Objective To use the backward propagation(BP)neural network to establish an appropriate consumption ratio prediction model,to help the hospital management department evaluate whether the use of consumables in each department is reasonable.Methods The operating data of the First Affiliated Hospital of Anhui Medical University from January 2021 to May 2023 were selected to construct the dataset.The network model was trained through the training set and its performance was evaluated through the validation set and the test set.Results The BP neural network model was established to predict the consumption ratio.The explained variance was 0.998604 in the validation set,and the mean absolute error was 0.006219.The evaluation index decreased slightly in the test set,the explained variance was 0.962396,and the mean absolute error was 0.027858.All evaluation indexes were better than other models.Conclusion The consumption ratio prediction model based on BP neural network can realize the nonlinear relationship description of the department,total revenue,drug ratio,and hospital visits etc.,and the model output can accurately predict the consumption ratio,which provides quantitative data support for the assessment and evaluation of consumables in each department.
关键词
医用耗材/耗占比/反向传播神经网络/回归模型
Key words
medical consumables/consumption ratio/backward propagation neural network/regression model
维普
万方数据