Establishment method of HIV patient talaromyces marneffei disease incidence probability prediction model

Abstract

The invention provides an establishment method of a HIV patient talaromyces marneffei disease incidence probability prediction model, and belongs to the technical field of disease prediction models. The prediction model is a model based on a random forest algorithm. The method comprises the following steps: selecting and extracting patient information data; performing standardization processing on different batches of source data; establishing a random forest model for predicting the incidence probability of the talaromyces marneffei disease in the HIV patient; the model is tested and evaluated, independent variables with small influences are removed, and an optimized prediction model is obtained. According to the method, the collected patient information data is subjected to unified standardization processing and then input into software, the relation between the independent variable and the dependent variable is established, the obtained model prediction result is more reliable and accurate, and an effective method is provided for predicting the incidence rate of the talaromyces marneffei disease of the HIV patient.

Description

【Technical field】 [0001] The invention relates to the technical field of prediction models for disease incidence probability, in particular to a method for establishing a prediction model for the incidence of tamariferiosis in HIV patients. 【Background technique】 [0002] The epidemic situation of AIDS in my country is severe, and Guangxi is one of the hardest-hit areas of HIV / AIDS epidemic in China. With the popularity of ART combined with antiretroviral therapy, various opportunistic infections have gradually become the main cause of death in AIDS patients. In Guangxi, Talaromyces marneffei (TM) has become the most important opportunistic infection after tuberculosis, and it is also the opportunistic infection with the highest fatality rate. T. marneffei, formerly known as Penicillium marneffei, is an endemic infectious fungal disease. T. marneffei is mainly prevalent in Southeast Asian countries such as Thailand and Vietnam, as well as southern China. The onset of the d...

AI Technical Summary

Problems solved by technology

However, when the random forest model algorithm is directly used to predict the incidence of T. marneffei in HIV patients, the data of the independent variables include some biochemical and immune test indicators of the patients, while the same indicators in clinical practice are tested under different instruments and reagents. , there are differences in the range of reference values, if these data are directly input into the model, the established prediction model cannot reflect the real situation, resulting in inaccurate and reliable prediction results

Images