CT image lung lobe image segmentation system based on attention mechanism

Abstract

A CT image lung lobe image segmentation system based on an attention mechanism comprises a foreground preprocessing module, a significance feature analysis module, a morphological feature extraction module, a segmentation reconstruction module and an I / O management module. According to the invention,the image saliency features related to the lung fissure region are extracted through the saliency region segmentation network, so that thelung fissure feature space is positioned; the network realizes fusion of high-level semantic features and low-level semantic features on the basis of an attention mechanism, so that the attention degree of a model on unrelated features is reduced, the detection efficiency of a lung fissure image is improved, meanwhile, three-dimensional direction information and saliency features are fused into a DoS (Delivery of Stick, narrow strip differential) filtering process, therefore, extraction of morphological characteristics of the lung fissure is guided, and the accuracy of lung lobe image segmentation is improved.

Description

technical field [0001] The present invention relates to a technique in the field of image processing, in particular to a CT image lung lobe image segmentation system based on an attention mechanism. Background technique [0002] The lung lobe image segmentation technology based on the physiological structure of the lung fissure can realize the visualization of the lesion area in different lung lobes and the calculation of the volume ratio, thereby assisting doctors in the diagnosis and monitoring of the disease. The existing lung lobe image segmentation technology has the following problems: First, the fissure structure has the problem that the imaging features are not obvious, especially in some CT layers due to factors such as focus interference and lung tissue occlusion. difficulty. Existing schemes usually screen fissure features based on artificial prior knowledge or adding additional post-processing operations. Such methods rely on a large amount of supervised data o...

AI Technical Summary

Problems solved by technology

The existing lung lobe image segmentation technology has the following problems: First, the fissure structure has the problem that the imaging features are not obvious, especially in some CT layers due to factors such as focus interference and lung tissue occlusion. difficulty

Such methods rely on a large amount of supervised data or human interaction, which is difficult to meet the needs of clinical applications; second, the proportion of the lung fissure area in the whole lung is relatively low, and the existing feature extraction algorithm based on iterative scanning of the whole lung does not detect irrelevant features. Information is screened out, resulting in a large amount of calculation for the algorithm as a whole and low execution efficiency

Images