Search

CN-121999094-A - Method, system and application for removing CT image bed image

CN121999094ACN 121999094 ACN121999094 ACN 121999094ACN-121999094-A

Abstract

The invention relates to a method, a system and an application for removing CT image bed images, which sequentially comprise the following steps of obtaining a continuous CT tomographic image sequence of a patient, constructing three-dimensional volume data, carrying out projection analysis on the three-dimensional volume data along the z-axis direction, respectively calculating to obtain a maximum projection image and a minimum projection image, carrying out first threshold segmentation on the basis of the minimum projection image to generate a first mask, carrying out second threshold segmentation on the basis of the maximum projection image to generate a second mask, and generating an accurate bed region mask which does not contain human tissues on the basis of the first mask and the second mask. The invention overcomes the segmentation ambiguity of a single threshold method in an overlapped CT value interval by fusing the maximum value projection (reflecting the upper limit of the tissue) and the minimum value projection (reflecting the characteristics of the bed and realizes the accurate separation of the bed and the human tissue.

Inventors

  • HA DA
  • LV GUOZHENG

Assignees

  • 影诺高新科技(苏州)有限公司

Dates

Publication Date
20260508
Application Date
20260123

Claims (10)

  1. 1. A method for removing CT image bed images is characterized by comprising the following steps in sequence: Step 1, acquiring a continuous CT tomographic image sequence of a patient, and constructing three-dimensional volume data V (x, y, z), wherein an x-axis and a y-axis represent pixel coordinates, and a z-axis is the head-foot direction of the patient; Step 2, carrying out projection analysis on the three-dimensional data along the z-axis direction, and respectively calculating to obtain a maximum projection image MIP (x, y) and a minimum projection image MinIP (x, y); Step 3, performing first threshold segmentation based on the minimum projection image, and generating a first mask which aims at completely covering the scanning bed area and allowing to contain first type of interference tissues; Step 4, performing second threshold segmentation based on the maximum projection image, and generating a second mask, wherein the second mask aims at completely covering a human tissue region; Step 5, generating an accurate bed area mask which does not contain human tissues through a series of preset morphological operations and logic operations based on the first mask and the second mask; And 6, expanding the accurate bed region mask to a three-dimensional space, and removing or transparentizing corresponding voxels from the original three-dimensional volume data for subsequent three-dimensional visual rendering.
  2. 2. The method for removing CT image table image as described in claim 1, wherein said step 5 comprises the steps of: step 51, performing subtraction operation on the second mask and the first mask to generate a third mask; Step 52, filling air outside the image boundary of the human tissue region in the third mask to obtain a fourth mask after filling; And 53, performing arithmetic logic operation on the first mask and the fourth mask, and deducting the determined non-bed area represented by the fourth mask from the first mask so as to generate the accurate bed area mask.
  3. 3. The method of claim 2, further comprising performing at least one dilation-erosion operation on the generated third mask in step 51 to eliminate fine lines caused by mask alignment problems.
  4. 4. A method for removing CT image bed images as recited in claim 2, wherein the filling operation in step 52 is performed by performing a reverse filling operation starting from the frame of the image and filling all background areas connected to the frame with foreground.
  5. 5. The method for removing CT images from a bed as claimed in claim 1, wherein when the CT value of the bed is lower than the CT value of the skin of the human body, the step 5 sequentially comprises the steps of: step 51, performing hole filling operation on a discontinuous region inside a human tissue region in the second mask to obtain a filled third mask; and 52, performing logical NAND operation on the first mask and the third mask, and removing the determined human tissue region represented by the third mask from the first mask so as to generate the accurate bed region mask.
  6. 6. The method of removing CT image bed images as recited in claim 5, wherein the filling operation in the step 51 is performed by using the foreground as a seed region, starting the back filling inside the human tissue, and filling all the background regions connected to the frame as the foreground.
  7. 7. A method of removing CT image table images as defined in claim 1, wherein the second threshold value used for the second threshold segmentation is higher than the first threshold value.
  8. 8. A system for removing CT image table images, comprising: A memory for storing a computer program; A processor for executing the computer program to implement a method for removing CT image table images according to any one of claims 1 to 7.
  9. 9. A computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor performs the steps of a method of removing CT image table images according to any of claims 1 to 7.
  10. 10. Use of a method for removing CT image table images according to any of the claims 1-7 in CT three-dimensional visualization.

