Syndromic surveillance early warning method based on symptom proportion R value

Abstract

The invention discloses a syndromic surveillance early warning method based on a symptom proportion R value. The syndromic surveillance early warning method comprises the following steps of acquiring diagnosis and treatment data of a hospital; screening to-be-analyzed hospital data; determining a to-be-analyzed target disease and a corresponding target symptom thereof by adopting a Delphi method and a user-defined diagnosis name searching method; dividing the to-be-analyzed hospital data into symptomatic field data and asymptomatic field data according to the condition whether a symptomatic field is empty or not; counting data volume according with target symptoms in the symptomatic field data, and extracting symptom characteristics according with the target disease; solving a symptom proportion R value matrix by taking disease entities and a time period as dimensionalities according to the symptom characteristics according with the target disease; solving the case number of the target disease from the symptomatic field data and the asymptomatic field data according to the symptom proportion R value matrix; performing modeling analysis and early warning on the target disease according to the case number of the target disease. The syndromic surveillance early warning method disclosed by the invention is high in efficiency, high in generality and good in transportability, and can be widely applied to the field of disease monitoring.

Description

technical field [0001] The invention relates to the field of symptom monitoring, in particular to a symptom monitoring and early warning method based on the symptom ratio R value. Background technique [0002] In recent years, with the frequent emergence of new infectious diseases, the threat of ancient infectious diseases and bioterrorism has intensified, and the control of infectious diseases is facing a more severe test. An epidemic of infectious diseases in one country or region may soon develop into a disease in many countries or even the whole world. The traditional disease monitoring method refers to the disease control department formulating corresponding monitoring plans to monitor the incidence of certain infectious diseases or syndromes, and selecting monitoring hospitals (ie sentinel hospitals) to obtain monitoring data regularly. The traditional disease monitoring method is based on disease diagnosis. The disease needs to be diagnosed before the monitoring resu...

AI Technical Summary

Problems solved by technology

Poor data quality or wrong data can even lead to false positives and false positives

Therefore, when processing data, the symptom monitoring and early warning system has relatively high requirements for data sources, but in fact, because the data standards used by each hospital are not uniform and different doctors describe the same symptom differently, the diagnosis and treatment data from hospitals are very different. Large, how to accurately extract the incidence of the target disease from the hospital diagnosis and treatment data through data preprocessing before comprehensive analysis (used to count the incidence of symptoms) has become an urgent problem to be solved in the symptom monitoring and early warning system

The current data preprocessing method used in the symptom monitoring and early warning system needs to design a set of extraction schemes for the number of target diseases for each hospital, which has low processing efficiency, poor versatility, and poor portability.

Images