Radiotherapy positioning image recognition method, computer program and computer storage medium

Abstract

An embodiment of the present invention provides a radiotherapy CT positioning image processing method, comprising: defining pixels in a CT original image whose CT values ​​are greater than or equal to a first threshold as human body pixels; counting the number of human body pixels in each pixel row in the CT original image in order from top to bottom; and determining a human body boundary based on the statistical results.

Description

Technical Field

[0001] The present invention relates to the field of medical imaging, and in particular to a radiotherapy positioning CT image processing method, a computer program and a computer storage medium. Background Art

[0002] Before performing radiotherapy (RT) on a patient, a doctor may use a CT (Computing Tomography) imaging device to obtain a positioning image of the patient in advance, and formulate a radiotherapy plan based on the positioning image.

[0003] When performing CT positioning imaging, patients sometimes need to maintain various specific postures, such as raising their arms, leaning their chest forward, etc. In addition, in addition to the above-mentioned bed board for supporting the patient, these postures need to be maintained with the help of various accessories, such as a headrest for supporting the head, an armrest for supporting the arms, and even a mesh support membrane that can be hardened and formed after accommodating the human head.

[0004] During radiotherapy, the patient needs to maintain a position and posture consistent with that during CT positioning imaging. Therefore, the bed board used to support the patient is usually compatible with both CT imaging equipment and radiotherapy equipment. After completing CT positioning imaging, the bed board and the patient, accessories, etc. on it are transferred to the radiotherapy equipment as is to ensure the above consistency.

[0005] Unlike traditional CT imaging methods, during the CT imaging process used for radiotherapy positioning, the patient usually exceeds the scanning field of view due to the need to assume these specific postures (the aperture of the CT imaging equipment used for radiotherapy positioning is generally larger than that of traditional CT imaging equipment), resulting in truncation of the image data. Subsequently, special image processing is required to reconstruct the image of the truncated part to ensure the integrity of the image of the patient's radiotherapy area. This first requires distinguishing the patient's body image from the bed board, auxiliary accessories, etc. Otherwise, the reconstructed body image will have large errors, seriously affecting the subsequent radiotherapy plan.

[0006] The practice in the prior art is to remove the images of the supporting body under the bed board, the bed board, and the accessories on the bed board from bottom to top in the scanned image. The removal method is to pre-set the height of these components in the image, and then remove the image data at this height after acquiring the image, and only retain the part above this height as the human body image.

[0007] This method has difficulty in accurately identifying human body data when the number of accessories is small and the shapes are simple. It is even more impractical when the number of accessories increases and the shapes of the bed boards or accessories become more complex. Therefore, merchants can only limit this function to specific accessories and recommend that hospitals do not use other accessories, which creates difficulties for actual radiotherapy work.

[0008] Those skilled in the art have also devoted themselves to researching more complex algorithms to remove image data other than the human body. However, on the one hand, the algorithms are too complex, and on the other hand, they still need to be adjusted for different accessories or bed boards. In some cases, such as the increase in CT value caused by the change of the material of the accessories, the tilted placement of the accessories, etc., this method still cannot accurately identify the human body in the image.

[0009] The above situation makes it impossible to further reconstruct accurate human body positioning images through interpolation and other methods, making it impossible to properly plan radiotherapy plans.

[0010] Therefore, it is necessary to provide a new radiotherapy positioning CT image processing method, so that no matter what complex environment the patient is in (mainly including the environment formed by the patient support bed and auxiliary accessories, etc.), the patient's body image can be accurately identified. Summary of the invention

[0011] An object of the present invention is to provide a radiotherapy CT positioning image processing method, which can accurately identify human body image data in CT original images.

[0012] An exemplary embodiment of the present invention provides a radiotherapy CT positioning image processing method, comprising: defining pixels in a CT original image whose CT values ​​are greater than or equal to a first threshold as human body pixels; counting the number of human body pixels in each pixel row in the CT original image in order from top to bottom; and determining a human body boundary based on the statistical results.

[0013] An exemplary embodiment of the present invention further provides a computer program, which is applied to a radiotherapy CT imaging system and used to control the radiotherapy CT imaging system to execute the above-mentioned radiotherapy CT positioning image processing method.

