MSI prediction model construction method based on gastric cancer histopathology image texture features

Abstract

The invention discloses an MSI prediction model construction method based on gastric cancer histopathological image texture features. The method comprises the following steps: obtaining histopathological images of a gastric cancer patient, marks of focus parts and clinical pathological information; extracting texture features of the original image and texture features after wavelet transform for the histopathological image of the patient with gastric cancer; performing feature selection on the obtained texture features by using LASSO, and selecting non-zero coefficient features corresponding to the lambda value when 10 times of cross validation error is minimum to obtain screened texture features; carrying out linear fitting according to the characteristic value of the selected texture characteristic and the coefficient weight of the characteristic value so as to obtain an MSI label of the gastric cancer patient; and combining the MSI label and the clinical pathological information ofthe patient to construct an MSI prediction model. According to the method, the MSI state of the gastric cancer patient is directly predicted on the basis of the histopathological image which is easy to obtain, an additional laboratory is not required to carry out gene detection and immunohistochemical analysis, and the MSI state can be detected at a lower cost.

Description

technical field [0001] The invention relates to the technical field of computer medical image information processing, in particular to a method for constructing an MSI prediction model based on the texture features of histopathological images of gastric cancer. Background technique [0002] There are two main traditional MSI detection methods: Immunohistochemistry (IHC) and polymerase chain reaction (PCR). Genetic analysis of gene markers at nucleotide sites; however, both IHC and PCR detection methods need to be carried out in large-capacity tertiary medical centers, and require high economic and time costs, and it is difficult to extend to every clinical practice. a patient. As a result, timely immune checkpoint inhibitor therapy cannot be provided to a large number of potentially immunotherapy-sensitive individuals, thereby losing the opportunity to control the disease. [0003] Histopathology has been an important tool for cancer diagnosis and prediction, and its pheno...

AI Technical Summary

Problems solved by technology

[0002] There are two main traditional MSI detection methods: immunohistochemistry (Immunohistochemistry IHC) and polymerase chain reaction (PCR). However, both IHC and PCR detection methods need to be carried out in large-capacity tertiary medical centers, and require high economic and time costs, so it is difficult to extend to every patient in clinical practice. a patient

As a result, timely immune checkpoint inhibitor therapy cannot be provided to a large number of potentially immunotherapy-sensitive individuals, thereby losing the opportunity to control the disease

Images