CN-122023204-A - Tumor radiotherapy auxiliary imaging image enhancement method and system

CN122023204ACN 122023204 ACN122023204 ACN 122023204ACN-122023204-A

Abstract

The embodiment of the application provides a method and a system for enhancing an auxiliary imaging image of tumor radiotherapy, which relate to the technical field of medical image processing, and the method comprises the steps of receiving medical image data to be processed; the method comprises the steps of computing and extracting a local area of medical image data to obtain feature marks, comparing the feature marks with a preset reference mode set to determine the category of the local area, wherein the category of the local area comprises physiological motion artifact, metal artifact or tumor edge, and executing corresponding image processing operation on the local area according to the category of the local area to output processed medical image data, wherein the image processing operation comprises deformation correction, image reconstruction, equalization processing, moderate smoothing and moderate enhancement. The application can improve the image quality of the auxiliary imaging of the tumor radiotherapy and the identification of tumor boundaries.

Inventors

SUN SHASHA
HUANG HAI
ZHU XINRONG
SUN GUOQIANG

Assignees

无锡睿影生物科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260203

Claims (10)

1. A method for enhancing an auxiliary imaging image of tumor radiotherapy, comprising: Receiving medical image data to be processed; calculating and extracting a local area of the medical image data to obtain a characteristic mark, wherein the characteristic mark comprises gray level change trend, texture pattern and frequency distribution; Comparing the characteristic marks with a preset reference mode set to determine the category of the local area, wherein the category of the local area comprises physiological motion artifact, metal artifact or tumor edge; And according to the category of the local area, performing corresponding image processing operation on the local area, and outputting processed medical image data, wherein the image processing operation comprises deformation correction, image reconstruction, equalization processing, moderate smoothing and moderate enhancement.
2. The method according to claim 1, wherein the computing the extraction of the local region of the medical image data to obtain a feature footprint comprises: calculating and extracting a local area of the medical image data by adopting a Sobel operator to obtain a gray level change trend; Calculating and extracting a local area of the medical image data by adopting a gray level co-occurrence matrix to obtain a texture pattern; And carrying out calculation and extraction on the local area of the medical image data by adopting two-dimensional Fourier transform to obtain frequency distribution.
3. The method of claim 1, wherein said comparing the signature with a set of preset reference patterns to determine the category of the local area comprises: and comparing the feature marks with a preset reference pattern set, evaluating similarity through a distance measure or a classifier, and determining the category of the local area.
4. The method of claim 1, wherein comparing the signature to a set of preset reference patterns to determine the category of the local region comprises: comparing the characteristic marks with a preset reference mode set to obtain a comparison result; When the comparison result represents that the gradient strength of the gray level change trend is lower than the preset strength in the motion direction and bilateral fuzzy edges appear in the vertical direction, determining the category of the local area as physiological motion artifact; when the comparison result represents that the texture pattern is a high-density stripe and the texture pattern is between smooth tissue and rough tissue, determining that the type of the local area is metal artifact; when the contrast result represents that the gray level change trend is that the definition is larger than the preset definition and the contrast is smaller than the preset contrast, determining that the category of the local area is tumor edge; When the comparison result represents that the frequency is distributed in the tumor and is a low-density uniform mark, determining that the type of the local area is a necrosis area; When the comparison results indicate that the frequency distribution is inside the tumor and is a high-density irregular imprint, the type of the local area is determined to be a hemorrhagic focus.
5. The method of claim 4, wherein performing a corresponding image processing operation on the local region according to the class of the local region comprises: Under the condition that the category of the local area is physiological motion artifact, analyzing the tiny displacement of the pixels in the local area to obtain a pixel motion vector; and carrying out deformation correction on the medical image data to be processed according to the pixel motion vector.
6. The method of claim 4, wherein performing a corresponding image processing operation on the local region according to the class of the local region, outputting processed medical image data, comprises: Under the condition that the category of the local area is metal artifact, interpolation filling is carried out on metal projections in the local area to obtain corrected projection values; and carrying out image reconstruction on the medical image data to be processed according to the corrected projection values.
7. The method of claim 4, wherein performing a corresponding image processing operation on the local region according to the class of the local region, outputting processed medical image data, comprises: calculating a gray level histogram of a local area of the tumor edge under the condition that the category of the local area is the tumor edge; and carrying out equalization processing on the medical image data to be processed according to the gray level histogram.
8. The method of claim 4, wherein performing a corresponding image processing operation on the local region according to the class of the local region, outputting processed medical image data, comprises: acquiring the gradient of the pixel points of the local area under the condition that the type of the local area is a necrosis area; And moderately smoothing the medical image data to be processed by adopting anisotropic diffusion filtering according to the gradient magnitude of the pixel points.
9. The method of claim 4, wherein performing a corresponding image processing operation on the local region according to the class of the local region, outputting processed medical image data, comprises: Subtracting the smooth version from the image of the local area to obtain an initial version under the condition that the type of the local area is a hemorrhagic focus; and moderately enhancing the medical image data to be processed by adopting non-sharpening masking according to the initial version.
10. A tumor radiation therapy assisted imaging image enhancement system, comprising: The receiving module is used for receiving medical image data to be processed; The computing and extracting module is used for computing and extracting the local area of the medical image data to obtain feature marks, wherein the feature marks comprise gray level change trend, texture patterns and frequency distribution; The determining module is used for comparing the characteristic marks with a preset reference mode set and determining the category of the local area, wherein the category of the local area comprises physiological motion artifact, metal artifact or tumor edge; And the output module is used for executing corresponding image processing operation on the local area according to the category of the local area and outputting processed medical image data, wherein the image processing operation comprises deformation correction, image reconstruction, equalization processing, moderate smoothing and moderate enhancement.

