Liver focus diagnosis method and device based on CT image

Abstract

The invention provides a liver focus diagnosis method and device based on a CT image. The method comprises the following steps: finding an image belonging to the same lesion in three phases through liver lesion matching on the three phases of plain scanning, artery and vein of a CT image of the same case; respectively determining a three-dimensional space occupied by each focus on three phases, and if one phase lacks an image of a certain focus, determining a three-dimensional space occupied by the focus on the phase based on the three-dimensional spaces occupied by the focus on other phases; digging the focus down from each phase and scaling to the same target size; and respectively carrying out focus feature extraction on each phase, fusing the focus features extracted from the three phases, and inputting the fused features into a classifier to obtain the category to which each focus belongs. The invention can make full use of the expressions and signs of the lesion in multiple phases for lesion diagnosis, and improves the accuracy of lesion diagnosis.

Description

technical field [0001] The invention belongs to the technical field of medical imaging, in particular to a method and device for diagnosing liver lesions based on CT images. Background technique [0002] In recent years, with the development of deep learning and medical imaging technology, applying deep learning technology to medical images has become a relatively hot research field. At present, deep learning technology has achieved good results in related image processing of lungs, breasts and brains. [0003] A complete CT image contains at least three phases of plain scan, arterial and venous data. At present, the diagnostic technology of liver lesions using CT images is mainly based on plain CT scans to make a diagnosis of the nature of the lesions. Generally, the liver segmentation model is first used to obtain the segmented area of ​​the liver, and then the liver area on the plain scan is extracted as the input of the classification model, and the entire liver is dia...

AI Technical Summary

Problems solved by technology

Diagnosis based on a single phase (such as plain scan) cannot make full use of lesion information in other phases, so it is difficult to accurately diagnose the type of lesion, especially in the case of the same shadow but different diseases

Images