Kidney stone three-dimensional coordinate determination method, system and device and storage medium

Abstract

The invention discloses a kidney stone three-dimensional coordinate determination method, system and device and a storage medium, and the method comprises the steps: inputting a plurality of two-dimensional CT grayscale images into an improved active contour model, obtaining a corresponding improved energy functional value through an improved energy functional, and after the improved energy functional value is converged, determining a three-dimensional coordinate of a kidney stone; the kidney contour internal pixel point information is output, the improved active contour model is constructed by the energy function of the improved Chan-Vese model, and kidney three-dimensional point cloud data is generated according to the kidney contour internal pixel point information corresponding to each two-dimensional CT gray level image; and selecting a plurality of pieces of kidney stone pixel point information from the kidney three-dimensional point cloud data based on a kidney stone pixel threshold constraint condition, and determining three-dimensional coordinates of the kidney stone according to the plurality of pieces of kidney stone pixel point information. According to the method, the kidney contour is extracted through the improved active contour model fusing the texture features and the local energy fitting item, the boundary of the kidney and calculi is captured more accurately, and misdiagnosis and missed diagnosis conditions are reduced.

Description

Technical Field

[0001] The present invention belongs to the field of medical image technology processing technology, and in particular relates to a method, system, device and storage medium for determining the three-dimensional coordinates of a kidney stone. Background Art

[0002] Kidney stones are a common urinary system disease, and early and accurate diagnosis is of great significance for treatment and prevention. Traditional kidney stone detection methods mainly rely on the doctor's experience and naked eye observation, which are highly subjective and have a high rate of missed diagnosis.

[0003] In recent years, with the development of medical imaging technology, automatic kidney stone extraction methods based on 3D CT images have gradually become a research hotspot. However, existing methods still have deficiencies in data preprocessing, contour extraction, and stone localization, which affects the accuracy and reliability of diagnosis.

[0004] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of the present invention is to provide a method, system, device and storage medium for determining the three-dimensional coordinates of kidney stones, aiming to solve the technical problem of how to efficiently and accurately extract the three-dimensional coordinates of kidney stones.

[0006] To achieve the above object, the present invention provides a method for determining the three-dimensional coordinates of a kidney stone, the method comprising: Acquire a three-dimensional CT image of a patient with kidney stones, and perform slice preprocessing on the three-dimensional CT image to obtain a plurality of two-dimensional CT grayscale images; Inputting a plurality of two-dimensional CT grayscale images into an improved active contour model, obtaining an improved energy functional value corresponding to each two-dimensional CT grayscale image through an improved energy functional, wherein the improved active contour model is constructed by an energy functional of an improved Chan-Vese model, and the improved energy functional is constructed by a combination of an overall energy term, a local energy fitting term, and a regularization term; After the improved energy functional value converges, the improved active contour model outputs the internal pixel point information of the kidney contour corresponding to each two-dimensional CT grayscale image; Generate kidney three-dimensional point cloud data according to the internal pixel information of kidney contour corresponding to each two-dimensional CT grayscale image; Based on the kidney stone pixel threshold constraint condition, multiple kidney stone pixel point information is selected from the kidney three-dimensional point cloud data, and the three-dimensional coordinates of the kidney stone are determined according to the multiple kidney stone pixel point information.

[0007] Optionally, the step of inputting a plurality of two-dimensional CT grayscale images into the improved active contour model and obtaining an improved energy functional value corresponding to each two-dimensional CT grayscale image by improving the energy functional comprises: Input multiple two-dimensional CT grayscale images into the improved active contour model, and calculate the overall energy value, local energy fitting value and regularization value of each two-dimensional CT grayscale image; The improved energy functional value of each two-dimensional CT grayscale image is calculated by improving the energy functional according to the overall energy value, the local energy fitting value and the regularization value.

