CN111462150A

CN111462150A - Method, system and medium for segmenting uterine fibroids in nuclear magnetic images

Info

Publication number: CN111462150A
Application number: CN202010177452.2A
Authority: CN
Inventors: 王鑫; 朱威桢; 王文波
Original assignee: Zhonghui Medical Technology Shanghai Co ltd
Current assignee: Zhonghui Medical Technology Shanghai Co ltd
Priority date: 2020-02-21
Filing date: 2020-03-13
Publication date: 2020-07-28
Anticipated expiration: 2040-03-13
Also published as: CN111462150B

Abstract

The invention provides a method, a system and a medium for segmenting hysteromyoma of a nuclear magnetic image, which comprise the following steps: step 1: taking MRI as an initial image for segmentation, and drawing an initial hysteromyoma contour; step 2: segmenting the uterine fibroid by using an RSF model according to the initial uterine fibroid contour to obtain the uterine fibroid contour; and step 3: and judging the segmented hysteromyoma contour. The invention solves the problem of easy over-segmentation when the RSF model is used alone by combining the algorithm of the RSF model of the Hessian matrix, and improves the accuracy of segmentation; the algorithm of the invention only needs a doctor to provide an initial hysteromyoma contour, so that the manual intervention is less and the time of the doctor is saved.

Description

Method, system and medium for segmenting uterine fibroids in nuclear magnetic images

Technical Field

The invention relates to the technical field of medical imaging, in particular to a method, a system and a medium for segmenting hysteromyoma of a nuclear magnetic image. In particular to a method for segmenting the uterine fibroid of a nuclear magnetic image by using a Hessian matrix and an RSF model.

Background

Currently, there are some MRI uterine fibroid segmentation methods based on level set algorithms, which use level sets to segment uterine fibroids with a given initial contour. However, when the level set algorithm segments the uterine fibroids, the segmentation problem is easily caused, so that the segmentation result has a large error, and the desired result cannot be obtained.

Patent document CN109741441A (application number: 201811558314.8) discloses a three-dimensional uterine fibroid model reconstruction method, which includes the steps of defining regions: according to the voxel value of the nuclear magnetic image, dividing a region of a specific voxel value range, and recording the region as an myoma region; interference removing step: selecting a point of the myoma area, and removing an interference area outside the myoma area by adopting area growth; and (3) myoma area correction: defining a revised myoma zone with an erasure padding or a scope; an image conversion step: extracting a nuclear magnetic image of the myoma area, recording the nuclear magnetic image as an extracted nuclear magnetic image, and performing image binarization on the extracted nuclear magnetic image to obtain a converted image; three-dimensional reproduction: and performing surface reconstruction on the converted image to obtain a three-dimensional myoma model of grid data consisting of triangular surfaces.

Disclosure of Invention

In view of the deficiencies in the prior art, it is an object of the present invention to provide a method, system and medium for segmenting uterine fibroids in nuclear magnetic images.

The method for segmenting the uterine fibroid of the nuclear magnetic image comprises the following steps:

step 1: taking MRI as an initial image for segmentation, and drawing an initial hysteromyoma contour;

step 2: segmenting the uterine fibroid by using an RSF model according to the initial uterine fibroid contour to obtain the uterine fibroid contour;

and step 3: and judging the segmented hysteromyoma contour.

Preferably, the step 2 includes: segmenting the initial hysteromyoma contour to obtain a first hysteromyoma contour;

performing enhancement treatment on the first hysteromyoma contour to obtain a second hysteromyoma contour;

and intersecting the first uterine fibroid contour with the second uterine fibroid contour to obtain a third uterine fibroid contour, and dividing the uterine fibroid as an initial contour of the RSF model to obtain a uterine fibroid contour.

Preferably, enhancement processing is performed on the uterine fibroid contour in the previous image and the next image of the initial image according to a Hessian matrix.

Preferably, the step 3 comprises: and when the average gray scale of the pixels in the segmented hysteromyoma contour is smaller than a set value, the area of the segmented hysteromyoma contour is smaller than a set value, or the total layer thickness of the segmented image is larger than a set value, the segmentation is judged to be finished.

The system for segmenting the uterine fibroid of the nuclear magnetic image provided by the invention comprises the following components:

module M1: taking MRI as an initial image for segmentation, and drawing an initial hysteromyoma contour;

module M2: segmenting the uterine fibroid by using an RSF model according to the initial uterine fibroid contour to obtain the uterine fibroid contour;

module M3: and judging the segmented hysteromyoma contour.

