Method for searching typical tiny features for radiotherapy quality control image

Abstract

The invention aims to provide a method for searching typical tiny features for a radiotherapy quality control image. The method comprises the following steps: converting a to-be-detected image into a matrix; performing singular value decomposition on the matrix; and obtaining a processing result of the to-be-detected image according to the decomposed matrix. Identified features are generally small and are small pits, small holes or lead points dug out on the uniform die body. And the uniform die body is simple and square in structure, and the opened radiation field is also a square field. The features to be identified are generally relatively complex and are circles or arcs. According to the method, the features to be recognized are separated from the uniform motif image without the features in a decomposition mode, so that the difficulty of feature recognition work is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method for finding typical micro features in radiotherapy quality control images. Background Art

[0002] Radiotherapy is the abbreviation of radiation therapy. The working principle is to irradiate tumor cells with high-energy radiation, such as X-rays, and use the different sensitivity of cells to kill abnormal cells. This kind of radiation can be used for internal and external radiation therapy of tumor cells, which mainly plays a role in treating cancer. External radiotherapy: External radiotherapy refers to a large machine that releases high-energy rays to irradiate a part of the body, such as the area where the tumor is concentrated. Generally, the number of irradiations will be more. Internal radiotherapy: Internal radiotherapy is the injection of radionuclides containing radiation into the body. Radionuclides have the ability to clear tumors and can bind to tumor cells, killing tumor cells through short-distance radiation.

[0003] In radiotherapy quality control, the accelerator often opens a square field to take images of a specified phantom, and obtains corresponding quality control data by identifying small features on the phantom. Figure 2 , we need to identify the center position of the ball in the image; Figure 3 , we need to identify the center positions of the 5 solid circles and 5 small rings in the image; Figure 4 It is an MQC phantom. It is necessary to identify the center position of the circle where the two large arcs are located, etc. As can be seen from the figure, the features to be identified are quite fuzzy. Without zooming in and adjusting the window width and window position, they are almost invisible. In addition, the irradiation conditions (field size, FFF or non-FFF), imaging equipment (noise level, background, bad lines of different imaging equipment), and positioning (i.e. the position of the phantom during shooting, which affects the size and position of the features in the image) will have a greater impact on the image. If each image is cropped and enlarged, the window width and window position are adjusted, and denoised, the processing is quite cumbersome and unintelligent. New situations need to be adjusted again, and the after-sales maintenance cost is high. Summary of the invention

[0004] The purpose of the present invention is to provide a method for finding typical tiny features in radiotherapy quality control images. The method can separate the features to be identified from the uniform model image with the features removed by decomposition, thereby reducing the difficulty of image recognition.

[0005] A method for finding typical tiny features in radiotherapy quality control images, comprising: Convert the image to be detected into a matrix; Performing singular value decomposition on the matrix; The processing result of the image to be detected is obtained according to the decomposed matrix.

[0006] Preferably, converting the image to be detected into a matrix comprises: Consider the image with length h and width w as Two-dimensional matrix , expressed as: ; in They are The matrix of , is a diagonal matrix.

[0007] Preferably, reconstructing the singular value decomposition result includes: Bundle as k column vectors , As k row vectors , and they are all unit vectors, then ; Decompose A into the superposition of k matrices, each of which can be expressed as the product of a column vector and a row vector. is the corresponding diagonal element of S, that is, the superposition coefficient.

[0008] Preferably, the processing result of obtaining the image to be detected according to the decomposed matrix includes: Determine two numbers j and g according to the image to be detected, and , the image is divided into a uniform model background part, expressed as: .

[0009] Preferably, the processing result of obtaining the image to be detected according to the decomposed matrix includes: Determine two numbers j and g according to the image to be detected, and , divide the image into small feature parts to be identified, expressed as: .

[0010] Preferably, the processing result of obtaining the image to be detected according to the decomposed matrix includes: Determine two numbers j and g according to the image to be detected, and , the image is divided into noise parts, expressed as: .

[0011] A system for finding typical tiny features in radiotherapy quality control images, comprising: An image conversion module, used for converting the image to be detected into a matrix; A matrix decomposition module, used for performing singular value decomposition on the matrix; The matrix processing module is used to obtain the processing result of the image to be detected according to the decomposed matrix.

[0012] An electronic device comprises: a chip, a processor and a memory, wherein the memory is used to store computer program codes, and the computer program codes include computer instructions. When the chip executes the computer instructions, the electronic device executes a method for finding typical tiny features in radiotherapy quality control images.

[0013] A computer-readable storage medium stores a computer program, wherein the computer program includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes a method for finding typical tiny features in a radiotherapy quality control image.