[0008] Optionally, the step of calculating the overall energy value of each two-dimensional CT grayscale image includes: Selecting a target CT grayscale image from a plurality of two-dimensional CT grayscale images; Determine the initial kidney contour parameters of the target CT grayscale image, and determine the grayscale value of the pixel points in each two-dimensional CT grayscale image, the average grayscale value inside the contour, the average grayscale value outside the contour, and the texture feature value of the pixel points; The overall energy value of each two-dimensional CT grayscale image is calculated through the overall energy term according to the initial kidney contour parameters, the grayscale value of the pixel point, the average grayscale value inside the contour, the average grayscale value outside the contour and the pixel point texture feature value.

[0009] Optionally, the step of determining the texture feature value of the pixel points of each two-dimensional CT grayscale image includes: Perform Gaussian blur preprocessing on each two-dimensional CT grayscale image to obtain multiple two-dimensional smooth images; Perform kidney windowing processing on multiple two-dimensional smooth images respectively to obtain multiple windowed two-dimensional images; The local texture features of each windowed two-dimensional image are extracted by LBP algorithm to obtain multiple LBP feature maps; Multiple two-dimensional smooth images and multiple LBP feature maps are linearly combined to obtain the pixel texture feature values ​​corresponding to each two-dimensional CT grayscale image.

[0010] Optionally, the step of calculating the local energy fitting value of each two-dimensional CT grayscale image includes: Determine the grayscale value of the pixel points in each two-dimensional CT grayscale image and the local area mean value of the pixel points; The local energy fitting value of each two-dimensional CT grayscale image is calculated according to the grayscale value of the pixel point and the local area mean value of the pixel point through the local energy fitting term.

[0011] Optionally, the step of calculating the regularization value of each two-dimensional CT grayscale image includes: Determine the kidney contour balance curve length and kidney contour area corresponding to each two-dimensional CT grayscale image; The regularization value of each two-dimensional CT grayscale image is calculated through a regularization term according to the length of the kidney contour balance curve and the area of ​​the kidney contour region.

[0012] In addition, to achieve the above-mentioned purpose, the present invention also proposes a three-dimensional coordinate determination system for kidney stones, the three-dimensional coordinate determination system for kidney stones comprising: A processing module, used for acquiring a three-dimensional CT image of a patient with kidney stones, and performing slice preprocessing on the three-dimensional CT image to obtain a plurality of two-dimensional CT grayscale images; A kidney contour extraction module, used for inputting a plurality of two-dimensional CT grayscale images into an improved active contour model, and obtaining an improved energy functional value corresponding to each two-dimensional CT grayscale image through an improved energy functional, wherein the improved active contour model is constructed by an energy functional of an improved Chan-Vese model, and the improved energy functional is constructed by a combination of an overall energy term, a local energy fitting term, and a regularization term; The kidney contour extraction module is further used to enable the improved active contour model to output the internal pixel point information of the kidney contour corresponding to each two-dimensional CT grayscale image after the improved energy functional value converges.

[0013] The processing module is also used to generate kidney three-dimensional point cloud data according to the internal pixel information of the kidney contour corresponding to each two-dimensional CT grayscale image; The coordinate determination module is used to select multiple kidney stone pixel point information from the kidney three-dimensional point cloud data based on the kidney stone pixel threshold constraint condition, and determine the three-dimensional coordinates of the kidney stone according to the multiple kidney stone pixel point information.

[0014] In addition, to achieve the above-mentioned purpose, the present invention also proposes a device for determining the three-dimensional coordinates of kidney stones, the device comprising: a memory, a processor, and a three-dimensional coordinate determination program for kidney stones stored in the memory and executable on the processor, the three-dimensional coordinate determination program for kidney stones being configured to implement the steps of the method for determining the three-dimensional coordinates of kidney stones as described above.

[0015] In addition, to achieve the above-mentioned purpose, the present invention also proposes a storage medium, on which a program for determining the three-dimensional coordinates of kidney stones is stored. When the program for determining the three-dimensional coordinates of kidney stones is executed by a processor, the steps of the method for determining the three-dimensional coordinates of kidney stones as described above are implemented.

