Method for evaluating human tissue calcification based on CT image
By defining and calculating calcification foci individually, the bias problem in calcification assessment in existing technologies has been solved, resulting in more accurate calcification assessment that is applicable to CT image evaluation of human tissue calcification.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WEST CHINA HOSPITAL SICHUAN UNIV
- Filing Date
- 2026-01-20
- Publication Date
- 2026-05-01
AI Technical Summary
In existing technologies, the screening of calcifications often uses the default CT value standard, which does not take into account individual differences, resulting in an unscientific and comprehensive assessment. Furthermore, the calculation of calcification area, volume, and mass may be biased.
A personalized definition of calcification lesions is adopted. By setting CT values and continuous area lower limits, the calcification score is calculated by combining pixel point screening, mean area method, layered accumulation method and CT value summation method, and then correction is performed to calculate the total calcification area, volume, mass and Agatston score.
It improves the accuracy of calcification assessment, can more accurately reflect tissue characteristics and individual differences, capture subtle changes, and contribute to more accurate disease diagnosis.
Smart Images
Figure CN121962092A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of calcification assessment technology, and more particularly to a method for assessing human tissue calcification based on CT images. Background Technology
[0002] CT, or computed tomography, is an advanced medical imaging technique. It can non-destructively probe the fine internal structures of objects, offering advantages unmatched by other non-destructive testing equipment. Calcification refers to the necrosis of certain tissues in the human body under the influence of certain factors, followed by the deposition of calcium salts within the necrotic lesion, thus localizing and stabilizing the lesion. On CT images, calcification typically appears as a high-density shadow, with a CT value generally greater than 100 HU, and sometimes even higher.
[0003] Current technologies for assessing calcification in biological tissues are mostly limited to the human coronary arteries, aortic arch, ascending aorta, and descending aorta. Two main methods are employed: 1. The sum of the calcified area, volume, or mass in multi-slice CT images of the observed tissue; 2. The Agatston score, commonly used for coronary artery calcification assessment, which calculates the sum of the products of the calcification intensity grade and the calcification area at each slice.
[0004] Existing calcification assessment methods include two main approaches: the first considers only the area of calcification, neglecting different calcification intensities, resulting in an insufficiently scientific and comprehensive assessment of calcification. The second approach considers differences in calcification intensity, but uses the highest CT value of pixels within the calcification lesion for intensity grading. This highest CT value only represents the maximum calcification intensity achievable by pixels within that calcification lesion, not the true calcification intensity of all pixels within the entire calcification lesion. Common problems with these methods include: the screening of calcification lesions often relies on software that uses a default definition of calcification lesions (e.g., CT value > 130 HU) for initial screening and display of pixels; this CT value standard may be inapplicable if the calcification is not vascular. Furthermore, the calculation of the sum of area, volume, and mass at each slice, or the calcification score, does not take into account the individual differences in the number of CT scan slices for a particular human tissue, potentially leading to bias. Therefore, this paper proposes a method for assessing human tissue calcification based on CT images. Summary of the Invention
[0005] The purpose of this invention is to address the problems existing in the prior art, such as the screening of calcifications often relying on software to initially screen and display pixels according to the software's default definition of calcifications (e.g., CT value > 130 HU). If it is not vascular calcification, this CT value standard may not be applicable. Furthermore, the calculation of the sum of area, volume, and mass of each layer or the calcification score does not take into account the differences in the number of CT scan layers of a certain human tissue between individuals, which may cause bias. Therefore, this invention proposes a method for evaluating human tissue calcification based on CT images.
[0006] To achieve the above objectives, the present invention adopts the following technical solution: A method for assessing human tissue calcification based on CT images includes the following steps: S1: Personalized definition of calcifications: Determine the observation range on CT images based on the tissue to be evaluated, and set the CT value and lower limit of the continuous area according to the tissue characteristics; S2: Pixel Filtering: Filter calcification pixels according to personalized definitions; S3: Calculate the calcification score: Calculate the calcification score based on the selected calcification pixels using appropriate methods, including the mean area method, the layered accumulation method, and the CT value summation method. S4: Calcification fraction correction: Correct the calcification fraction based on the number of CT slices; S5: Other related calculations: Calculate the total calcification area, total volume, total mass, and Agatston score.
