Construction method and construction system for critical risk prediction model of COVID-19 patient

Abstract

The invention discloses a COVID-19 patient critical risk prediction model construction method and system, and the method comprises the following steps: S1, collecting the basic information of a COVID-19 patient, including three pieces of information, i.e., the age, whether the COVID-19 patient suffers from an aggregative disease or not, and the number of days of developing from a common disease to a critical disease; s2, a sample of a COVID-19 patient is collected and processed; s3, measuring clinical indexes of the sample, including five clinical index values, namely serum albumin, platelet count, glutamic-pyruvic transaminase, C-reactive protein and neutrophil percentage; and S4, fitting data of age, whether aggregation disease occurs or not, serum albumin, platelet count, glutamic-pyruvic transaminase, C-reactive protein and neutrophil percentage by adopting a multi-factor proportional risk regression model, constructing a prediction model, and displaying a result of the prediction model in the form of a column diagram. The prediction model has readability, and is convenient for evaluating whether a common disease patient progresses into a critical patient or not.

Description

technical field [0001] The invention relates to the technical field of severe risk prediction for patients with COVID-19, in particular to a method and system for constructing a severe risk prediction model for COVID-19 patients based on clinical indicators. Background technique [0002] COVID-19 is the seventh coronavirus discovered so far that can infect humans. From the existing epidemiological data, the transmission speed of COVID-19 is significantly faster than that of SARS-CoV, and its infection rate is stronger than that of SARS-CoV. In terms of pathogenicity, COVID-19 is weaker than SARS-CoV. Most people infected with COVID-19 have mild or no symptoms, and only about 20% of infected people have severe clinical manifestations. However, COVID-19 is more transmissible, and the incidence of infection has obvious clustering phenomenon. Therefore, early detection, early isolation, early diagnosis, and early treatment of patients infected with the new coronavirus are the ...

AI Technical Summary

Problems solved by technology

If the hazard ratio is close to 1, the predictor does not affect the survival event; if the hazard ratio is less than 1, the predictor is protective; if the hazard ratio is greater than 1, the predictor increases the risk

Images