[0014] The beneficial effect of the present invention is that the features recognized by the present invention are usually relatively small, such as small pits or holes or lead dots removed from a uniform model. The uniform model has a simple and square structure, and the field of view is also square. The features to be recognized are usually relatively complex, such as circles or arcs. The present invention separates the features to be recognized from the uniform model image with the features removed by decomposition, thereby reducing the difficulty of feature recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0017] Figure 1 A flow chart of a method for finding typical micro features in radiotherapy quality control images according to the present invention; Figure 2 A diagram of a small ball phantom according to the present invention; Figure 3 The penta guide motif diagram of the present invention; Figure 4 is the MQC motif diagram of the present invention; Figure 5 This is a diagram of the MQC phantom of the present invention under FFF irradiation; Figure 6 This is a schematic diagram of the results of decomposing the penta guide motif of the present invention by the method of the present invention; Figure 7 This is a schematic diagram of the result after the MQC motif of the present invention is decomposed by the method of the present invention; Figure 8 The figure is a schematic diagram of the hardware structure of an electronic device of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0020] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0021] In radiotherapy quality control, the accelerator often opens a square field to take images of a specified phantom, and obtains corresponding quality control data by identifying small features on the phantom. Figure 2 , we need to identify the center position of the ball in the image; Figure 3 , we need to identify the center positions of the 5 solid circles and 5 small rings in the image; Figure 4 It is an MQC phantom. It is necessary to identify the center position of the circle where the two large arcs are located, etc. As can be seen from the figure, the features to be identified are quite fuzzy. Without zooming in and adjusting the window width and window position, they are almost invisible. In addition, the irradiation conditions (field size, FFF or non-FFF), imaging equipment (noise level, background, bad lines of different imaging equipment), and positioning (i.e. the position of the phantom during shooting, which affects the size and position of the features in the image) will have a greater impact on the image. If each image is cropped and enlarged, the window width and window position are adjusted, and denoised, the processing is quite cumbersome and unintelligent. New situations need to be adjusted again, and the after-sales maintenance cost is high.

[0022] The features identified by the present invention are usually relatively small, such as small pits or holes or lead dots removed from the uniform model. The uniform model has a simple and square structure, and the field of view is also square. The features to be identified are usually relatively complex, such as circles or arcs. The present invention separates the features to be identified from the uniform model image with the features removed by decomposition, thereby reducing the difficulty of feature recognition.

[0023] A method for finding typical tiny features in radiotherapy quality control images, referring to Figure 1 ,include: S100, converting the image to be detected into a matrix; By extracting features from an image, the image can be converted into a numerical matrix that can represent the key points, edges, textures and other information of the image. This information can be used for tasks such as image recognition, classification, and retrieval.

[0024] S200, performing singular value decomposition on the matrix; ‌Singular value‌ is a mathematical concept of a matrix, which is mainly obtained through singular value decomposition (SVD). Singular value decomposition is to decompose a matrix into the product of three matrices: an orthogonal matrix, a diagonal matrix and the transpose of an orthogonal matrix.

[0025] S300, obtaining a processing result of the image to be detected according to the decomposed matrix.

[0026] In image data analysis, matrix decomposition can reduce high-dimensional data to a low-dimensional space, thereby simplifying calculations and increasing processing speed. For example, techniques such as singular value decomposition (SVD) and principal component analysis (PCA) can decompose data matrices into smaller matrices and extract important eigenvectors, which are helpful for data visualization and analysis. The present invention decomposes the matrix to separate the features to be identified from the uniform model image with the features removed, thereby reducing the difficulty of image feature recognition.

[0027] Preferably, converting the image to be detected into a matrix comprises: Consider the image with length h and width w as Two-dimensional matrix , expressed as: ; in They are The matrix of , is a diagonal matrix.

[0028] Preferably, reconstructing the singular value decomposition result includes: Bundle as k column vectors , as k row vectors , and they are all unit vectors, then ; Decompose A into the superposition of k matrices, each of which can be expressed as the product of a column vector and a row vector. is the corresponding diagonal element of S, that is, the superposition coefficient.

[0029] ‌Singular value decomposition (SVD) decomposes a non-zero real matrix A into the product of three matrices: an orthogonal matrix, a diagonal matrix, and the transpose of an orthogonal matrix. ‌‌‌An orthogonal matrix refers to a square matrix that satisfies the following conditions: the transpose multiplied by itself is equal to the identity matrix, and the row vectors and column vectors are both unit vectors‌: The row vectors and column vectors of an orthogonal matrix are both unit vectors, and they are orthogonal to each other. The absolute value of the determinant of an orthogonal matrix is ​​1; ‌Diagonal matrix‌ is a special square matrix whose elements outside the main diagonal are all 0. A diagonal matrix is ​​a symmetric matrix. When all the elements in the diagonal matrix are 1, it is called the identity matrix. When all the elements in the diagonal matrix are equal, it is called a scalar matrix or a scalar matrix. The transposed matrix‌ refers to a new matrix obtained by swapping the rows and columns of a matrix. Specifically, for an m×n matrix A, its transposed matrix is ​​an n×m matrix. The elements in the transposed matrix indicate that the element in the i-th row and j-th column of the original matrix A is located in the j-th row and i-th column of the transposed matrix.

