Brain tumor MR image segmentation method based on local-global adaptive information learning

Abstract

The invention relates to a brain tumor MR image segmentation method based on local-global adaptive information learning. According to the method, image information in a brain tumor MR image is extracted, and a tumor in the MR image is segmented into a tumor overall area, a tumor core area and an enhanced tumor core area by adopting a semi-supervised learning method. The method comprises the following steps: firstly, extracting features from four modals by using a method of combining a spatial domain and a frequency domain to obtain enhanced features for expressing brain structure information; the features extracted from the four modes are fused to obtain a final fused feature; then efficient feature selection is carried out; and finally, segmenting the brain tumor MR image into a tumor overall region, a tumor core region and an enhanced tumor core region. Brain tumor MR image segmentation is a difficulty in brain tumor MR image segmentation, and the method of the invention reduces the workload of manual marking while ensuring the segmentation precision, thereby improving the working efficiency.

Description

technical field [0001] The invention belongs to the field of image processing, in particular to a brain tumor MR image segmentation method based on local-global adaptive information learning. Background technique [0002] Brain tumor MR image segmentation technology is a process of extracting features from the image, and then segmenting each brain tumor structure in the image based on the features. The main basis of the brain tumor MR image segmentation technology is to use the distribution of the texture structure inside the tumor in the brain tumor MR image, and then segment the tumor into the overall tumor, tumor core and enhanced tumor core according to the different textures of different tumor types. Then output relevant indicators for research and analysis. [0003] Brain tumor MR image segmentation technology is one of the most practical technologies in medical image processing technology. Due to the increasing number of brain tumor MR images in the database, higher...

AI Technical Summary

Problems solved by technology

[0004] For brain tumor MR image segmentation, the disadvantages of traditional MR image segmentation methods are: because MR images are easily affected by artifacts during the imaging process, resulting in poor imaging quality

And there is no clear difference in the texture features inside the tumor, so the segmentation effect on the tumor type is poor

Due to the huge amount of data in brain MR sequence images, manual segmentation is not only time-consuming but also inefficient, and the results of manual segmentation are affected by the professional knowledge and operational proficiency of doctors, which may produce widely different results

Images