[0007] The above further includes: Furthermore, in S1, the tissue characteristics include the tissue density, composition, and structure, as well as the characteristics of the lesion. The tissue density, composition, and structure determine its grayscale representation on the CT image, thereby determining the CT value. The characteristics of the lesion determine the setting of the lower limit of the continuous area.
[0008] Furthermore, in S2, the pixel filtering is performed on the CT image based on the CT value and the lower limit of the continuous area set in S1.
[0009] Furthermore, in S2, the specific steps for pixel selection are as follows: Image loading and preprocessing: Preprocessing operations are performed on the CT image files to be processed, including noise removal (through filtering algorithms such as Gaussian filtering, median filtering, etc.) and image enhancement (contrast enhancement, histogram equalization, etc.) to improve image quality and increase the visibility and detection accuracy of calcifications. Set screening criteria: Based on the density, composition and structure of the tissue, set a CT value threshold. The CT value threshold is used to distinguish calcified tissue from other surrounding tissues. Set a continuous area lower limit based on the characteristics of the lesion. The continuous area lower limit is used to exclude isolated noise points or artifacts and ensure that the detected calcifications have a certain area size. Automatic pixel scanning: Automatically traverses every pixel in the image and records the coordinates of the currently traversed pixel; Threshold judgment: Read the CT value of the current pixel and compare it with the CT value set in S1. If the CT value of the current pixel exceeds the set threshold, the pixel is considered to be part of the calcification foci and the area is judged; otherwise, continue to traverse the next pixel. Area determination: For pixels exceeding the CT value threshold, check the CT values of their adjacent pixels (usually pixels in the top, bottom, left, and right directions). If the CT values of adjacent pixels also exceed the set threshold, add them to the current calcification area and continue to expand outward until no more adjacent pixels meet the condition. Calculate the area of the formed continuous region and compare it with the lower limit of the continuous area set in S1. If the area of the continuous region exceeds the set lower limit, the region is considered a calcification; otherwise, it is regarded as noise or artifact and is not retained. Marking and visualization involves marking the detected calcification areas on the image. Different colors or outlines can usually be used to distinguish calcifications from other tissues. As needed, a report containing information such as the location, size, and CT value of the calcifications can be generated and may be displayed in the form of images or tables.
[0010] Furthermore, in the automatic pixel scanning, the image width (Width) and height (Height) are set, along with a matrix or data structure for storing the results (e.g., a two-dimensional array for marking calcifications). Two nested loops are used: the outer loop iterates through each row (Height) of the image, and the inner loop iterates through each column (Width). Within the loop, the coordinates of the currently traversed pixel can be determined by its row index. and column indexes express.
[0011] Furthermore, in S3, the mean area method assesses the degree of calcification by comprehensively considering the area and density of the calcification foci. The mean area method calculates the total area and the mean CT value of the selected pixels that meet the definition of calcification foci. ,in, S is the average CT value of the pixels in S2 that meet the definition of calcification, where S is the total area occupied by the pixels selected in S2. for , This represents the area corresponding to a single pixel.
[0012] Furthermore, in S3, the layered accumulation method reflects the distribution of calcifications in three-dimensional space in more detail:
[0013]
[0014] in, This represents the total number of slices in the CT scan. For the first The contribution of calcification fraction in the layer, For the first The area occupied by pixels in the layer that conform to the definition of calcification. For the first The average CT value of these pixels in the layer. The area corresponding to a single pixel. The number of pixels in the i-th layer that meet the definition of a calcification foci.
[0015] Furthermore, in S3, the CT value summation method is applicable to cases emphasizing calcification density: ,in, This is the sum of CT values of pixels in S2 that meet the definition of calcification. For pixel spacing, This represents the area corresponding to a single pixel.
[0016] Furthermore, in S4, the calcification fraction calculated by any of the methods in S3 is corrected:
[0017] in, It represents the total number of slices in a CT scan.