Preferably, said module M2 comprises: segmenting the initial hysteromyoma contour to obtain a first hysteromyoma contour;

Preferably, said module M3 comprises: and when the average gray scale of the pixels in the segmented hysteromyoma contour is smaller than a set value, the area of the segmented hysteromyoma contour is smaller than a set value, or the total layer thickness of the segmented image is larger than a set value, the segmentation is judged to be finished.

Compared with the prior art, the invention has the following beneficial effects:

1. the invention solves the problem of easy over-segmentation when the RSF model is used alone by combining the algorithm of the RSF model of the Hessian matrix, and improves the accuracy of segmentation;

2. the algorithm of the invention only needs a doctor to provide an initial hysteromyoma contour, so that the manual intervention is less and the time of the doctor is saved.

Drawings

Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:

FIG. 1 is a flow chart of two-dimensional initial image segmentation;

FIG. 2 is a flow chart of three-dimensional image segmentation;

FIG. 3 is a graph of the segmentation results;

fig. 4 is a graph of the segmentation results.

Detailed Description

The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.

Referring to fig. 1 and fig. 2, the method of the present invention is a flowchart, including:

the method comprises the following steps: a doctor selects an MR image as an initial image for segmentation, and the outline of the uterine fibroid in the image is drawn;

step two: the RSF model utilizes the initial hysteromyoma contour drawn by a doctor to segment the hysteromyoma in the MRI to obtain a hysteromyoma contour;

step three: enhancing uterine fibroids in the previous image and the next image of the initial image by using a Hessian matrix;

step four: intersecting the enhanced image obtained in the third step with the hysteromyoma contour obtained in the second step, and taking the obtained contour as an initial contour of the RSF model to segment the two images to obtain the hysteromyoma contour in the two images;

step five: by analogy, the uterine fibroid contour combined by the uterine fibroid contour of the image enhanced by the Hessian matrix and the previous image (located after the initial image) or the next image (located before the initial image) is respectively used as the initial contour of the RSF model to segment the uterine fibroid contour;

step six: when the average gray scale of the pixels in the segmented hysteromyoma contour is smaller than the set value, or the area of the segmented contour is smaller than the set value, or the total layer thickness of the segmented image is larger than the set value, the segmentation is finished, as shown in fig. 3 and 4, which are graphs of the segmentation results.

In fig. 2, the three-dimensional segmentation starts from the initial image and is performed on both sides, which is a segmentation process on one side, and the other side is the same as the process. And when the segmented image is an image before and after the initial image is adjacent, the reference outline is used for taking the segmentation result of the initial image, otherwise, the segmentation result of the RSF model of the previous image is used.

Those skilled in the art will appreciate that, in addition to implementing the systems, apparatus, and various modules thereof provided by the present invention in purely computer readable program code, the same procedures can be implemented entirely by logically programming method steps such that the systems, apparatus, and various modules thereof are provided in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system, the device and the modules thereof provided by the present invention can be considered as a hardware component, and the modules included in the system, the device and the modules thereof for implementing various programs can also be considered as structures in the hardware component; modules for performing various functions may also be considered to be both software programs for performing the methods and structures within hardware components.

The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.

Claims

1. A method of segmenting a uterine fibroid in a nuclear magnetic image, comprising:

and step 3: and judging the segmented hysteromyoma contour.

2. The method for segmenting uterine fibroids according to claim 1, wherein said step 2 comprises: segmenting the initial hysteromyoma contour to obtain a first hysteromyoma contour;

3. The method for segmenting uterine fibroids according to claim 2, wherein enhancement processing is performed on uterine fibroid contours in a previous image and a next image of the initial image according to a Hessian matrix.

4. The method for segmenting uterine fibroids according to claim 1, wherein said step 3 comprises: and when the average gray scale of the pixels in the segmented hysteromyoma contour is smaller than a set value, the area of the segmented hysteromyoma contour is smaller than a set value, or the total layer thickness of the segmented image is larger than a set value, the segmentation is judged to be finished.

5. A system for segmenting uterine fibroids in a nuclear magnetic image, comprising:

module M3: and judging the segmented hysteromyoma contour.

6. The system for segmenting uterine fibroids according to claim 5, characterized in that said module M2 comprises: segmenting the initial hysteromyoma contour to obtain a first hysteromyoma contour;

7. The system for segmenting uterine fibroids according to claim 6, wherein enhancement processing is performed on uterine fibroid contours in a previous image and a next image of the initial image according to a Hessian matrix.

8. The system for segmenting uterine fibroids according to claim 5, characterized in that said module M3 comprises: and when the average gray scale of the pixels in the segmented hysteromyoma contour is smaller than a set value, the area of the segmented hysteromyoma contour is smaller than a set value, or the total layer thickness of the segmented image is larger than a set value, the segmentation is judged to be finished.

9. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 4.