[0016] The present invention first obtains a three-dimensional CT image of a patient with kidney stones, and slices and pre-processes the three-dimensional CT image to obtain multiple two-dimensional CT grayscale images, and then inputs the multiple two-dimensional CT grayscale images into an improved active contour model, obtains the corresponding improved energy functional value by improving the energy functional, and outputs the internal pixel point information of the kidney contour corresponding to each two-dimensional CT grayscale image after the improved energy functional value converges, and the improved active contour model is constructed by the energy functional of the improved Chan-Vese model, and the improved energy functional is constructed by the combination of the overall energy term, the local energy fitting term and the regularization term, and then generates kidney three-dimensional point cloud data according to the internal pixel point information of the kidney contour corresponding to each two-dimensional CT grayscale image, and finally selects multiple kidney stone pixel point information from the kidney three-dimensional point cloud data based on the kidney stone pixel threshold constraint condition, and determines the three-dimensional coordinates of the kidney stone according to the multiple kidney stone pixel point information. The present invention combines texture features and local energy fitting technology, can more accurately capture the boundary of kidney and stone, and greatly improves the accuracy and efficiency of kidney stone identification by combining the improved active contour model with multi-stage image preprocessing, and reduces misdiagnosis and missed diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic structural diagram of a device for determining the three-dimensional coordinates of kidney stones in a hardware operating environment involved in an embodiment of the present invention; Figure 2 It is a schematic flow chart of a first embodiment of a method for determining three-dimensional coordinates of a kidney stone according to the present invention; Figure 3 This is a structural block diagram of the first embodiment of the system for determining the three-dimensional coordinates of kidney stones of the present invention.

[0018] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0019] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0020] Reference Figure 1 , Figure 1 The present invention is a schematic diagram of the structure of a device for determining the three-dimensional coordinates of kidney stones in the hardware operating environment involved in the embodiment of the present invention.

[0021] like Figure 1As shown, the device for determining the three-dimensional coordinates of the kidney stone may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (RandomAccess Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage system independent of the aforementioned processor 1001.

[0022] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the device for determining the three-dimensional coordinates of kidney stones, and may include more or less components than those shown in the figure, or a combination of certain components, or a different arrangement of components.

[0023] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a program for determining the three-dimensional coordinates of kidney stones.

[0024] exist Figure 1 In the three-dimensional coordinate determination device for kidney stones shown, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the three-dimensional coordinate determination device for kidney stones of the present invention can be set in the three-dimensional coordinate determination device for kidney stones, and the three-dimensional coordinate determination device for kidney stones calls the three-dimensional coordinate determination program for kidney stones stored in the memory 1005 through the processor 1001, and executes the three-dimensional coordinate determination method for kidney stones provided in an embodiment of the present invention.

[0025] The embodiment of the present invention provides a method for determining the three-dimensional coordinates of a kidney stone, referring to Figure 2 , Figure 2 Schematic diagram of the flow chart of the first embodiment of the method for determining the three-dimensional coordinates of kidney stones of the present invention.

[0026] In this embodiment, the method for determining the three-dimensional coordinates of the kidney stone comprises the following steps: Step S10: Acquire a three-dimensional CT image of a patient with kidney stones, and perform slice preprocessing on the three-dimensional CT image to obtain a plurality of two-dimensional CT grayscale images.

[0027] It is easy to understand that the execution subject of this embodiment can be a three-dimensional coordinate determination system for kidney stones with functions such as data processing, network communication and program running, or it can be other computer equipment with similar functions, etc., and this embodiment is not limited.

[0028] In this embodiment, it is necessary to obtain the DICOM format CT slice file (i.e., DICOM file) of the kidney stone patient in advance, and then use the medical image processing software SimpleITK to parse the DICOM format CT slice file to obtain a three-dimensional CT image and DICOM metadata, and its DICOM metadata includes detailed information such as patient information, scanning parameters, and CT layer number. Specific parameters include image resolution, scanning area size, and scanning time.

[0029] In a specific implementation, it is necessary to resample the three-dimensional CT image and adjust the resolution of the three-dimensional CT image to meet the needs of subsequent processing.

[0030] The resampling process adjusts all images to the same voxel size of 1mm×1mm×1mm to eliminate the impact of equipment differences. In addition, the image is normalized to adjust the grayscale value of the image to a uniform range of 0 to 255 to enhance the image contrast and improve the visualization effect.

