Symptom Monitoring And Early Warning Data Analysis And Method Research

Symptom monitoring has an increasingly significant preemptive advantage for disease prevention,which can perceive abnormal tendency before large-scale disease outbreaks,leaving more adequate preparation for health care prevention.In recent years,symptom monitoring,as a new means of monitoring,is attracting more and more attention.However,there is serious hysteresis in the existing methods of disease surveillance which is inconvenient and low information level in the form of information input,and its collected sample information has no accurate data processing and early warning analysis method.In order to establish a feasible symptom monitoring and early warning system and realize real-time monitoring and data analysis of key areas,this paper uses the accurate early warning model and large data association information mining method to analyze the early warning and response system and obtain fast and accurate warning results.This paper mainly includes the following sections:(1)The expert evaluation method was used to select the data source of the symptom monitoring system,and the data was cleaned by rough data cleaning.(2)This paper designs the Shewhart control charts from the perspective of the likelihood ratio.The CUSUM control chart is proposed as it is not sensitive to small offset in monitoring data.And meanwhile,it completes the design of the CUSUM control chart of monitoring control graph model of the variance and puts forward the Markov chain method to calculate the ATS and ARL under the combination the control chart parameters to compare the sensitivity of model.(3)The paper designs the improved CUSUM control chart with dynamic changes in sampling intervals to find the better detection effect of the dynamic control chart through comparing with the value of ATS and ANSS and puts forward the concept of difference rate.By comparison,the dynamic control chart is superior to the static control chart warning effect under different parameters.(4)CUSUM control chart with variable sampling interval will monitor the variance of the sample data for symptom monitoring.This paper analyzes the symptom data weighted by gray correlation,changes the sampling variance of data.In the last,and obtains that the dynamic control chart has high accuracy and high sensitivity of detection.This paper uses the method of Markov chain to calculate the ATS and the difference rate and gets the conclusion that changing the length of sampling interval control chart has better effect of monitoring results.Hence,its application in symptom monitoring and early warning system effectively improves the sensitivity of the alarm,the alarm time reduced,monitoring efficiency improved.