Research And Implementation Of Artificial Intelligence-assisted Infection Monitoring And Early Warning System

The monitoring and detection of nosocomial infection is a very important issue faced by hospitals.Although infection monitoring and alarm systems have been used in hospitals in many countries and regions,outbreaks of nosocomial infection still occur from time to time.The main reason is that the traditional infection monitoring and alarm system mainly implements infection monitoring through statistical analysis of the information of infected patients actively reported by doctors.The reliability and real-time nature of infection monitoring depends on the timeliness and consciousness of the doctor’s initiative to report.In addition,there are problems such as inability to monitor new infections caused by changes in microbial resistance and inability to accurately determine whether infection has occurred in infected patients.How to use the computer system to improve the efficiency of infection monitoring and reliability of infection monitoring,realize the monitoring of microbial drug resistance and timely send out the infection alarm,make full use of the monitoring data for intervention prompt,so as to improve the efficiency of hospital infection monitoring and the rationality of clinician treatment,is the urgent demand of the current medical staff and infection control personnel.In this study,we designed and implemented an artificial intelligence assisted infection monitoring and alarm system which utilize the artificial intelligence Jess(Java expert system shell)inference engine and data mining technology.On the one hand,it solves the problem of low accuracy and efficiency of traditional infection monitoring system.On the other hand,the system sends out an infection alarm after monitoring the infection event and generates an intelligent report to provide clinicians with the best antibiotic list for adjuvant treatment.In addition,the system also integrates a statistical analysis module to help users intuitively find the trend of infection data and the characteristics of infection events in the hospital.In the process of system implementation,firstly,the requirement analysis and function description of AI aided infection monitoring system,including business process analysis,system function requirement description and system non function requirement analysis.Then the system is designed and implemented,in which the Jess infection monitoring module and knowledge discovery module are designed in detail.In the Jess infection monitoring module,Jess inference engine is used to realize infection monitoring and alarm,and generate intelligent report.The infection monitoring module is divided into initialization sub module,Jess infection monitoring sub module and Jess communication sub module for detailed design;In the knowledge discovery module,the association rule algorithm aprior algorithm is applied to realize rule mining,and the discovered rules are used to update the knowledge base of Jess expert system.Finally,the software test of each functional module and the data verification of the reliability of the intelligent report and data mining under the guidance of domain experts,verify that the task of generating the intelligent report by the JESS infection monitoring module has reached a high degree of reliability.Among them,antibiotics the accuracy of reporting changes in resistance was 94%,and the sensitivity was97.5%.