[0014] An exemplary embodiment of the present invention further provides a computer storage medium for storing the above-mentioned computer program.

[0015] Other features and aspects will become apparent from the following detailed description, drawings, and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention may be better understood by describing exemplary embodiments of the present invention in conjunction with the accompanying drawings, in which:

[0017] Figure 1 A flowchart of a radiotherapy CT positioning image processing method provided by an embodiment of the present invention;

[0018] Figure 2 An example diagram of binarizing a CT original image and identifying human body pixels in one embodiment of the present invention;

[0019] Figure 3 This is an example diagram of binarizing an original CT image and identifying human body pixels in another embodiment of the present invention. DETAILED DESCRIPTION

[0020] The specific embodiments of the present invention will be described below. It should be noted that in the specific description of these embodiments, in order to provide a concise description, it is impossible for this specification to provide a detailed description of all the features of the actual embodiments. It should be understood that in the actual implementation of any embodiment, just as in the process of any engineering project or design project, in order to achieve the specific goals of the developer and to meet system-related or business-related restrictions, various specific decisions are often made, and this will also change from one embodiment to another. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for ordinary technicians in the field related to the content disclosed by the present invention, some changes such as design, manufacturing or production based on the technical content disclosed in this disclosure are just conventional technical means, and should not be understood as insufficient content of this disclosure.

[0021] Unless otherwise defined, the technical or scientific terms used in the claims and the specification shall have the usual meaning understood by persons with ordinary skills in the technical field to which the invention belongs. The words "first", "second" and similar words used in the patent application specification and the claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "One" or "one" and other similar words do not indicate a quantitative limitation, but indicate the existence of at least one. "Include" or "comprises" and other similar words mean that the elements or objects appearing before "include" or "comprises" include the elements or objects listed after "include" or "comprises" and their equivalent elements, and do not exclude other elements or objects. "Connected" or "connected" and other similar words are not limited to physical or mechanical connections, nor are they limited to direct or indirect connections.

[0022] The embodiments of the present invention can be used in a radiotherapy CT imaging system. In one embodiment, the radiotherapy CT imaging system can include a scanning system, which includes a scanning frame, the scanning frame is formed with a cylindrical scanning cavity to accommodate a patient, and a bed board supporting the patient, and the bed board is further provided with supporting accessories, such as a headrest, an armrest, etc., to keep the patient in a fixed posture, and a supporting body is provided under the bed board. A tube and a detector are relatively arranged on the scanning frame, and the X-rays emitted by the tube penetrate the human tissue, accessories, bed board and supporting body in sequence and are received by the detector, and the X-rays received by the detector are converted into digital image signals.

[0023] The radiotherapy CT imaging system further includes a data acquisition system and an image reconstruction system, wherein the data acquisition system is used to acquire the digital image signal and transmit it as CT raw image data to the image reconstruction system for image reconstruction. The CT raw image data includes human image data and non-human image data, and therefore, the reconstructed CT raw image includes human images and non-human images. The image reconstruction system can be set on a computer, and the reconstructed image can be displayed on a display connected to the computer. The reconstructed image is composed of a plurality of pixels, and the plurality of pixels are arranged in the form of a digital matrix.

[0024] The radiotherapy CT imaging system further includes a control system, which can also be arranged on a computer and is used to control a scanning system, a data acquisition system and an image reconstruction system.

[0025] The radiotherapy CT positioning image processing method of the embodiment of the present invention is used to process the CT original image to more accurately identify the human body image data therein. The above-mentioned CT original image can be a CT image reconstructed according to the original scanning data when performing radiotherapy positioning imaging. In order to reconstruct the human body data of the truncated part in the CT original image, it is necessary to identify the existing human body image data therein for image reconstruction. Therefore, the less non-human body image data of the parts mixed in the human body image data, the more accurate the recognition result.

