Research On Diagnosis Resource Scheduling Based On Y Hospital Outpatient Volume Prediction

As an important window connecting the society and the hospital,outpatient visits have an important impact on the operation and management of the hospital.Accurate prediction of cardiovascular disease outpatient volume is very necessary,which can provide objective basis and rationality for their own management decisions and future planning,reasonable use of medical and health resources to improve medical and health standards.Precisely predict the number of outpatient visits in the next cycle(next week,next month or next year),provide theoretical support for outpatient operation management decisions,and arrange the budget and investment amount of each department according to the predicted number of patients;reasonable allocation of various internal resource scheduling to improve resource utilization efficiency,an important foundation for medical operation management.This article starts from the actual scene of the outpatient clinic of Y Hospital,and proposes outpatient diagnosis resource scheduling based on prediction for the outpatient consultation.Conduct research from two aspects:(1)In order to improve the prediction accuracy of cardiovascular disease outpatient volume,consider that cardiovascular disease is affected by many factors,resulting in complex,frequent,nonlinear,non-stationary characteristics of the outpatient volume time series,this paper constructs a data features combined prediction model based on the idea of "decomposition-reconstruction-ensemble".First,an improved ensemble empirical mode decomposition that avoids end-point effects is used to decompose the complex cardiovascular disease outpatient time series,and several relatively simple components are obtained,which simplifies the data with complex characteristics.In order to avoid data redundancy,according to the fuzzy entropy reconstruction algorithm,the decomposed data components are reconstructed,and three subsequences of high frequency,low frequency and trend are obtained.According to the data characteristics,the autoregressive integrated moving average model is adopted for the high-frequency sub-sequences,and the back-propagation neural network is used for the low-frequency and trend sub-sequences for prediction.Finally,the prediction results are integrated by linear superposition.(2)Based on the prediction results of cardiovascular diseases,aiming at the limited medical diagnosis resources,establish a mathematical model aiming at minimizing the total examination time of patients 'outpatient projects,and provide reasonable outpatient diagnosis resources optimization configuration for patients' examination projects.Reduce the waste of medical resources and make reasonable procurement suggestions for some key bottleneck resources to reduce the imbalance between supply and demand between patients and hospitals.Finally,the time series data of outpatient volume of cardiovascular disease and data of outpatient diagnosis resources of Y Hospital of Dalian City were selected as samples for empirical research.In order to verify the prediction accuracy of the established decompositionreconstruction-integrated combined prediction model,the experimental results show that it is compared with other single prediction models and decomposition-integrated models that do not consider reconstruction.The combined prediction model proposed in this paper is more accurate in predicting the outpatient volume of cardiovascular diseases.And according to the predicted number of patients,an example of outpatient diagnosis resource scheduling is provided to provide data support for Y hospital's resource allocation,medical operation management,and future planning.