Research and development of a platform based on digital twin technique for predicting fault,and managing and controlling risk of medical equipment
10.3969/j.issn.1672-8270.2025.08.023
- VernacularTitle:基于数字孪生技术的医疗设备故障预测与风险管控平台研发
- Author:
Song HAN
1
;
Yan HE
;
Jia YANG
;
Li HUANG
Author Information
1. 首都医科大学附属北京同仁医院总务处 北京 100176
- Publication Type:Journal Article
- Keywords:
Digital twin;
Medical equipment management;
Fault prediction;
Risk management and control;
Intelligent maintenance;
Deep tensor network
- From:
China Medical Equipment
2025;22(8):125-129,135
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To develop a platform based on digital twin technique for predicting fault,and managing and controlling risk of medical equipment,so as to resolve the existed problems of the blind spots of spatiotemporal monitoring and the data fragmentation in conventional maintenance mode for medical equipment.Methods:This study established a bidirectional interaction system of integrating physical equipment with virtual models,and fused multi-source heterogeneous data to establish a prediction model that was driven by a deep tensor network for faults,and developed a platform for predicting fault,and managing and controlling risk so as to conduct operation,maintenance and management for equipment.Twenty imaging equipment in clinical use at Beijing Tongren Hospital of Capital Medical University from January 2022 to December 2024 were included.Conventional maintenance mode was implemented to maintain 15 equipment in use during January 2022 and June 2023.The platform(platform maintenance mode)based on digital twin technique for predicting fault,and managing and controlling risk of medical equipment was adopted to maintain 20 equipment(including five equipment that were newly added)from July 2023 to December 2024.The differences of predictive performance,efficacy of managing and controlling risk,and cost-effectiveness assessment for fault of equipment between the two kinds of maintenance modes were compared.Results:The average accuracy rate,recall rate,and average weighted accuracy rate of the fault severity index(FSI)of maintaining equipment of using the platform maintenance mode were respectively(92.52±2.33)%,(89.23±3.12)%and(94.12±1.83)%,all of which were higher than those of the conventional maintenance mode,while the average false alarm rate of the platform maintenance mode was(7.83±1.52)%,which was lower than that of the conventional maintenance mode,and the differences of the above indicators between two modes were statistically significant(t=19.234,17.256,20.976,18.365,P<0.05).The average response time for equipment maintenance and the times of occurring events with risk in using platform maintenance mode were lower than those in using conventional maintenance mode,and the differences were significant(t=15.273,37.454,P<0.05).The manpower cost of emergency in using platform maintenance mode was also lower than that in using the conventional maintenance mode,with a statistically significant difference(U=215.783,P<0.05).The preventive maintenance cost of equipment,repair cost of fault,and the cost of downtime losses of equipment in using the platform maintenance mode were all lower than those in using the conventional maintenance mode,while the average utilization rate of equipment was higher than that in using the conventional maintenance mode,and the differences were statistically significant(t=13.058,8.962,10.465,10.513,P<0.05).Conclusion:The platform based on digital twin technique for predicting fault,and managing and controlling risk of medical equipment can reduce unplanned downtime of equipment,and achieve precise prediction for fault of equipment,and full-process management and control for risk of equipment.