[0026] Figure 1 A flowchart of a radiotherapy CT positioning image processing method provided by an embodiment of the present invention, Figure 2 This is an example diagram of binarizing a CT original image and identifying human body pixels in an embodiment of the present invention, wherein Figure 2 The left side is the original CT image, and the right side is the binarized image; Figure 3 This is an example diagram of binarizing the original CT image and identifying human body pixels in another embodiment of the present invention, wherein Figure 3 The left side is the original CT image, and the right side is the binarized image.

[0027] like Figure 1As shown, the method includes steps S12, S14 and S16. In step S12, pixels in the CT original image whose CT values ​​are greater than or equal to a first threshold are defined as human body pixels. The first threshold is an empirical value of the human body CT value. Since most of the human body is composed of water, the CT value of water is close to or equal to 0, and the first threshold is set to be close to 0, for example, 200HU, which can cover most of the human body CT values.

[0028] Further, in step S12, human body pixels can be identified by a binarization method. For example, in the original CT image, the pixel values ​​of pixels whose CT values ​​are greater than or equal to the first threshold are defined as 1, and the pixel values ​​of pixels whose CT values ​​are less than the first threshold are defined as 0, so as to obtain Figure 2 The binarized CT original image is shown.

[0029] In step S14, the number of human body pixels in each pixel row in the original CT image is counted in order from top to bottom.

[0030] It should be noted that the "from top to bottom" mentioned in this application corresponds to the positional relationship between the human body and its supporting components. For example, the human body is on top, and below it are the supporting accessories, supporting bed boards, supporting main body, etc. Therefore, in the original CT image, "up and down" are also defined according to such a positional relationship.

[0031] by Figure 2 For example, in step S14, further processing can be performed based on the binarized CT original image. Figure 2 In the image, the ordinate represents the row of the pixel matrix, and the abscissa represents the column of the pixel matrix. Since the human body pixels have been defined in step S12, the number of human body pixels in each row can be counted in sequence, and in step S16, the human body boundary is determined according to the statistical results. Figure 2 The curve 20 in FIG. 2 shows the distribution of the number of human body pixels on the pixel row.

[0032] In one application, combined with a specific human body shape, the following conclusion can be drawn, that is, the number of human body pixels in the row where the lower boundary is located is significantly less than the number of pixels in the row above it. Therefore, in step S16, the human body boundary can be determined in the following manner:

[0033] It is determined whether the number of human body pixels in the current pixel row is reduced to a specific degree compared with the number of human body pixels in the counted pixel rows. If so, the current pixel row is identified as the lower boundary of the human body.

[0034] In order to quantify the specific degree, the embodiment of the present invention further proposes that in the process of counting from top to bottom, the pixel row with the largest number of human body pixels can be continuously determined as the pixel row with the largest number of human body pixels according to the current statistical result;

[0035] When continuing to count the number of human body pixels in the current row, calculate its ratio to the number of human body pixels in the row with the most human body pixels. If the ratio of the number of human body pixels in the current pixel row to the number of human body pixels in the row with the most human body pixels is less than or equal to a second threshold, the current pixel row is identified as the lower boundary of the human body. The second threshold can be, for example, 10%.

[0036] By using the above method, the human body image can be accurately identified in the original CT image. Figure 2 The human lung shown in the figure has an imaged part as a whole. By further reconstructing the image of the truncated part according to the image data of the human lung image, a more accurate and complete lung image can be obtained.

[0037] like Figure 3 As shown, when the imaging part of the human body includes multiple independent body parts (for example, including the head and shoulders connected to each other), after determining the lower boundary, it is not enough to determine the overall contour of the imaging part of the human body. However, due to the particularity of the human body structure, generally speaking, the imaging part can only include a limited number (generally no more than 3) of body parts (for example, the head and shoulders, or the head and raised arms). Therefore, based on this situation, the embodiment of the present invention proposes that the above-mentioned radiotherapy CT positioning image processing method also includes the following steps: recording the starting point, intermediate breakpoints and end point of the human body pixels of each pixel row, and identifying the contour formed by all the starting points, intermediate breakpoints and end points as the human body part contour. In this way, all connected or independent areas representing body parts can be obtained in the image.

[0038] For example, when counting the human body pixels of each pixel row from left to right, the first human body pixel is the starting point, the last human body pixel is the ending point, and the middle points adjacent to the non-human body points are all middle points.

