Respiratory system disease patient number prediction method based on lag analysis and LSTM

Abstract

The invention relates to a respiratory system disease patient number prediction method based on lag analysis and LSTM. The influence of atmospheric pollutants and meteorological conditions on human respiratory system diseases is combined, a deep learning technology is adopted to predict the target disease patient number, and the prediction precision of the respiratory system disease patient numbercan be effectively improved. The hysteresis effect of atmospheric pollutants on respiratory system diseases is brought into analysis, a deep learning time step length setting method based on hysteresis analysis is provided, and the adjustment and optimization process of LSTM network parameters can be effectively simplified on the premise of ensuring the rationality of the model.

Description

technical field [0001] The invention relates to the field of disease prevention and control, in particular to a method for predicting the incidence of respiratory diseases based on lag analysis and LSTM. Background technique [0002] In recent years, scholars at home and abroad have conducted a lot of research on the impact of air pollution on the health of the human respiratory system. However, most of these studies use time series research methods to focus on the correlation between the two, based on the relative sensitivity of pollutant concentrations and meteorological elements. Research on disease prediction is still relatively rare, and most of the few related studies are directly based on multiple linear regression models and autoregressive integral moving average models. However, the prediction results of these prediction models have great uncertainty and accuracy Very low, it is difficult to meet the actual needs of business applications. Contents of the invention...

AI Technical Summary

Problems solved by technology

[0002] In recent years, scholars at home and abroad have conducted a lot of research on the impact of air pollution on the health of the human respiratory system. However, most of these studies use time series research methods to focus on the correlation between the two, based on the relative sensitivity of pollutant concentrations and meteorological elements. Research on disease prediction is still relatively rare, and most of the few related studies are directly based on multiple linear regression models and autoregressive integral moving average models. However, the prediction results of these prediction models have great uncertainty and accuracy Very low, it is difficult to meet the actual needs of business applications

Images