[0030] Singular value decomposition can reduce the dimensionality of data by retaining a few of the largest singular values, thereby reducing data storage and processing time. This method is particularly effective when processing large-scale data sets. By removing small singular values, the noise in the matrix can be eliminated and the quality of the data can be improved. This is very important in image processing and signal processing. It can help remove background noise and improve the clarity of images and signals. In image processing, SVD can help remove noise in images and improve image quality. By retaining the main singular values ​​and singular vectors, high-quality images can be reconstructed.

[0031] Preferably, obtaining a processing result of the image to be detected according to the decomposed matrix includes: Determine two numbers j and g according to the image to be detected. If the background shape of the model is simpler, j is smaller; if the tiny features are more complex, g is larger, and , the image is divided into a uniform model background part, expressed as: .

[0032] Bundle Arrange from large to small. For general images, the size of s varies greatly, and the first few items contain most of the image information. ,The image can be divided into three parts, the uniform motif part, the small feature part to be identified, and the noise part.

[0033] A homogeneous phantom is a phantom with uniform density and characteristics used to test the basic performance of radiotherapy equipment. Its main purpose is to evaluate the performance of radiotherapy equipment under different conditions and ensure the stability and accuracy of the equipment. A homogeneous phantom is a phantom with uniform density and characteristics used to test the basic performance of radiotherapy equipment. It can provide a standardized environment to help evaluate the stability and accuracy of the equipment. By using a homogeneous phantom, it can be ensured that the performance of radiotherapy equipment is consistent under different conditions, thereby improving the safety and effectiveness of treatment.

[0034] In the embodiment of the present invention, the uniform phantom has a simple and square structure, and the radiation field is also a square field.

[0035] Preferably, obtaining a processing result of the image to be detected according to the decomposed matrix includes: Determine two numbers j and g according to the image to be detected, and , divide the image into small feature parts to be identified, expressed as: .

[0036] In the embodiment of the present invention, the features to be identified are usually relatively complex, such as circles or arcs, and the features to be identified are usually relatively small, such as small pits or holes or lead dots scratched off from a uniform model. Since the features to be identified have the above characteristics, it is difficult for conventional image recognition algorithms to accurately find all the features to be identified. In the embodiment of the present invention, by decomposing the image matrix, three matrix expressions are obtained, each matrix expressing one image feature, so that the image features to be identified can be easily found.

[0037] refer to Figure 5 and Figure 6 , respectively, are the results of the decomposition of the penta guide motif by the method of the present invention and the results of the decomposition of the MQC motif by the method of the present invention, Figure 2 It is a small ball model, and the center position of the ball in the image needs to be identified. Figure 3 This is the Pentaguide model. We need to identify the center positions of the 5 solid circles and 5 small rings in the image. The left one is the original image, and the right one is after zooming in and adjusting the window width and window position. Figure 4 This is the MQC model. To identify the center of the two large arcs, the left one is the original image, and the right one is after zooming in and adjusting the window width and window position. Figure 5 This is the MQC model under FFF irradiation. There is a bad line on the right. It can be seen from the attached figure that the method of the present invention can more accurately identify tiny features.

[0038] Preferably, obtaining a processing result of the image to be detected according to the decomposed matrix includes: Determine two numbers j and g according to the image to be detected, and , the image is divided into noise parts, expressed as: .

[0039] The selection of j is related to the complexity of the uniform model background. For cropped and enlarged images, j is 1, and for general square field images, j is 2 or 3, which will produce good results. The selection of g is related to the image size. Generally, half of k or 100 will produce good results.

[0040] Image noise refers to unnecessary or redundant interference information in image data, which is introduced during the image capture, transmission or processing process, resulting in a decrease in image quality. Image noise can develop from tiny noise to almost full-frame noise. Image noise can be divided into the following categories: Random noise: Noise that exists regardless of the exposure time, mainly affected by the camera ISO value. Banding noise: Noise that is strongly related to the camera, mainly generated during the process of reading data from the chip shift register. Sensor noise: Thermal noise generated by the camera sensor when receiving light, especially in low-light environments. Electronic component noise: Thermal noise generated by the camera's electronic components when working, such as when exposed for a long time. Signal processing noise: Noise introduced during image processing due to algorithms, software, etc., such as when compressing or converting images. The main causes of image noise include: Electronic noise: Generated by the sensor and circuit of a scanner or digital camera. Shot noise: Unavoidable noise in an ideal photodetector. Thermal noise: The heat generated by sensors and electronic components when working is converted into noise.

