Research On Early Warning Model Of Hospital Audit Based On Information Fusion

With the continuous expansion and improvement of hospital business,the hospital audit work will be more and more complicated.The traditional audit methods will not satisfy today's digital information technology development requirements.Therefore,continuous audit work will become the future development trend.Although the information fusion technology has widely applied in civil fields,but they are mostly concentrated in medical diagnosis and financial crisis.The application of continuous audit is still on the stage of fundamental research.According to the present situation of hospital audit,this paper takes neural network Back Propagation(BP)algorithm and Dempster-Shafer theory of evidence(D-S evidence)as the research core to build a model of hospital audit early warning which bases on information fusion.This research has achieved the function of judge audit data and audit crisis warning,in order to reach the goal of real-time monitoring hospital audit activities.The main contents of the paper are summarized as follows:(1)Through browsing the research status of domestic and foreign,this paper analyzes and summarizes the problems exists in the audit warning era where it used the information fusion technology,and then the main contents of the paper are determined.This paper makes a brief introduction of the main theory as well as technique of the early warning model in order to provide the theoretical basis of the final system design and implementation.(2)In view of the current problems in the hospital audit process,this paper analyzes the main attribute of audit data and contrasts the advantages and disadvantage of financial forecasting algorithm.The neutral network BP algorithm is selected to construct the model of audit early-warning.At the same time,the optimization method which covers Opposition-Based Genetic Algorithm(OBGA)combining number of fixed hidden layer neurons has been put forward according to the flaw which exists in the neutral network BP algorithm.Using OBGA to find the optimal combination of input variables and the number of hidden layer neurons to improve the prediction accuracy,the model of audit early warning caller OBGA-FHBP is established.Experiments show that the model which bases on OBGA-FHBP algorithm has a significant improvement in accuracy rate,modeling time,networking error accuracy as well as the stability.(3)According to the abnormal data generated by auditing activities,the audit crisis early-warning index prioritization framework has been established based on the theory of D-S evidence.Basing on that framework,we find the audit crisis warning indicator that obtained the most abnormal data and establish the DR-Z audit crisis early-warning model.Experiments prove that audit crisis early warning model can correctly alert the hospital operating condition and make a more objective judgment on the crisis risk,and achieve warning function in advance or in the matter.(4)According to the design goal of the early-warning hospital audit system,the overall design and detailed design of the system are carried out,and the development of the system is completed.Through the display of the system,it is proved that the system can not only manage and judge the data,but also effectively realize the hospital audit crisis early-warning and risk assessment.The system includes four functional modules: audit data review,abnormal data processing,crisis warning and audit results publicity.Finally,this system achieves the intelligent and integrated of the audit process operation.