[0031] Furthermore, since the three-dimensional CT images to be sliced ​​are in DICOM format, they need to be converted into PNG format three-dimensional CT images, and then the PNG format three-dimensional CT images are sliced ​​according to the number of CT layers (i.e., the number of CT slices), and the multiple CT images after slicing (multiple CT images in PNG format) are rotated, weakened, and contrast enhanced to standardize the image data and optimize subsequent stone detection performance, so as to obtain multiple two-dimensional CT grayscale images.

[0032] Assuming that the number of CT slices is 256, 256 two-dimensional CT grayscale images are generated.

[0033] It should also be noted that the steps of rotating, weakening and contrast enhancing the sliced ​​multiple CT images are as follows: First, the image is rotated. All images will be rotated 180 degrees to ensure that the head of the image is always facing upwards, which helps to ensure the consistency of the image during the standardization process. The rotation operation will be performed around the center point of the image; Image weakening was performed by applying a 3x3 Gaussian filter to each CT image, where the standard deviation of the Gaussian kernel was set to 1.5 to smooth the image and reduce noise, thereby reducing the recognition error of the subsequent algorithm; Contrast enhancement was performed by processing the images using contrast-limited adaptive histogram equalization (CLAHE), with the clipping limit set to 2.0 and the applied grid size of 8x8. This was used to improve the contrast between the calculi and the surrounding tissues in the images, making the calculi more visible and easier to identify.

[0034] Step S20: Input multiple two-dimensional CT grayscale images into the improved active contour model, and obtain the improved energy functional value corresponding to each two-dimensional CT grayscale image through the improved energy functional, wherein the improved active contour model is constructed by the energy functional of the improved Chan-Vese model, and the improved energy functional is constructed by combining the global energy term, the local energy fitting term and the regularization term.

[0035] It should be noted that the improved active contour model improves the overall energy functional of the Chan-Vese (CV) model. It no longer simply calculates the average grayscale value of the internal and external areas of the image. Instead, the difference between the grayscale value of the image area and its average grayscale value is directly expressed by the square difference, and combined with saliency mapping to enhance the edge recognition ability.

[0036] The improved active contour model effectively integrates the global energy term and the local energy fitting term, and enhances the smoothness and adaptability of the model through the regularization term. This model can accurately identify the edge of the target without resetting, especially in medical images with complex backgrounds and uneven grayscale.

[0037] The energy functional of the improved model consists of three parts: global energy term, local energy fitting term and regularization term.

[0038] Furthermore, multiple two-dimensional CT grayscale images are input into the improved active contour model to calculate the overall energy value, local energy fitting value and regularization value of each two-dimensional CT grayscale image; and the improved energy functional value of each two-dimensional CT grayscale image is calculated through the improved energy functional according to the overall energy value, local energy fitting value and regularization value.

[0039] The improved energy functional is defined as:

[0040] Where E is the improved energy functional value, is the overall energy value, is the local energy fitting value, is the regularization value, and w is a constant ranging from 0 to 1, which is adjusted according to the degree of grayscale non-uniformity. Here w is initialized to 0.7.

[0041] It should be understood that the information of the inner pixel point of the kidney contour is the grayscale value of the inner pixel point of the kidney contour in the range of 0 to 255 after normalization.

[0042] Furthermore, the processing method for calculating the overall energy item of each two-dimensional CT grayscale image is to select a target CT grayscale image from multiple two-dimensional CT grayscale images; determine the initial kidney contour parameters of the target CT grayscale image, and determine the grayscale value of the pixel points in each two-dimensional CT grayscale image, the average grayscale value inside the contour, the average grayscale value outside the contour, and the texture eigenvalue of the pixel points; calculate the overall energy value of each two-dimensional CT grayscale image through the overall energy item based on the initial kidney contour parameters, the grayscale value of the pixel points, the average grayscale value inside the contour, the average grayscale value outside the contour, and the texture eigenvalue of the pixel points.

[0043] It should also be noted that the target CT grayscale image may be the first two-dimensional CT grayscale image or the last two-dimensional CT grayscale image corresponding to the CT cut in the normal order.