[0041] Example 2 A system for finding typical tiny features in radiotherapy quality control images, comprising: An image conversion module, used for converting the image to be detected into a matrix; Matrix decomposition module, used to perform singular value decomposition on the matrix; The matrix processing module is used to obtain the processing result of the image to be detected according to the decomposed matrix.

[0042] Example 3 An electronic device includes: a chip, a processor and a memory, the memory is used to store computer program codes, the computer program codes include computer instructions, and when the chip executes the computer instructions, the electronic device executes a method for finding typical tiny features in radiotherapy quality control images.

[0043] refer to Figure 8, the electronic device 2 includes a processor 21, a memory 22, an input device 23, and an output device 24. The processor 21, the memory 22, the input device 23, and the output device 24 are coupled via a connector, and the connector includes various interfaces, transmission lines, or buses, etc., which are not limited in the embodiments of the present invention. It should be understood that in various embodiments of the present invention, coupling refers to mutual connection in a specific manner, including direct connection or indirect connection through other devices, for example, through various interfaces, transmission lines, buses, etc.

[0044] The processor 21 may be one or more graphics processing units (GPUs). When the processor 21 is a GPU, the GPU may be a single-core GPU or a multi-core GPU. Optionally, the processor 21 may be a processor group consisting of multiple GPUs, and the multiple processors are coupled to each other via one or more buses. Optionally, the processor may also be other types of processors, etc., which are not limited in the embodiments of the present invention.

[0045] The memory 22 can be used to store computer program instructions and various computer program codes including program codes for executing the scheme of the present invention. Optionally, the memory includes but is not limited to random access memory (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM), or portable read only memory (CD-ROM), which is used for related instructions and data.

[0046] The input device 23 is used to input data and / or signals, and the output device 24 is used to output data and / or signals. The output device 24 and the input device 23 may be independent devices or an integrated device.

[0047] Example 4 A computer-readable storage medium stores a computer program, wherein the computer program includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes a method for finding typical micro features in a radiotherapy quality control image.

[0048] The features identified by the present invention are usually relatively small, such as small pits or holes or lead dots removed from the uniform model. The uniform model has a simple and square structure, and the field of view is also square. The features to be identified are usually relatively complex, such as circles or arcs. The present invention separates the features to be identified from the uniform model image with the features removed by decomposition, thereby reducing the difficulty of feature recognition.

[0049] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for finding typical micro features in radiotherapy quality control images, characterized in that: include: Convert the image to be detected into a matrix; Performing singular value decomposition on the matrix; The processing result of the image to be detected is obtained according to the decomposed matrix.

2. A method for finding typical micro features in radiotherapy quality control images according to claim 1, characterized in that: The step of converting the image to be detected into a matrix comprises: Consider the image with length h and width w as Two-dimensional matrix , expressed as: ； in They are The matrix of , is a diagonal matrix.

3. A method for finding typical micro features in radiotherapy quality control images according to claim 2, characterized in that: Reconstructing the singular value decomposition results includes: Bundle As k column vectors , as k row vectors , and they are all unit vectors, then ； Decompose A into the superposition of k matrices, each of which can be expressed as the product of a column vector and a row vector. is the corresponding diagonal element of S, that is, the superposition coefficient.

4. The method for finding typical micro features in radiotherapy quality control images according to claim 3, characterized in that: The processing result of obtaining the image to be detected according to the decomposed matrix includes: Determine two numbers j and g according to the image to be detected, and , the image is divided into a uniform model background part, expressed as: 。 5. The method for finding typical micro features in radiotherapy quality control images according to claim 3, characterized in that: The processing result of obtaining the image to be detected according to the decomposed matrix includes: Determine two numbers j and g according to the image to be detected, and , divide the image into small feature parts to be identified, expressed as: 。 6. The method for finding typical micro features in radiotherapy quality control images according to claim 3, characterized in that: The processing result of obtaining the image to be detected according to the decomposed matrix includes: Determine two numbers j and g according to the image to be detected, and , the image is divided into noise parts, expressed as: 。 7. A system for finding typical micro features in radiotherapy quality control images, characterized in that: include: An image conversion module, used for converting the image to be detected into a matrix; A matrix decomposition module, used for performing singular value decomposition on the matrix; The matrix processing module is used to obtain the processing result of the image to be detected according to the decomposed matrix.

8. An electronic device, characterized in that: include: A chip, a processor and a memory, wherein the memory is used to store computer program codes, wherein the computer program codes include computer instructions. When the chip executes the computer instructions, the electronic device executes a method for finding typical tiny features in radiotherapy quality control images as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes a method for finding typical tiny features in a radiotherapy quality control image according to any one of claims 1 to 6.

Images