[0039] In order to eliminate the noise caused by independent pixels or independent small areas, before further determining the effective area in the multiple areas, the following operations can be performed: corroding the contour of the human body part, and dilating the corroded contour of the human body part. The above-mentioned effective area can be understood as the body area that actually needs to be imaged and detected, such as the head, shoulders, etc.

[0040] Furthermore, determining the effective area may include an area division step, a human body pixel area calculation step, and an effective area determination step. In the area division step, the human body part contour is divided into a plurality of areas according to the shape of the human body part contour. When dividing, the division may be performed according to the approximate boundaries of each area constituting the human body part contour. In the human body pixel area calculation step, the human body pixel area of ​​each area is calculated. In the effective area determination step, an area whose human body pixel area is greater than or equal to a third threshold is determined as an effective area. The third threshold may be an empirical value, such as the imaging area of ​​a specific body part. In the effective area determination step, the N areas with the largest human body pixel area among all areas may also be taken as effective areas according to the number N of imaged body parts.

[0041] Furthermore, calculating the area of ​​each region may include a pixel number calculation step, a physical pixel area calculation step, and an image area calculation step. In the pixel number calculation step, the number of human body pixels in each region is calculated, for example, Figure 3 The points in the human body part contour represent the human body pixels in the human body part contour. Since the number of human body pixels in each pixel row has been counted, it is only necessary to sum the number of human body pixels in the pixel rows covered in each area. In the physical pixel area calculation step, the physical pixel area of ​​a single human body pixel is calculated, which is the square of the quotient of the scanning field of view defined in the imaging process and the reconstruction matrix size. In the image area calculation step, the image area of ​​each area can be obtained by multiplying the results obtained in the above two steps, that is, multiplying the number of human body pixels in each area by the physical pixel area of ​​a single human body pixel to obtain the image area of ​​each area.

[0042] The embodiment of the present invention, through a completely opposite idea to that in the prior art, predefines human body pixels in the CT original image according to the CT value, and then counts the number of human body pixels in the pixel row from top to bottom. According to this number, combined with some generalized experience of human body imaging, human body image data can be accurately identified from the CT original image, so that when reconstructing the truncated data later, accurate input values ​​are available, so that a more accurate and complete human body image can be reconstructed. In addition, by sequentially identifying human body image data from top to bottom, the restrictions on the recognition function due to the shape, material, position state (such as tilt, overlap with human body parts in height, etc.) of accessories, bed boards, etc. are avoided, and accurate recognition results can be obtained regardless of the complex support environment in which the patient is in.

[0043] An embodiment of the present invention further provides a computer program, which is applied to a radiotherapy CT imaging system and is used to control the radiotherapy CT imaging system to execute the radiotherapy CT positioning image processing method of any of the above embodiments.

[0044] Specifically, the computer program can be installed on a computer system of a radiotherapy CT imaging system, and is used to execute the following instructions:

[0045] Instruction 1: Define pixels in the CT original image whose CT values ​​are greater than or equal to the first threshold as human body pixels;

[0046] Instruction 2: Count the number of human body pixels in each pixel row in the original CT image from top to bottom;

[0047] Instruction 3: Determine the human body boundary based on the statistical results.

[0048] Specifically, in instruction 1, the pixel values ​​of pixels whose CT values ​​in the CT original image are greater than or equal to the first threshold are defined as 1, and the pixel values ​​of pixels whose CT values ​​are less than the first threshold are defined as 0.

[0049] Instruction 3 may include: determining whether the number of human body pixels in the current pixel row is reduced to a specific degree compared with the number of human body pixels in the counted pixel rows, and if so, identifying the current pixel row as the lower boundary of the human body.

[0050] Specifically, instruction 3 may include the following sub-instructions:

[0051] Instruction 3-1: According to the current statistical result, determine the pixel row with the largest number of human body pixels as the pixel row with the largest number of human body pixels;

[0052] Instruction 3-2: If the ratio of the number of human body pixels in the current pixel row to the number of human body pixels in the maximum number of human body pixel rows is less than or equal to a second threshold, identify the current pixel row as the lower boundary of the human body.

