Tumor minimally invasive treatment system based on pathological data analysis
Through the minimally invasive tumor treatment and processing system based on pathological data analysis, three-dimensional tumor image data is generated and overlapping is calculated, the problem of difficulty in resection of complex shapes is solved, accurate identification of tumor residues and reasonable allocation of medical resources is achieved, and diagnosis is improved and treatment efficiency.
Patent Information
- Application Number
- CN202411213380.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
During the minimally invasive process of tumors, it is difficult to accurately remove tumors of complex shapes, making it difficult to detect unremoved residual tumors, resulting in wasted medical resources and delayed treatment time.
The minimally invasive tumor treatment processing system based on pathological data analysis generates three-dimensional tumor image data through data image units, models, scanning units, shape overlap units and shape priority units, calculates overlap and prioritizes residual tumors to ensure accurate identification. and rationally allocate medical resources.
It improves the identification accuracy and diagnostic sensitivity of tumor residual conditions, rationally allocates medical resources, reduces overexamination of low-priority tumors, saves time and reduces medical costs.
Smart Images

Figure CN119092150A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of tumor image processing, and in particular to a minimally invasive tumor treatment processing system based on pathological data analysis. Background Art
[0002] Minimally invasive tumor treatment based on pathological data analysis collects detailed information about the tumor from pathological sections and imaging examination data sources. The detailed information includes the size, shape, location, tissue type and gene mutation of the tumor. Due to the complex and irregular shapes of tumors in different patients, with protrusions, depressions, uneven or unclear boundaries, the complex shape makes it more difficult to completely remove the tumor. It is difficult to accurately determine the entire boundary of the tumor, resulting in residual tumors that have not been completely removed after minimally invasive treatment of brain tumors. When there are multiple residual tumors that are difficult to find in the area or boundary after minimally invasive treatment of the tumor, it is impossible to accurately prioritize the residual tumors, resulting in unreasonable allocation of medical resources and waste of medical resources. At the same time, if low-priority residual tumors are over-examined, it will lead to delays in treatment time and increased medical costs. Therefore, we provide a minimally invasive tumor treatment system based on pathological data analysis. Summary of the invention
[0003] The purpose of the present invention is to provide a minimally invasive tumor treatment system based on pathological data analysis to solve the problems raised in the above background technology.
[0004] To achieve the above-mentioned object, the present invention provides a minimally invasive tumor treatment processing system based on pathological data analysis, comprising a data image unit, a model and scanning unit, a shape overlapping unit, and a shape priority unit;
[0005] The data image unit acquires the pathological data to generate a three-dimensional tissue model, and performs segmentation and surface extraction on the image data in the pathological data, and then takes three vertices of a triangle from the extracted surface data, and generates point cloud data by taking the three-dimensional tumor boundary coordinate points of the vertices, and generates three-dimensional tumor image data before minimally invasive surgery based on the point cloud data;
[0006] The model and scanning unit is used to receive data from the data image unit, input the three-dimensional tumor boundary coordinate points into the three-dimensional tissue model to connect and obtain a complete tumor shape, and generate a three-dimensional model of the tumor and its tissue, perform minimally invasive surgery on the patient according to the tumor and tissue three-dimensional model data, and then perform CT and MRI scanning on the patient after minimally invasive surgery;
[0007] The shape overlapping unit is used to receive data from the data image unit and the model and scanning unit, and to cut out the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery, align the three-dimensional tumor boundary coordinate points of the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery, generate an intersection area and a union area, divide the generated intersection area and the union area into a plurality of grid areas to calculate the overlap, and when the calculated overlap is lower than the standard overlap, it indicates that there is residual tumor in the grid area;
[0008] The shape priority unit is used to receive data from the shape overlap unit and the data image unit, obtain the mapped three-dimensional coordinate points through the space transformation matrix, connect the mapped three-dimensional coordinate points to generate the three-dimensional shape of the residual tumor, decompose the three-dimensional shape of the residual tumor into a number of tetrahedrons and sum up the three-dimensional shape volume of the residual tumor, and then prioritize the three-dimensional shape of the residual tumor from high to low according to the volume, and prioritize the three-dimensional shape of the residual tumor according to the result of the priority sorting.
[0009] As a further improvement of the technical solution, the data image unit includes an image conversion module, a surface extraction module and a tumor image module;
[0010] The image conversion module obtains CT and MRI image data before minimally invasive surgery in the pathological data. Medical images (CT and MRI scans) are composed of many continuous two-dimensional slices, and performs image segmentation on the CT and MRI image data before minimally invasive surgery to identify the tumor area, and then uses a binarization method to convert the segmented image data into a binary image of the slice. The binary image has a small amount of data and high calculation and storage efficiency, which can speed up image processing and improve calculation efficiency;
[0011] Pathological data include CT and MRI imaging data, tissue section images, histological features, and pathological diagnosis information;
[0012] The implementation principle of using the binarization method to convert the segmented image into a sliced binary image:
[0013] Collect segmented image data Fg tx , and extract the segmented image data Fg tx The pixel grayscale value at the position is determined according to the extracted pixel grayscale value I(x,y). The specific algorithm formula is:
[0014]
[0015] When the extracted pixel grayscale value I(x, y) is greater than or equal to the preset fixed grayscale value T, the tumor area is marked as 1;
[0016] When the extracted pixel grayscale value I(x,y) is less than the preset fixed grayscale value T, the tumor area is marked as 0;
[0017] The surface extraction module is used to receive the binary image data of the slices in the image conversion module, and then stack the binary images of the slices in sequence by stacking slices to form three-dimensional volume data, and then extract the surface from the three-dimensional volume data by using Marching Cubes algorithm;
[0018] The tumor image module is used to receive the surface data extracted in the surface extraction module and the pathological data in the image conversion module, generate a three-dimensional tissue model according to the case data, use a three-dimensional grid to take the three vertices of each triangle from the extracted surface data, the extracted surface data is composed of a number of triangles, each triangle has three vertices, each vertex represents a three-dimensional tumor boundary coordinate point on the tumor surface, and generate point cloud data by taking the three-dimensional tumor boundary coordinate points of the vertices, and then use point cloud processing technology to generate three-dimensional tumor image data before minimally invasive based on the point cloud data. The point cloud data can accurately represent objects in three-dimensional space, and can generate high-resolution three-dimensional tumor images, clearly showing the shape, boundaries and structure of the tumor.