[0044] In the specific implementation, the initial contour parameters are set first. For each 512×512 resolution 2D CT grayscale image, the initialized level set contour parameters are represented as two circular contours, where the centers are (320, 170) and (300, 350) respectively, and the radius is 75 pixels. They correspond to the approximate initial positions of the left and right kidneys of the human body.

[0045] In order to improve the processing capability of images with uneven grayscale and simplify the calculation process, the calculation of the overall energy term is optimized for the global energy term of the simplified CV model. The square difference is directly used in the calculation, and the difference between the grayscale value of the image area and its average grayscale value is directly expressed by the square difference, which avoids complex numerical calculations and improves stability.

[0046] The overall energy term is:

[0047] In the formula, are the initial contour parameters, is the gray value of the pixel position (x, y) in each 2D CT grayscale image, c1 is the average gray value inside the contour, and c2 is the average gray value outside the contour. and is the weight coefficient, controlling the importance of the energy terms in the inner and outer regions. Set to 0.4, Set to 0.6, It is the texture feature value of the pixel position (x, y), which is used to enhance the recognition ability of the target edge.

[0048] Furthermore, the processing method for determining the pixel texture feature value of each two-dimensional CT grayscale image is to perform Gaussian blur preprocessing on each two-dimensional CT grayscale image to obtain multiple two-dimensional smooth images; perform kidney windowing processing on the multiple two-dimensional smooth images to obtain multiple windowed two-dimensional images; extract the local texture features of each windowed two-dimensional image by LBP algorithm to obtain multiple LBP feature maps; linearly combine the multiple two-dimensional smooth images and the multiple LBP feature maps to obtain the pixel texture feature value corresponding to each two-dimensional CT grayscale image.

[0049] In the specific implementation, for each two-dimensional CT grayscale image Perform Gaussian blur preprocessing to obtain multiple smoothed images (i.e. multiple 2D smooth images).

[0050] It should also be noted that in order to highlight the texture feature information of the kidney, the image smoothed by Gaussian blur , kidney windowing was performed on the two-dimensional smoothed image: Windowing is to adjust the window width and window level of the image based on the CT value range of the kidney, so that the contrast of the kidney tissue is enhanced, while the contrast of the non-target tissue is relatively reduced. Based on the CT value range of the kidney itself, the distribution of the CT value of the kidney is concentrated between -75~325Hu, so the window width is selected as 390Hu and the window level is 120Hu, and the two-dimensional smoothed image is windowed.

[0051] After windowing, the LBP algorithm is applied to extract local texture features based on the windowed two-dimensional image. Among them, the LBP algorithm parameters include neighborhood size and sampling radius. The neighborhood size and sampling radius are for each pixel in the windowed image. Here, the neighborhood size parameter is set to 8, the sampling radius parameter is set to 1, and the mode parameter uses the uniform mode. The output of the LBP algorithm is the LBP feature map .

[0052] Next, the smoothed image With LBP feature map Perform linear combination to form pixel texture feature value α is a weight coefficient used to balance the importance of grayscale information and texture information. Here α is set to 0.5.

[0053]

[0054] Furthermore, the processing method for calculating the local energy fitting item of each two-dimensional CT grayscale image is to determine the grayscale value of the pixel points in each two-dimensional CT grayscale image and the local area mean of the pixel points; and calculate the local energy fitting value of each two-dimensional CT grayscale image through the local energy fitting item according to the grayscale value of the pixel points and the local area mean of the pixel points.

[0055] It should be noted that the feature extraction of local energy terms is achieved through the local binary pattern (LBP), which is a powerful texture descriptor that can capture local texture information in the image. In two-dimensional CT grayscale images, texture information plays a key role in distinguishing kidney tissue from surrounding abdominal adhesion tissue. The idea of ​​the Local Binary Pattern (LBP) algorithm is adopted here to capture the local features of the image through the Gaussian kernel function.

[0056] It should also be noted that the local energy fitting term can better reflect the local features of the image and help improve the segmentation accuracy of the model.