[0053] The computer program may further include instruction 4: recording the starting point, intermediate breakpoints and end point of human body pixels in each pixel row, and identifying the contour formed by all the starting points, intermediate breakpoints and end points as the human body part contour.

[0054] The above computer program may further include instruction 5, which includes the following sub-instructions:

[0055] 5-1: Corrosion treatment of the contours of human body parts;

[0056] 5-2: Expand the contour of the human body part after corrosion treatment.

[0057] The above computer program may further include instruction 6, which includes the following sub-instructions:

[0058] 6-1: Divide the contour of the human body part into multiple areas according to the shape of the contour of the human body part;

[0059] 6-2: Calculate the human pixel area in each region;

[0060] 6-3: Determine a region whose area is greater than or equal to the third threshold as a valid region.

[0061] Among them, sub-instruction 6-2 includes the following sub-instructions:

[0062] 6-2-1: Calculate the number of human pixels in each area;

[0063] 6-2-2: Calculate the physical pixel area of ​​a single human body pixel;

[0064] 6-2-3: Multiply the number of human body pixels in each region by the physical pixel area of ​​a single human body pixel to obtain the image area of ​​each region.

[0065] An embodiment of the present invention further provides a computer storage medium for storing the above-mentioned computer program.

[0066] Some exemplary embodiments have been described above, however, it should be understood that various modifications may be made. For example, if the described techniques are performed in a different order and / or if the components in the described systems, architectures, devices, or circuits are combined in different ways and / or replaced or supplemented by other components or their equivalents, suitable results may be achieved. Accordingly, other implementations also fall within the scope of protection of the claims.

Claims

1. A radiotherapy CT positioning image processing method, comprising: The pixels whose CT values ​​in the CT original image are greater than or equal to the first threshold are defined as human body pixels; From top to bottom, count the number of human body pixels in each pixel row in the original CT image; Determine the human body boundary based on the statistical results. Wherein, the step of determining the human body boundary according to the statistical results comprises: It is determined whether the number of human body pixels in the current pixel row is reduced to a preset level compared with the number of human body pixels in the counted pixel rows. If so, the current pixel row is identified as the lower boundary of the human body.

2. The method according to claim 1, characterized in that: In the CT original image, the pixel values ​​of pixels whose CT values ​​are greater than or equal to the first threshold are defined as 1, and the pixel values ​​of pixels whose CT values ​​are less than the first threshold are defined as 0.

3. The method according to claim 1, characterized in that: The steps of determining the human body boundary based on the statistical results include: According to the current statistical result, the pixel row with the largest number of human body pixels is determined as the pixel row with the largest number of human body pixels; If the ratio of the number of human body pixels in the current pixel row to the number of human body pixels in the maximum number of human body pixel rows is less than or equal to a second threshold, the current pixel row is identified as a lower boundary of the human body.

4. The method according to claim 1, characterized in that: The method further includes: recording the starting point, the middle breaking point and the ending point of the human body pixels in each pixel row, and identifying the contour formed by all the starting points, the middle breaking points and the ending points as the human body part contour.

5. The method according to claim 4, characterized in that Also includes: Performing corrosion processing on the contour of the human body part; The contour of the human body part after corrosion treatment is expanded.

6. The method according to claim 4, characterized in that Also includes: Dividing the human body part contour into a plurality of regions according to the shape of the human body part contour; Calculate the human pixel area of ​​each region; A region whose area is greater than or equal to the third threshold is determined as a valid region.

7. The method according to claim 6, characterized in that Calculating the area of ​​each region involves: Calculate the number of human pixels in each region; Calculate the physical pixel area of ​​a single human body pixel; The image area of ​​each region is obtained by multiplying the number of human body pixels in each region by the physical pixel area of ​​a single human body pixel.

8. A computer program, which is applied to a radiotherapy CT imaging system and is used to control the radiotherapy CT imaging system to execute the radiotherapy CT positioning image processing method according to any one of claims 1 to 7.

9. A computer storage medium for storing the computer program according to claim 8.

Images