Description

Method, system and application for removing CT image bed image Technical Field The invention relates to the technical field of medical image processing, in particular to a method, a system and application for removing CT image bed images. Background In CT examinations, a patient is usually lying on a scanning bed for data acquisition. When three-dimensional volume rendering (Volume Rendering Technique, VRT) reconstruction is performed, the bed (typically made of high-strength, low-density material such as carbon fiber) will appear as a solid structure in the image, severely interfering with the doctor's view of the patient anatomy, especially the 3D image of the abdomen, pelvis, lower extremities, etc. Currently, common methods of bed removal rely primarily on simple global threshold segmentation. Because the CT value (Hunter unit, HU) range of the bed material is overlapped with the soft tissue (such as muscle and viscera) and liquid of the human body part, the single threshold method is difficult to distinguish accurately. If the threshold value is too high, the bed structure is easily kept or removed by mistake, and if the threshold value is too low, the body tissues of the patient can be eroded or removed by mistake, so that the 3D model is incomplete, and the diagnosis accuracy is affected. In addition, some methods based on region growing or edge detection also tend to cause segmentation leakage or shortfall due to the blurring of the boundary of the bed with the human body contact region and the irregular shape of the bed in the cross section (XY plane). Thus, there is a need for a method that can robustly, automatically and accurately separate a bed from human tissue. In view of the above-mentioned drawbacks, the present inventors have actively studied and innovated to create a method, system and application for removing CT image table images, which makes the method and system have more industrial utility value. Disclosure of Invention In order to solve the technical problems, the invention aims to provide a method, a system and an application for removing CT image bed images. The invention relates to the technical field of medical image processing, in particular to a three-dimensional visualization preprocessing method for Computed Tomography (CT) images, and specifically relates to a method for automatically and accurately dividing and removing scanning bed images based on image morphological projection and logic operation. In order to achieve the above purpose, the invention adopts the following technical scheme: One of the objects of the present invention is: a method for removing CT image bed images sequentially comprises the following steps: Step 1, acquiring a continuous CT tomographic image sequence of a patient, and constructing three-dimensional volume data V (x, y, z), wherein an x-axis and a y-axis represent pixel coordinates, and a z-axis is the head-foot direction of the patient; Step 2, carrying out projection analysis on the three-dimensional data along the z-axis direction, and respectively calculating to obtain a maximum projection image MIP (x, y) and a minimum projection image MinIP (x, y); Step 3, performing first threshold segmentation based on the minimum projection image to generate a first mask, wherein the first mask aims to completely cover the scanning bed area and allow the first type of interference tissues (namely 'partial non-bed tissue structures' in the first mask); Step 4, performing second threshold segmentation based on the maximum projection image to generate a second mask, wherein the second mask aims to completely cover the human tissue region; step 5, generating an accurate bed area mask which does not contain human tissues through a series of preset morphological operations and logic operations based on the first mask and the second mask; And 6, expanding the accurate bed region mask to a three-dimensional space, and removing or transparentizing corresponding voxels from the original three-dimensional volume data for subsequent three-dimensional visual rendering. As a further improvement of the present invention, step 5 comprises the following steps in order: step 51, performing subtraction operation on the second mask and the first mask to generate a third mask; step 52, filling air outside the image boundary of the human tissue region in the third mask to obtain a fourth mask after filling; and 53, performing arithmetic logic operation on the first mask and the fourth mask, and subtracting the determined non-bed area represented by the fourth mask from the first mask so as to generate an accurate bed area mask. As a further improvement of the present invention, at least one expansion etching operation is performed on the generated third mask in step 51 to eliminate fine lines caused by mask alignment problems. As a further improvement of the present invention, the filling operation in step 52 is specifically to perform a reverse filling from the frame of t