[0018] Furthermore, in S5, the total calcification area refers to the total area occupied by all calcification foci on the two-dimensional image:
[0019] in, Yes, yes, the first The area of each calcification foci This represents the total number of calcifications. The total calcification volume refers to the total volume occupied by all calcification foci in three-dimensional space. or
[0020] in, The calcifications are spherical. It is the first The radius of each calcification (assuming it is spherical). It is its area on a two-dimensional slice. It is the average thickness of calcifications on the slice or the distance between adjacent slices; The total mass of calcification refers to the total mass of all calcification foci:
[0021] in, It is the density of the calcified material; The Agatston score is a method for assessing coronary artery calcification only.
[0022] in, It is the first The area of each calcification foci The weights are determined based on the CT values of the calcifications. Calcifications with CT values in the range of 130HU-199HU are assigned a weight of 1, those with CT values in the range of 200HU-299HU are assigned a weight of 2, those with CT values in the range of 300HU-399HU are assigned a weight of 3, and those with CT values ≥400HU are assigned a weight of 4.
[0023] The present invention has the following beneficial effects: In this invention, by setting the CT value threshold and continuous area lower limit of calcification foci in a personalized manner, different tissue characteristics and individual differences can be reflected more accurately, thereby improving the accuracy of calcification assessment. Compared with the traditional unified standard assessment method, this method can better capture subtle calcification changes and helps to diagnose diseases more accurately. Attached Figure Description
[0024] Figure 1 This diagram illustrates the steps of the method for assessing human tissue calcification based on CT images proposed in this invention. Detailed Implementation
[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0026] Please see Figure 1 As shown, this invention is a method for assessing calcification in human tissue based on CT images, comprising the following steps: S1: Personalized definition of calcifications: Determine the observation range on CT images based on the tissue to be evaluated, and set the CT value and lower limit of the continuous area according to the tissue characteristics; S2: Pixel Filtering: Filter calcification pixels according to personalized definitions; S3: Calculate the calcification score: Calculate the calcification score based on the selected calcification pixels using appropriate methods, including the mean area method, the layered accumulation method, and the CT value summation method. S4: Calcification fraction correction: Correct the calcification fraction based on the number of CT slices; S5: Other related calculations: Calculate the total calcification area, total volume, total mass, and Agatston score.
[0027] In one embodiment, for the above S1, the tissue characteristics include the density, composition and structure of the tissue and the characteristics of the lesion. The density, composition and structure of the tissue determine its grayscale representation on the CT image, and thus determine the CT value. The characteristics of the lesion determine the setting of the lower limit of the continuous area.
[0028] In one embodiment, for the above S2, pixel filtering is performed by filtering CT images based on the CT value and the lower limit of the continuous area set in S1.
[0029] In one embodiment, for the above S2, the specific steps of pixel selection in S2 are as follows: Image loading and preprocessing: Preprocessing operations are performed on the CT image files to be processed, including noise removal (through filtering algorithms such as Gaussian filtering, median filtering, etc.) and image enhancement (contrast enhancement, histogram equalization, etc.) to improve image quality and increase the visibility and detection accuracy of calcifications. Set screening criteria: Based on the density, composition and structure of the tissue, set the CT value threshold. The CT value threshold is used to distinguish calcified tissue from other surrounding tissues. Set the lower limit of continuous area based on the characteristics of the lesion. The lower limit of continuous area is used to exclude isolated noise points or artifacts and ensure that the detected calcifications have a certain area size. Automatic pixel scanning: Automatically traverses every pixel in the image and records the coordinates of the currently traversed pixel; Threshold judgment: Read the CT value of the current pixel and compare it with the CT value set in S1. If the CT value of the current pixel exceeds the set threshold, the pixel is considered to be part of the calcification foci and the area is judged; otherwise, continue to traverse the next pixel. Area determination: For pixels exceeding the CT value threshold, check the CT values of their adjacent pixels (usually pixels in the top, bottom, left, and right directions). If the CT values of adjacent pixels also exceed the set threshold, add them to the current calcification area and continue to expand outward until no more adjacent pixels meet the condition. Calculate the area of the formed continuous region and compare it with the lower limit of the continuous area set in S1. If the area of the continuous region exceeds the set lower limit, the region is considered a calcification; otherwise, it is regarded as noise or artifact and is not retained. Marking and visualization involves marking the detected calcification areas on the image. Different colors or outlines can usually be used to distinguish calcifications from other tissues. As needed, a report containing information such as the location, size, and CT value of the calcifications can be generated and may be displayed in the form of images or tables.
