Superpixel-based heart MRI image segmentation method and MRI equipment applied to medical treatment

Abstract

The invention relates to a superpixel-based MRI image segmentation method and MRI equipment, for processing heart MRI images. The MRI equipment comprises a main magnet system, a gradient magnetic field system, a radio frequency system and an operation and image processing system, wherein the radio frequency system comprises a plurality of radio frequency coils which are arranged in coil fixing devices respectively, and different radio frequency receiving coils correspond to different detection parts on a patient. The superpixel-based MRI image segmentation method comprises the following steps:1, preprocessing an input fused MRI image; 2, setting the number of initial superpixels; 3, according to the characteristics of the heart image, extracting super-pixel characteristics by adopting a measurement criterion for fusing gray scale, texture and adjacent relations, and combining the super-pixels of which the similarity meets a set condition; and 4, after the step 2 is continuously iterated, gradually expanding the initially segmented superpixel blocks until the similarity reaches a set value, i.e., the target area and the background area are not changed any more, iteration is stopped, and then obtaining a target image. According to the superpixel-based MRI image segmentation method, detail features such as contours and textures of the to-be-fused images can be fully mined, so that the superpixel-based MRI image segmentation method is more suitable for processing the singularity of the images, and the fused images with richer information amount, higher definition and better quality are obtained.

Description

technical field [0001] The invention relates to the technical field of MRI image acquisition and medical image processing, in particular to a method for segmenting cardiac MRI images and MRI equipment. Background technique [0002] The heart is one of the most important organs in the human body, and it is the hub of power and blood transport in the circulatory system. Cardiovascular disease (CVD), also known as circulatory system disease, is the response and manifestation of systemic vascular disease or systemic vascular disease in the heart. With the irregular life and rest of people, cardiovascular disease with the characteristics of high morbidity, disability and mortality has gradually come into people's sight, and has become the "number one killer" that endangers human health. [0003] With the rapid development of medical technology, the diagnosis of CVD is mainly based on medical imaging and various indicators of patients to make comprehensive judgments. Doctors use ...

AI Technical Summary

Problems solved by technology

[0004] In the prior art, segmentation methods based on improved registration accuracy can usually achieve better segmentation results. However, these precise registration methods usually require a long calculation time, and these methods often perform a large number of parameter settings for the target image, thereby reducing the Robustness of the algorithm

The method based on map selection can effectively reduce the computing resources required for automatic segmentation. However, these algorithms lose some useful information in the map during the map selection process, thus affecting the segmentation accuracy.

The methods based on image fusion can effectively use the local feature information of the target to be segmented. However, these methods cannot globally consider the difference between the atlas and the target image, which affects the segmentation accuracy.

Images