[0057] The local energy fitting term is:

[0058] In the formula, is the gray value of the i-th pixel, Ω represents the set of all pixels in the two-dimensional CT gray image, is the mean of the local area where the pixel is located (i.e. the local area mean of the pixel point), It is a weight factor dynamically calculated based on the pixel position and grayscale value, and its initial value is set to 0.2.

[0059] Furthermore, the processing method for calculating the regularization term of each two-dimensional CT grayscale image is to respectively determine the kidney contour balance curve length and the kidney contour area area corresponding to each two-dimensional CT grayscale image; and calculate the regularization value of each two-dimensional CT grayscale image through the regularization term according to the kidney contour balance curve length and the kidney contour area area.

[0060] In this embodiment, the regularization term is mainly used to maintain the smoothness of the contour and prevent excessive details from appearing during the curve evolution process. Two different regularization strategies are used here, namely curve length regularization and region area regularization. Curve length regularization encourages smooth boundaries by penalizing the length of the contour, and region area regularization encourages compact segmentation results by penalizing the area of ​​the region.

[0061] Regularization term:

[0062] In the formula, μ and ν are regularization parameters used to balance the influence of curve length and area. Here, μ is set to 0.3 and ν is set to 0.7. The length of the balanced curve of the kidney contour, is the area of ​​kidney contour.

[0063] Step S30: After the improved energy functional value converges, the improved active contour model is enabled to output the internal pixel point information of the kidney contour corresponding to each two-dimensional CT grayscale image.

[0064] In the specific implementation, it is determined whether the multiple improved energy function values ​​in the improved active contour model converge. If not, the number of iterations is increased, and the steps of calculating the overall energy term, the local energy fitting term and the regularization term are repeated. If converged, the pixel information obtained after convergence is output (i.e., the pixel information inside the kidney contour, the pixel information includes the pixel position information, the pixel gray value or the HU value).

[0065] Step S40: Generate kidney three-dimensional point cloud data according to the internal pixel information of the kidney contour corresponding to each two-dimensional CT grayscale image.

[0066] It should be noted that the kidney three-dimensional point cloud data consists of pixel point information inside the kidney contour.

[0067] Step S50: selecting a plurality of kidney stone pixel point information from the kidney three-dimensional point cloud data based on the kidney stone pixel threshold constraint condition, and determining the three-dimensional coordinates of the kidney stone according to the plurality of kidney stone pixel point information.

[0068] In a specific implementation, according to clinical experience and previous studies, kidney stones usually present a higher CT value in CT images. This is based on the fact that stones are usually harder than surrounding soft tissues and show higher density in CT scans.

[0069] The kidney stone pixel threshold constraint condition is to pre-set and select pixel points higher than the kidney stone pixel threshold (eg, 110) from the kidney three-dimensional point cloud data.

[0070] In this embodiment, the threshold condition is set to a CT value higher than 300 Hounsfield units (HU). Since the normalization process of the two-dimensional CT grayscale image is performed above, the CT value is normalized to a range of 0-255, and the image grayscale value corresponding to the CT value of the stone here is about 110. The grayscale value of each pixel point inside the obtained kidney contour C (i.e., the three-dimensional point cloud data of the kidney) is checked, and only those points with grayscale values ​​higher than 110 are retained.

[0071] It should also be noted that the pixels above the kidney stone pixel threshold (e.g. 110) need to be sorted and summarized, and the output stone three-dimensional coordinates include the location information of each stone point cloud, which is convenient for direct navigation of surgical equipment or further image reconstruction.

