基于反向传播神经网络-数据包络分析的康复医疗设备质量风险预警与服务效率优化研究OA

Objective:To study application value of the early warning for quality risks,and the optimal mode of service efficiency of medical equipment for rehabilitation on the basis of back propagation(BP)neural network-data envelopment analysis(DEA)in refined management for equipment.Methods:An early warning mechanism was established through sensitive characteristics of quality risks of mining equipment of BP neural network,and the DEA model was used to conduct multi-input-multi-output efficiency assessment.The optimal strategies were determined on the basis of key factors of risks to implement management for medical equipment for rehabilitation.A total of 120 rehabilitation equipment in clinical use at Ningbo Rehabilitation Hospital from January 2024 to December 2024 were selected.They were divided into two groups according to the random number table method,with 60 equipment in each group.The two groups were respectively managed by conventional management mode and optimal mode of early warning and serve efficiency for quality risks of medical equipment for rehabilitation on the basis of BP neural network-DEA(optimal management mode).The differences of the indicators included early warning for risks,DEA efficiency,output and increase of cost-benefit between two kinds of management modes were compared.The self-made survey questionnaire was adopted to investigate satisfaction rates of 200 patients,who used equipment to conduct treatment for rehabilitation,for managing equipment,and each management mode involved 100 patients.Results:The average accuracy rate of early warning,average timely rate of response,and average coverage rate of real-time feedback of medical equipment for rehabilitation of adopting optimal management mode were respectively(93.68±2.45)%,(95.22±2.44)%,and(96.22±2.33)%,which were significantly higher than those of conventional management mode,and the differences of them between two groups were significant(t=20.520,15.097,22.619,15.676,P<0.05).The average fault rate of equipment of the optimal management mode was lower than that of the conventional management mode,and the difference was significant(t=15.676,P<0.05).The means of technical efficiency,pure technical efficiency,and scale efficiency of equipment of the optimal management mode group were significantly higher than those of the conventional management mode group,and the difference were significant(t=13.396,13.255,9.226,P<0.05).The person-times of annual serve of equipment for patients,the average satisfaction rate of patients,effectiveness rate of rehabilitation of the optimal management mode group were significantly higher than those of the conventional management mode group,and the differences were statistically significant(t=16.181,29.614,14.316,P<0.05).The average extent of cost-benefit of physical therapy equipment,exercise therapy equipment,occupational therapy equipment,speech therapy equipment,rehabilitation evaluation equipment,and other auxiliary equipment of the optimal management mode group were significantly higher than those of the conventional management mode group,and the differences were statistically significant(t=6.865,13.613,5.970,5.952,5.724,6.454,P<0.05).Conclusion:The application of the optimal mode of early warning and serve efficiency for quality risks of medical equipment for rehabilitation on the basis of BP neural network-DEA in management for medical equipment for rehabilitation can reduce risks of early warning for equipment,and enhance the increased extents of the indicators of DEA efficiency,output and cost-benefit of equipment.