[0030] In one embodiment, for the aforementioned automatic pixel scanning, the image width (Width) and height (Height) are set, along with a matrix or data structure for storing the results (e.g., a two-dimensional array for marking calcifications). Two nested loops are used: the outer loop iterates through each row (Height) of the image, and the inner loop iterates through each column (Width). Within the loop, the coordinates of the currently traversed pixel can be determined by the row index. and column indexes express.
[0031] In one embodiment, for S3 above, the mean area method assesses the degree of calcification by comprehensively considering the area and density of the calcification foci. The mean area method calculates the total area and the mean CT value of the selected pixels that meet the definition of calcification foci. ,in, S is the average CT value of the pixels in S2 that meet the definition of calcification, where S is the total area occupied by the pixels selected in S2. for , This represents the area corresponding to a single pixel.
[0032] In one embodiment, for S3 above, the layered accumulation method reflects the distribution of calcifications in three-dimensional space in more detail:
[0033]
[0034] in, This represents the total number of slices in the CT scan. For the first The contribution of calcification fraction in the layer, For the first The area occupied by pixels in the layer that conform to the definition of calcification. For the first The average CT value of these pixels in the layer.
[0035] In one embodiment, for S3 above, the CT value summation method is applicable to cases emphasizing calcification density: ,in, This is the sum of CT values of pixels in S2 that meet the definition of calcification. For pixel spacing, This represents the area corresponding to a single pixel.
[0036] In one embodiment, for S4 above, the calcification fraction calculated by any of the methods in S3 is corrected:
[0037] in, It represents the total number of slices in a CT scan.
[0038] In one embodiment, for the above S5, the total calcification area refers to the total area occupied by all calcification foci on the two-dimensional image:
[0039] in, Yes, yes, the first The area of each calcification foci This represents the total number of calcifications. The total calcification volume is the total volume occupied by all calcification foci in three-dimensional space. or
[0040] in, The calcifications are spherical. It is the first The radius of each calcification (assuming it is spherical). It is its area on a two-dimensional slice. It is the average thickness of calcifications on the slice or the distance between adjacent slices; Total calcification mass refers to the total mass of all calcification foci:
[0041] in, It is the density of the calcified material; The Agatston score is a method for assessing coronary artery calcification only.
[0042] in, It is the first The area of each calcification foci The weights are determined based on the highest CT value within the calcification. Calcifications with CT values in the range of 130 HU-199 HU are assigned a weight of 1, those with CT values in the range of 200 HU-299 HU are assigned a weight of 2, those with CT values in the range of 300 HU-399 HU are assigned a weight of 3, and those with CT values ≥400 HU are assigned a weight of 4. Suppose two calcifications are identified on a coronary CT image, one of which has an area of 10 mm. 2 The highest CT value within the calcification was 250 HU; another calcification had an area of 5 mm. 2 The highest CT value within the calcification was 150 HU, calculated using the Agatston score method: The weight of the first calcification is 2; The weight of the second calcification is 1 (assuming that the weight of CT value 150 HU is 1). Therefore, Agatston's score is 25.
[0043] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A method for assessing calcification in human tissue based on CT images, characterized in that, Includes the following steps: S1: Personalized definition of calcifications: Determine the observation range on CT images based on the tissue to be evaluated, and set the CT value and lower limit of the continuous area according to the tissue characteristics; S2: Pixel Filtering: Filter calcification pixels according to personalized definitions; S3: Calculate the calcification score: Calculate the calcification score based on the selected calcification pixels using appropriate methods, including the mean area method, the layered accumulation method, and the CT value summation method. S4: Calcification fraction correction: Correct the calcification fraction based on the number of CT slices; S5: Other related calculations: Calculate the total calcification area, total volume, total mass, and Agatston score.
2. The method for assessing human tissue calcification based on CT images according to claim 1, characterized in that, In S1, the tissue characteristics include the tissue density, composition and structure, as well as the characteristics of the lesion. The tissue density, composition and structure determine its grayscale representation on the CT image, and thus determine the CT value. The characteristics of the lesion determine the setting of the lower limit of the continuous area.
