Esophageal cancer risk prediction method based on som neural network and svm

Abstract

The present invention proposes a risk prediction method for esophageal cancer based on SOM neural network and SVM, the steps of which are as follows: firstly, collect M blood index information and survival period information of patients with esophageal cancer as the original data set; then, use SOM neural network The network clusters M blood indicators to obtain the clustering results of M blood indicators; then uses the COX risk regression model to perform regression verification on the clustering results, and obtains information on N blood indicators that are significantly related to the survival of patients with esophageal cancer; Then, by drawing the ROC curve, find the critical threshold of the survival period, and divide the risk level; finally, use the genetic algorithm to optimize the parameters of the SVM, and select the RBF kernel function to establish the risk prediction model of esophageal cancer. The present invention can simultaneously find a plurality of blood indexes significantly related to the survival period through SOM neural network clustering and SVM model construction, and reasonably, conveniently and effectively predict the risk level of esophageal cancer.

Description

technical field [0001] The invention relates to the technical field of cancer risk assessment, in particular to an esophageal cancer risk prediction method based on SOM neural network and SVM. Background technique [0002] Cancer has always been one of the leading causes of death in both developed and developing countries, causing a huge social and economic burden. Esophageal cancer is one of the tumor types with high morbidity and mortality worldwide, and ranks sixth among the causes of tumor-related death. More than 300,000 people die from esophageal cancer every year in the world, and 90% of the cases are Esophageal squamous cell carcinoma. my country is one of the regions with a high incidence of esophageal cancer in the world, and esophageal cancer has become an important disease affecting the health of our people. [0003] On the one hand, with the advancement of science and technology and the innovation of medical technology, the treatment methods and concepts of es...

AI Technical Summary

Problems solved by technology

[0004] In the contemporary social environment, with the continuous expansion of medical data, traditional technologies and methods can no longer cope with the processing of large amounts of complex data.

Images