Liver space-occupying lesion identification method, device and implementation device

Abstract

The present invention provides a liver space-occupying lesion identification method, device and implementation device, which relate to the technical field of medical imaging, including acquiring the first CT image; The position of the suspected lesion in the first CT image of the first stage; the position of the suspected lesion is merged horizontally and vertically to obtain the position of the detection frame, and the first CT image of each stage is cut vertically according to the position of the detection frame to obtain the second CT image; through the second recognition model Identify the type of lesion in the suspected lesion position in the second CT image, and judge the target lesion position from the suspected lesion position; segment the first CT image by the third recognition model to obtain the third CT image, according to the third CT image The position coordinates of the target lesion are obtained from the detection frame retained on the image, and the horizontal characteristics of the four-phase images and the vertical characteristics of the single-phase images are fully considered, which greatly improves the accuracy of the recognition results.

Description

technical field [0001] The invention relates to the technical field of medical imaging, in particular to a method, device and implementation device for identifying liver space-occupying lesions. Background technique [0002] Space-occupying lesion is a special term in medical imaging diagnostics. It usually appears in X-ray, B-ultrasound, CT and other examination results. , arterial phase, portal venous phase and delayed phase four-phase images. [0003] At present, liver space-occupying lesions are mainly identified manually from CT images, and automatic computer identification is mostly based on single-phase images of CT images, which cannot consider the horizontal characteristics between the four-phase images and the longitudinal characteristics of a single-phase image. The accuracy of the recognition result is not high. Contents of the invention [0004] In view of this, the purpose of the present invention is to provide a liver space-occupying lesion identification ...

AI Technical Summary

Problems solved by technology

[0003] At present, liver space-occupying lesions are mainly identified manually from CT images, and automatic computer identification is mostly based on single-phase images of CT images, which cannot consider the horizontal characteristics between the four-phase images and the longitudinal characteristics of a single-phase image. The accuracy of the recognition result is not high

Images