[0072] In this embodiment, a three-dimensional CT image of a patient with kidney stones is first obtained, and the three-dimensional CT image is sliced ​​and preprocessed to obtain a plurality of two-dimensional CT grayscale images, and then the plurality of two-dimensional CT grayscale images are input into an improved active contour model, and corresponding improved energy functional values ​​are obtained by improving the energy functional. After the improved energy functional value converges, the internal pixel point information of the kidney contour corresponding to each two-dimensional CT grayscale image is output, and the improved active contour model is constructed by the energy functional of the improved Chan-Vese model, and the improved energy functional is constructed by combining the overall energy term, the local energy fitting term and the regularization term. Then, according to the internal pixel point information of the kidney contour corresponding to each two-dimensional CT grayscale image, kidney three-dimensional point cloud data is generated, and finally, based on the kidney stone pixel threshold constraint condition, multiple kidney stone pixel point information is selected from the kidney three-dimensional point cloud data, and the three-dimensional coordinates of the kidney stone are determined according to the multiple kidney stone pixel point information. This embodiment combines texture features and local energy fitting technology to more accurately capture the boundaries of kidneys and stones. By combining the improved active contour model with multi-stage image preprocessing, the accuracy and efficiency of kidney stone recognition are greatly improved, reducing misdiagnosis and missed diagnosis.

[0073] Reference Figure 3 , Figure 3 This is a structural block diagram of the first embodiment of the system for determining the three-dimensional coordinates of kidney stones of the present invention.

[0074] like Figure 3 As shown, the three-dimensional coordinate determination system of kidney stones proposed in the embodiment of the present invention includes: The processing module 3001 is used to obtain a three-dimensional CT image of a patient with kidney stones, and perform slice preprocessing on the three-dimensional CT image to obtain a plurality of two-dimensional CT grayscale images; A kidney contour extraction module 3002 is used to input multiple two-dimensional CT grayscale images into an improved active contour model, and obtain an improved energy functional value corresponding to each two-dimensional CT grayscale image through an improved energy functional, wherein the improved active contour model is constructed by an energy functional of an improved Chan-Vese model, and the improved energy functional is constructed by a combination of an overall energy term, a local energy fitting term, and a regularization term; The kidney contour extraction module 3002 is further used to enable the improved active contour model to output the internal pixel point information of the kidney contour corresponding to each two-dimensional CT grayscale image after the improved energy functional value converges; The processing module 3001 is also used to generate kidney three-dimensional point cloud data according to the internal pixel information of the kidney contour corresponding to each two-dimensional CT grayscale image; The coordinate determination module 3003 is used to select multiple kidney stone pixel point information from the kidney three-dimensional point cloud data based on the kidney stone pixel threshold constraint condition, and determine the three-dimensional coordinates of the kidney stone based on the multiple kidney stone pixel point information.

[0075] Other embodiments or specific implementations of the system for determining the three-dimensional coordinates of kidney stones of the present invention can refer to the above-mentioned method embodiments and will not be described in detail here.

[0076] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0077] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0078] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0079] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for determining the three-dimensional coordinates of a kidney stone, characterized in that: The method for determining the three-dimensional coordinates of a kidney stone comprises the following steps: Acquire a three-dimensional CT image of a patient with kidney stones, and perform slice preprocessing on the three-dimensional CT image to obtain a plurality of two-dimensional CT grayscale images; Inputting a plurality of two-dimensional CT grayscale images into an improved active contour model, obtaining an improved energy functional value corresponding to each two-dimensional CT grayscale image through an improved energy functional, wherein the improved active contour model is constructed by an energy functional of an improved Chan-Vese model, and the improved energy functional is constructed by a combination of an overall energy term, a local energy fitting term, and a regularization term; After the improved energy functional value converges, the improved active contour model outputs the internal pixel point information of the kidney contour corresponding to each two-dimensional CT grayscale image; Generate kidney three-dimensional point cloud data according to the internal pixel information of kidney contour corresponding to each two-dimensional CT grayscale image; Based on the kidney stone pixel threshold constraint condition, multiple kidney stone pixel point information is selected from the kidney three-dimensional point cloud data, and the three-dimensional coordinates of the kidney stone are determined according to the multiple kidney stone pixel point information.

2. The method according to claim 1, characterized in that The step of inputting a plurality of two-dimensional CT grayscale images into the improved active contour model and obtaining an improved energy functional value corresponding to each two-dimensional CT grayscale image by improving the energy functional comprises: Input multiple two-dimensional CT grayscale images into the improved active contour model, and calculate the overall energy value, local energy fitting value and regularization value of each two-dimensional CT grayscale image; The improved energy functional value of each two-dimensional CT grayscale image is calculated by improving the energy functional according to the overall energy value, the local energy fitting value and the regularization value.

