Three-dimensional medical image segmentation method

Abstract

The invention discloses a three-dimensional medical image segmentation method. The method comprises the following steps: firstly performing three-dimensional superposition reconstruction on sequential two-dimensional medical image, and then performing Gaussian filter smoothing in a three-dimensional space; carrying out three-dimensional space contraction computation on the three-dimensional medical image, and calculating and recording the contraction orders and the energy characteristics of every three-dimensional pixel point; performing three-dimensional space extension on the three-dimensional medical image in a plurality of areas by a level set method; and finally extracting and segmenting the three-dimensional pixel points in the three-dimensional image according to information of interested areas, and respectively generating a plurality of three-dimensional images of the interested areas. The method has positive effects of strong timeliness, high computing efficiency, high segmentation accuracy, and clear and reliable image local features, and can achieve simultaneous segmentation of a plurality of areas.

Description

technical field [0001] The invention relates to a method for segmenting a three-dimensional medical image, in particular to a method for segmenting a three-dimensional medical image based on a multi-region contraction-expansion collision strategy. Background technique [0002] In the medical field, 3D medical image segmentation is used to segment the region of interest or lesion in the 3D medical image, to observe and analyze the morphology, characteristics and other pathological conditions of the region of interest or lesion, and to conduct 3D medical Image reconstruction and fusion, etc. Especially for medical images of the heart, it is necessary to segment and extract multiple regions of the medical image of the heart at the same time, such as the segmentation of left and right atria or left and right ventricles. [0003] Generally speaking, most medical image segmentation methods are basically based on CT / MRI sequence images to perform image segmentation in two-dimensio...

AI Technical Summary

Problems solved by technology

[0005] 1) It is difficult to determine the precise threshold value, and the difference of medical images leads to the difference of threshold value;

[0006] 2) In the case that the gray scales of multiple regions of the image are roughly the same, the characteristics of different regions cannot be accurately distinguished, and the segmentation effect cannot be achieved;

[0007] 3) In the case of obvious differences in the gray scales of multiple regions of the image, the single gray threshold can reduce the discrimination effect of different regions, and even damage the image quality of the segmented region

[0009] 1) The region growing method can only be segmented for a single region or multiple regions without connectivity

In the case of multi-regions with certain connectivity, the region growing method cannot distinguish the connected edges of multi-regions, and it cannot achieve the purpose of multi-region segmentation;

[0010] 2) The region growing method will cause inaccurate image segmentation and extraction when the image gray level is uneven and the difference is large.

[0012] 1) The calculation of energy characteristics is easy to fall into local minimization;

[0013] 2) The segmentation of the image region of interest with large shape changes is not ideal, and there is an over-segmentation phenomenon;

[0014] 3) When the shape of the region of interest in the sequence image is separated or merged, it cannot be accurately tracked, segmented and extracted

[0022] 1) Most image segmentation methods can only segment a single area in the image, but cannot segment multiple areas in the image at the same time, and cannot satisfy the extraction and segmentation of multiple areas of cardiac medical images;

[0023] 2) The discriminative performance for the edges of multiple regions with certain connectivity is weak, and it is prone to over-segmentation or under-segmentation between multiple regions;

For the region separation or merging of sequential medical images, most of the existing technologies cannot track the detailed changes of the image region morphology;

[0025] 4) The regional morphology, pathology and other information of cardiac medical images have great differences due to individual differences. Most of the existing image segmentation techniques have certain limitations to varying degrees. When there are large differences in medical images, Significant decline in practicality and adaptability;

[0026] 5) Although some existing image segmentation techniques and methods have good segmentation effects, they have the limitations of large amount of computing data, complex data structure or low computing efficiency;

[0027] 6) There are some image segmentation techniques that are not suitable for the segmentation of sequential cardiac medical images in three-dimensional space

Images