Hepatic venous pressure gradient classification method of CT image as well as computer equipment

Abstract

The embodiment of the invention relates to a hepatic venous pressure gradient classification method of a CT image as well as computer equipment. The method comprises the following steps: performing liver image cutting pretreatment on the acquired abdomen CT image of a tester to acquire a plurality of to-be-detected liver image blocks; input the plurality of to-be-detected liver image blocks into aliver predicting submodel which is trained in advance to perform hepatic venous pressure gradient classification test to acquire a plurality of predicting classification of each to-be-detected liverimage block as well as predicting classification probability corresponding to the classification, wherein the hepatic venous pressure gradient of the tester is the same predicting classification within the same preset hepatic venous pressure gradient range; and calculating the average value of the classification probability belonging to the same predicting classification, and determining the maximum value of each average value and the predicting classification corresponding to the maximum value as the hepatic venous pressure gradient classification test result of the tester. According to eachembodiment of the invention, multi-classification noninvasive test of the hepatic venous pressure gradient is realized, the test classification result is high in precision and the generalization ability is higher.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a hepatic vein pressure gradient classification method of CT images and computer equipment. Background technique [0002] Portal hypertension refers to a group of clinical syndromes caused by persistently elevated pressure in the portal venous system. It is common in the development of chronic liver disease into cirrhosis, and its poor clinical diagnosis has caused a certain degree of public health problems. In traditional techniques, the common non-invasive diagnosis and detection techniques of portal hypertension are mainly used for the identification of clinically significant portal hypertension, and are easily interfered by different factors. At the same time, in the traditional technology, although there is clinical diagnosis by building a predictive model, the generalization ability of the model is relatively limited, and it is not suitable for selecting bet...

AI Technical Summary

Problems solved by technology

In traditional techniques, the common non-invasive diagnosis and detection techniques of portal hypertension are mainly used for the identification of clinically significant portal hypertension, and are easily interfered by different factors

At the same time, in the traditional technology, although there is clinical diagno

Images