3. The method according to claim 2, characterized in that The step of calculating the overall energy value of each two-dimensional CT grayscale image comprises: Selecting a target CT grayscale image from a plurality of two-dimensional CT grayscale images; Determine the initial kidney contour parameters of the target CT grayscale image, and determine the grayscale value of the pixel points in each two-dimensional CT grayscale image, the average grayscale value inside the contour, the average grayscale value outside the contour, and the texture feature value of the pixel points; The overall energy value of each two-dimensional CT grayscale image is calculated through the overall energy term according to the initial kidney contour parameters, the grayscale value of the pixel point, the average grayscale value inside the contour, the average grayscale value outside the contour and the pixel point texture feature value.

4. The method according to claim 3, characterized in that The step of determining the texture feature value of the pixel points of each two-dimensional CT grayscale image comprises: Perform Gaussian blur preprocessing on each two-dimensional CT grayscale image to obtain multiple two-dimensional smooth images; Perform kidney windowing processing on multiple two-dimensional smooth images respectively to obtain multiple windowed two-dimensional images; The local texture features of each windowed two-dimensional image are extracted by LBP algorithm to obtain multiple LBP feature maps; Multiple two-dimensional smoothed images and multiple LBP feature maps are linearly combined to obtain the pixel texture feature values ​​corresponding to each two-dimensional CT grayscale image.

5. The method according to claim 2, characterized in that The step of calculating the local energy fitting value of each two-dimensional CT grayscale image includes: Determine the grayscale value of the pixel points in each two-dimensional CT grayscale image and the local area mean value of the pixel points; The local energy fitting value of each two-dimensional CT grayscale image is calculated according to the grayscale value of the pixel point and the local area mean value of the pixel point through the local energy fitting term.

6. The method according to claim 2, characterized in that The step of calculating the regularization value of each two-dimensional CT grayscale image comprises: Determine the kidney contour balance curve length and kidney contour area corresponding to each two-dimensional CT grayscale image; The regularization value of each two-dimensional CT grayscale image is calculated through a regularization term according to the length of the kidney contour balance curve and the area of ​​the kidney contour region.

7. A system for determining the three-dimensional coordinates of kidney stones, characterized in that: The three-dimensional coordinate determination system of the kidney stone comprises: A processing module, used for acquiring a three-dimensional CT image of a patient with kidney stones, and performing slice preprocessing on the three-dimensional CT image to obtain a plurality of two-dimensional CT grayscale images; A kidney contour extraction module, used for inputting a plurality of two-dimensional CT grayscale images into an improved active contour model, and obtaining an improved energy functional value corresponding to each two-dimensional CT grayscale image through an improved energy functional, wherein the improved active contour model is constructed by an energy functional of an improved Chan-Vese model, and the improved energy functional is constructed by a combination of an overall energy term, a local energy fitting term, and a regularization term; The kidney contour extraction module is further used to enable the improved active contour model to output the internal pixel point information of the kidney contour corresponding to each two-dimensional CT grayscale image after the improved energy functional value converges; The processing module is also used to generate kidney three-dimensional point cloud data according to the internal pixel information of the kidney contour corresponding to each two-dimensional CT grayscale image; The coordinate determination module is used to select multiple kidney stone pixel point information from the kidney three-dimensional point cloud data based on the kidney stone pixel threshold constraint condition, and determine the three-dimensional coordinates of the kidney stone according to the multiple kidney stone pixel point information.

8. A device for determining the three-dimensional coordinates of kidney stones, characterized in that: The device includes: a memory, a processor, and a program for determining the three-dimensional coordinates of kidney stones stored in the memory and executable on the processor, wherein the program for determining the three-dimensional coordinates of kidney stones is configured to implement the steps of the method for determining the three-dimensional coordinates of kidney stones as described in any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium stores a program for determining the three-dimensional coordinates of kidney stones, and when the program for determining the three-dimensional coordinates of kidney stones is executed by a processor, the steps of the method for determining the three-dimensional coordinates of kidney stones as described in any one of claims 1 to 6 are implemented.

Images