Prognostic And Survival Risk Prediction For Esophageal Cancer Research On Neural Network Model

Esophageal cancer is the ninth most common cancer in the world,with a high mortality rate,few early symptoms and poor prognosis.Reasonable and effective prognostic survival risk prediction can help doctors to take appropriate treatment methods to improve the prognosis and improve the survival rate of patients.In this study,the medical data of patients with esophageal squamous cell carcinoma are taken as the research object,and the relationship between patients’ medical indicators and prognosis is explored through a variety of methods,and prognostic survival risk prediction model is established.The main research contents of this paper are as follows:To solve the problem that traditional prognostic indicators are not accurate in the evaluation of the prognosis of patients with esophageal squamous cell carcinoma,a prognostic index model for esophageal squamous cell carcinoma is proposed based on statistical analysis and Least Absolute Selection Operator(LASSO).First,the correlation between the patient’s medical indicators and prognosis is analyzed by using medical statistics.Then,the LASSO method is used to construct the prognostic index,which could be used to judge the prognostic survival risk of patients.The prognostic index model is better than the traditional prognostic index prognostic nutrition index.The prognostic index model improved the prognostic survival risk prediction accuracy of esophageal squamous cell carcinoma patients.Aiming at the problem of low prediction accuracy of the Back Prapagation(BP)neural network prognostic survival risk prediction model for patients with esophageal squamous cell carcinoma,a prognostic survival risk prediction model based on the Hybrid Improvement Chicken Swarm Optimization(HICSO)algorithm is proposed.HICSO algorithm optimized BP neural network prognostic survival risk prediction model.Based on the traditional flock algorithm,the characteristics of the chaotic sequence are added to initialize the population,introducing the nonlinear change of inertia weights are used in exploration and exploitation ability of balance algorithm,using boundary mutation strategy,and "failure" rooster reverse learning search strategy to improve the search performance of the algorithm,get new HICSO algorithm,and use it to optimize the BP neural network.Finally,the HICSO-BP prediction model is established.The performance of the model is evaluated using real medical data of patients with esophageal squamous cell carcinoma.The results showed that the HICSO-BP prediction model has high prediction accuracy and could effectively predict the prognostic survival risk of patients based on their medical indicators.In order to solve the problems of insufficient data feature extraction and low prediction accuracy in the prognostic survival risk assessment model of esophageal squamous cell carcinoma patients.A prognostic survival risk prediction model for esophageal squamous cell carcinoma is proposed based on Relief feature selection algorithm and one-dimensional convolutional neural network(1D-CNN).First,the Relief feature selection algorithm is used to reduce the impact of redundant features unrelated to patient outcomes.Then,the significant features obtained are used as the input factors of 1D-CNN classifier to predict the prognostic survival risk of patients.This method makes full use of the effective characteristics of the data and improves the performance of the prediction model.Based on the prognosis of patients with esophageal squamous cell carcinomas survival risk prediction as the research object,based on the study of esophageal cancer patient medical data,use statistical methods to analyze medical data,use the LASSO regression analysis method,the optimized BP neural network and combining feature selection algorithm of onedimensional convolution neural network in three different algorithms.The prognostic survival risk prediction models of esophageal squamous cell carcinoma patients are established to effectively predict the prognostic status of esophageal squamous cell carcinoma patients,providing reference for the prognostic prediction of esophageal cancer.