Wuxi Jiangsu 214000,China)

机构地区：[1]南京医科大学附属无锡人民医院医学工程处,江苏无锡214000 [2]南京医科大学附属无锡人民医院信息大数据中心,江苏无锡214000

出　　处：《中国医疗设备》2024年第8期25-31,共7页China Medical Devices

基　　金：江苏省优势学科建设工程项目(YSHL0804-418)。

摘　　要：目的设计基于改进粒子滤波算法的医用超声设备寿命预测方法,解决传统医用超声设备寿命预测过程中不能对设备退化过程进行精准分析,从而导致预测模型参数不能实时更新、预测精准度下降的难题。方法首先将医用超声设备的整个生命周期划分为正常退化状态和非完美维护退化状态,分析设备退化过程,并依据医用超声设备的退化过程建立寿命预测模型;然后利用极大似然理论计算模型中的退化参数,并根据实际退化数据随机更新模型参数,获取更符合设备实际退化过程的寿命预测模型;最后利用改进粒子滤波算法求解该模型,获取医用超声设备寿命预测结果。结果利用本文提出方法建立的模型决定系数在0.80以上且更接近于1,具有较高的医用超声设备寿命预测准确度,均值为88.5%。结论改进粒子滤波算法在医用超声设备寿命预测中具有较高的准确性和可靠性,可以有效识别设备故障发生的早期迹象,提前采取维修或更换措施,避免可能出现的设备损坏和停机。Objective To design a method for predicting the lifespan of medical ultrasound equipment based on an improved particle filter algorithm,to solve the problems of inaccurate analysis of equipment degradation process in traditional medical ultrasound equipment lifespan prediction,which lead to the inability to update prediction model parameters in real time and a decrease in prediction accuracy.Methods Firstly,the entire lifecycle of medical ultrasound equipment was divided into normal degradation state and imperfect maintenance degradation state.The equipment degradation process was analyzed,and a life prediction model was established based on the degradation process of medical ultrasound equipment.Then,the maximum likelihood theory was used to calculate the degradation parameters in the model,and the model parameters were randomly updated based on actual degradation data to obtain a life prediction model that was more in line with the actual degradation process of the equipment.Finally,the improved particle filter algorithm was used to solve the model and obtain the life prediction results of medical ultrasound equipment.Results The model established by the proposed method had a coefficient of determination above 0.80 and was closer to 1,with a high accuracy in predicting the lifespan of medical ultrasound equipment,with an average of 88.5%.Conclusion The improved particle filter algorithm has a high accuracy and reliability in predicting the lifespan of medical ultrasound equipment,which can effectively identify early signs of equipment failure,take maintenance or replacement measures in advance,and avoid possible equipment damage and downtime.

关 键 词：改进粒子滤波算法 医用超声设备 寿命预测 退化过程 模型参数更新 

分 类 号：R197.39[医药卫生—卫生事业管理] TP206[医药卫生—公共卫生与预防医学]