3. The method for assessing human tissue calcification based on CT images according to claim 1, characterized in that, In S2, the pixel filtering is performed by filtering CT images based on the CT value and the lower limit of the continuous area set in S1.
4. The method for assessing human tissue calcification based on CT images according to claim 2, characterized in that, In S2, the specific steps for pixel selection are as follows: Image loading and preprocessing: Preprocessing the CT image files to be processed; Set screening criteria: Set CT value thresholds based on tissue density, composition, and structure; set lower limits for continuous area based on lesion characteristics. Automatic pixel scanning: Automatically traverses every pixel in the image and records the coordinates of the currently traversed pixel; Threshold judgment: Read the CT value of the current pixel and compare it with the CT value set in S1. If the CT value of the current pixel exceeds the set threshold, the pixel is considered to be part of the calcification foci and the area is judged; otherwise, continue to traverse the next pixel. Area determination: For pixels that exceed the CT value threshold, check the CT values of their neighboring pixels. If the CT values of neighboring pixels also exceed the set threshold, add them to the current calcification foci region and continue to expand outward until no more neighboring pixels meet the condition. Calculate the area of the formed continuous region and compare it with the lower limit of the continuous area set in S1. If the area of the continuous region exceeds the set lower limit, the region is considered a calcification foci; otherwise, it is regarded as a noise point or artifact and is not retained. Labeling and visualization: The detected calcification foci are labeled on the image.
5. The method for assessing human tissue calcification based on CT images according to claim 4, characterized in that, In automatic pixel scanning, the width and height of the image are set, along with a matrix or data structure to store the results. Two nested loops are used: the outer loop iterates through each row of the image, and the inner loop iterates through each column. Within the loop, the coordinates of the currently visited pixel can be determined by its row index. and column indexes express.
6. The method for assessing human tissue calcification based on CT images according to claim 1, characterized in that, In S3, the mean area method assesses the degree of calcification by comprehensively considering the area and density of the calcification foci. The mean area method calculates the total area and the mean CT value of the selected pixels that meet the definition of calcification foci. ,in, S is the average CT value of the pixels in S2 that meet the definition of calcification, where S is the total area occupied by the pixels selected in S2. for , This represents the area corresponding to a single pixel.
7. The method for assessing human tissue calcification based on CT images according to claim 1, characterized in that, In S3, the layered accumulation method reflects the distribution of calcifications in three-dimensional space in more detail: in, This represents the total number of slices in the CT scan. For the first The contribution of calcification fraction in the layer, For the first The area occupied by pixels in the layer that conform to the definition of calcification. For the first The average CT value of these pixels in the layer. The area corresponding to a single pixel. The number of pixels in the i-th layer that meet the definition of a calcification foci.
8. The method for assessing human tissue calcification based on CT images according to claim 1, characterized in that, In S3, the CT value summation method is applicable to cases where calcification density is emphasized: ,in, This is the sum of CT values of pixels in S2 that meet the definition of calcification. For pixel spacing, This represents the area corresponding to a single pixel.
9. The method for assessing human tissue calcification based on CT images according to claim 1, characterized in that, In S4, the calcification fraction calculated by any of the methods in S3 is corrected: in, It represents the total number of slices in a CT scan.
10. The method for assessing human tissue calcification based on CT images according to claim 1, characterized in that, In S5, the total calcification area refers to the total area occupied by all calcification foci on the two-dimensional image: in, Yes, yes, the first The area of each calcification foci This represents the total number of calcifications. The total calcification volume refers to the total volume occupied by all calcification foci in three-dimensional space. or in, The calcifications are spherical. It is the first The radius of each calcified foci It is its area on a two-dimensional slice. It is the average thickness of calcifications on the slice or the distance between adjacent slices; The total mass of calcification refers to the total mass of all calcification foci: in, It is the density of the calcified material; The Agatston score is a method for assessing coronary artery calcification only. in, It is the first The area of each calcification foci The weights are determined based on the CT values of the calcifications. Calcifications with CT values in the range of 130HU-199HU are assigned a weight of 1, those with CT values in the range of 200HU-299HU are assigned a weight of 2, those with CT values in the range of 300HU-399HU are assigned a weight of 3, and those with CT values ≥400HU are assigned a weight of 4.