Prostate magnetic resonance image segmentation method based on level set

Abstract

The invention provides a prostate magnetic resonance image segmentation method based on a level set and relates to the technical field of magnetic resonance image segmentation. Based on a prostate magnetic resonance image, the invention provides a prostate magnetic resonance image segmentation method based on edge distance adjustment level set evolution for the comprehensive segmentation of the internal and external contours of the prostate. Based on the establishment of a level set evolution equation, the method achieves the segmentation of the external contour based on a prostate magnetic resonance longitudinal relaxation time image, and achieves the segmentation of the internal contour based on a prostate magnetic resonance transverse relaxation time image under an external contour constraint condition, thereby further completing the full effective segmentation of the internal and external contours of the prostate. The method can achieve the comprehensive segmentation of the internal and external contours of the prostate, is very close to the ideal result of manual segmentation of a clinical expert, and has a good reference value for the clinical diagnosis and treatment of prostate diseases.

Description

technical field [0001] The invention relates to a method for segmenting a prostate magnetic resonance image based on a level set, and belongs to the technical field of magnetic resonance image segmentation. Background technique [0002] With population growth and changes in living habits, the incidence and mortality of prostate cancer have shown a clear upward trend in recent years. Clinical experience shows that prostate cancer has a high survival rate if it is detected early and treated in time. Therefore, research on the diagnosis and treatment of prostate cancer is of great significance. As one of the important means of diagnosis and treatment of prostate cancer, medical imaging is playing an increasingly important role. Magnetic resonance imaging is considered to be the best medical imaging for the diagnosis and adjuvant treatment of prostate cancer because of its high resolution of soft tissue, multi-parameter imaging, and ability to scan arbitrary slices. [0003] I...

AI Technical Summary

Problems solved by technology

Both of these methods first realize the outer contour segmentation manually, and use it as the initialization of the internal segmentation, which makes the full segmentation of the prostate time-consuming and laborious.

In 2013, Toth and others in the United States used the method of multiple coupling level set active contour model to perform full segmentation of inner and outer contours. The segmentation effect of this method depends more on the quality of the segmented image, and segmenting an image requires training a large number of atlases. , takes a long time

Due to the noise in the magnetic resonance image of the inner prostate area, the gray scale display is uneven, and the boundary of the area is blurred, it has always been a challenge to study the full segmentation of the inner and outer contours of the prostate based on magnetic resonance images.

Images