Description

Tumor radiotherapy auxiliary imaging image enhancement method and system Technical Field The application relates to the technical field of medical image processing, in particular to a tumor radiotherapy auxiliary imaging image enhancement method and system. Background In the related art, in the clinical practice of tumor radiotherapy, an auxiliary imaging technology is a key for accurately defining a treatment target area, and can provide visual information of tumor positions, sizes and relations with surrounding tissues for doctors. However, the imaging quality is affected by various factors, the tumor boundary identification difficulty is high, the boundary blurring is often caused by noise and insufficient contrast in the conventional technology, the demarcation accuracy of a target area is affected, and the radiation dose distribution deviation is caused. The initial planning of radiotherapy needs to rely on high-resolution CT to acquire a three-dimensional image, but because of equipment physical limitation, similar tissue density, scanning artifact and noise, tumors and healthy tissues are difficult to distinguish, the difficulty of a doctor in drawing a target area is increased, and deviation is easy to cause. The existing image enhancement system has obvious limitation, and when tumors are positioned in organs affected by physiological activities such as livers, motion artifacts can be enhanced by misjudgment, false textures are generated, and boundary recognition is interfered. When there are heterogeneous structures such as dead zone and hemorrhagic focus in the tumor, the existing method can not accurately distinguish active tumor cell areas, which affects the doctor to evaluate the tumor activity and formulate the dose distribution. In addition, when a metal bracket is arranged near the tumor, the metal artifact can be misjudged to be edge detail reinforcement, false high-contrast edges are generated, and the false high-contrast edges are confused with the real tumor boundary, distort the tumor morphology and seriously interfere with the correct identification of the anatomical structure. Disclosure of Invention The present application aims to solve at least one of the technical problems existing in the prior art. Therefore, the application provides a tumor radiotherapy auxiliary imaging image enhancement method and system, aiming at improving the image quality of tumor radiotherapy auxiliary imaging and the identification degree of tumor boundaries. In a first aspect, an embodiment of the present application provides a method for enhancing an auxiliary imaging image for tumor radiotherapy, including: Receiving medical image data to be processed; calculating and extracting a local area of the medical image data to obtain a characteristic mark, wherein the characteristic mark comprises gray level change trend, texture pattern and frequency distribution; Comparing the characteristic marks with a preset reference mode set to determine the category of the local area, wherein the category of the local area comprises physiological motion artifact, metal artifact or tumor edge; And according to the category of the local area, performing corresponding image processing operation on the local area, and outputting processed medical image data, wherein the image processing operation comprises deformation correction, image reconstruction, equalization processing, moderate smoothing and moderate enhancement. According to some embodiments of the application, the computing the local region of the medical image data to obtain the feature stamp includes: calculating and extracting a local area of the medical image data by adopting a Sobel operator to obtain a gray level change trend; Calculating and extracting a local area of the medical image data by adopting a gray level co-occurrence matrix to obtain a texture pattern; And carrying out calculation and extraction on the local area of the medical image data by adopting two-dimensional Fourier transform to obtain frequency distribution. According to some embodiments of the application, the comparing the feature stamp with a preset reference pattern set, determining the category of the local area includes: and comparing the feature marks with a preset reference pattern set, evaluating similarity through a distance measure or a classifier, and determining the category of the local area. According to some embodiments of the application, comparing the signature with a set of preset reference patterns, determining the category of the local area comprises: comparing the characteristic marks with a preset reference mode set to obtain a comparison result; When the comparison result represents that the gradient strength of the gray level change trend is lower than the preset strength in the motion direction and bilateral fuzzy edges appear in the vertical direction, determining the category of the local area as physiological motion artifact; when the comparison result rep