Colorectal cancer prediction method and device based on marker gene and hybrid kernel function SVM

Abstract

The invention discloses a support vector machine classifier construction method and a support vector machine classifier construction device for colorectal cancer prediction. The method comprises the steps of: acquiring health and colorectal cancer sample data and preprocessing the data; determining disease-related feature genes based on the two sets of sample data; utilizing a Gaussian kernel function, a polynomial kernel function and a linear kernel function to construct a hybrid kernel function support vector machine; and optimizing parameters of the hybrid kernel function support vector machine. The support vector machine constructed by adopting the method and the device is more suitable for performing classification based on marker genes and can save time for colorectal cancer judgment.

Description

technical field [0001] The invention belongs to the field of disease auxiliary prediction, and in particular relates to a colorectal cancer prediction method and device based on marker genes and mixed kernel function SVM. Background technique [0002] Colorectal cancer is one of the most common malignant tumors. About 1.2 million people around the world suffer from colorectal cancer every year, and as many as 600,000 patients die from the disease, which not only causes huge economic losses, but also has serious consequences for human health. serious threat. Early screening for the disease is key to successful treatment and patient survival, and is a major current public health challenge. The traditional diagnostic methods of colorectal cancer include X-ray examination, serum carcinoembryonic antigen (CEA) examination, endoscopy, etc. These methods play a vital role in the diagnosis of colorectal cancer, but the patient compliance of these methods is low , and it is necessa...

AI Technical Summary

Problems solved by technology

At present, the use of machine learning algorithms to predict colorectal cancer mainly has the following problems: (1) redundant disease feature factors (2) inability to correctly select the number of features (3) inability to correctly select an appropriate classifier

However, when solving practical problems, people's usual practice is to pre-select a single kernel function based on the prior knowledge of experts, but this method often results in the selected kernel function having only a single property, which cannot solve the problem well.

Images