[0019] As a further improvement of the technical solution, the model and scanning unit includes a tumor and tissue model module and a post-injury scanning module;
[0020] The tumor and tissue model module is used to receive the three-dimensional tumor boundary coordinate points and the tissue three-dimensional model in the tumor image module, input the three-dimensional tumor boundary coordinate points into the tissue three-dimensional model, and the tissue three-dimensional model connects the three-dimensional tumor boundary coordinate points to obtain a complete tumor shape, and automatically generates a three-dimensional model of the tumor and its tissue, which can clearly show the tumor boundary and its relationship with surrounding structures, and provide more detailed anatomical information;
[0021] The post-injury scanning module is used to receive the tumor and tissue three-dimensional model data in the tumor and tissue model module, perform minimally invasive tumor on the patient according to the tumor and tissue three-dimensional model data, and then perform CT and MRI scans on the patient after minimally invasive tumor to obtain CT and MRI image data after minimally invasive tumor.
[0022] As a further improvement of the technical solution, the shape overlapping unit includes a shape cutout module, an intersection and union module and an overlap degree module;
[0023] The shape cutout module is used to receive the three-dimensional tumor boundary coordinate points and the three-dimensional tumor image data before minimally invasive surgery in the tumor image module, receive the CT and MRI image data after minimally invasive surgery in the post-injury scanning module, and perform image segmentation based on the CT and MRI image data after minimally invasive surgery to generate the three-dimensional tumor image data after minimally invasive surgery, mark the three-dimensional tumor boundary coordinate points on the three-dimensional tumor image data after minimally invasive surgery and the three-dimensional tumor image data before minimally invasive surgery, and connect the marked coordinate points to form a shape on the three-dimensional tumor image data after minimally invasive surgery and the three-dimensional tumor image data before minimally invasive surgery, and use the cutout segmentation method to cut out the connected shape to cut out the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery, so as to accurately extract the tumor area from the image, ensure that only the tumor itself is focused on rather than the surrounding tissue, and provide higher segmentation accuracy;
[0024] The intersection and union module is used to receive the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery in the shape cutout module, receive the tissue three-dimensional model and the three-dimensional tumor boundary coordinate points in the tumor image module, input the three-dimensional tumor boundary coordinate points, the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery into the tissue three-dimensional model, the tissue three-dimensional model amplifies the three-dimensional tumor boundary coordinate points of the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery to a preset value and stops, and then aligns the three-dimensional tumor boundary coordinate points of the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery to generate an intersection area and a union area;
[0025] The implementation principle of generating intersection areas and union areas:
[0026] First, collect the tumor shape area image W before minimally invasive surgery pre and the tumor shape area image W after minimally invasive surgery post Calculate the intersection area and the union area to obtain the calculated intersection area W overlap and the union area W union , the specific algorithm formula is:
[0027] W overlap =W pre ∩W post ;
[0028] W union =W pre ∪W post ;
[0029] Among them, this formula is used to generate intersection areas and union areas. By comparing the tumor shapes before and after minimally invasive surgery, the minimally invasive resection range can be planned more accurately, thereby reducing miscutting and omissions.
[0030] As a further improvement of the technical solution, the overlap module is used to receive the intersection area and the union area generated in the intersection and union module, divide the generated intersection area and the union area into a plurality of grid areas using a grid division method, and obtain the corresponding three-dimensional coordinate points of the grid areas, and calculate the overlap using the intersection grid area and the corresponding union grid area. When the calculated overlap is lower than the standard overlap, it indicates that there is residual tumor in the grid area, which can more accurately identify the residual situation of the tumor and improve the sensitivity and accuracy of diagnosis;
[0031] The implementation principle of calculating the overlap in the overlap module is:
[0032] First collect the grid area W of the intersection overlap,sub and the mesh area W of the union union,sub Calculate the overlap and get the calculated overlap C d , the specific algorithm formula is:
[0033]
[0034] Among them, C di Refers to the calculated i-th overlap C d , which is used to calculate the overlap and to determine whether there is residual tumor in the grid area.
[0035] As a further improvement of the technical solution, the shape priority unit includes a residual tumor shape module and a priority module;
[0036] The residual tumor shape module is used to receive the three-dimensional coordinate points of the grid area in the overlap module and the three-dimensional tumor boundary coordinate points in the tumor image module, and convert the three-dimensional coordinate points of the grid area into the three-dimensional tumor boundary coordinate points in the tumor shape area image before minimally invasive surgery through the space transformation matrix to obtain the mapped three-dimensional coordinate points, and then connect all the mapped three-dimensional coordinate points to generate the three-dimensional shape of the residual tumor. By connecting the mapped three-dimensional coordinate points, a clear three-dimensional shape of the residual tumor can be generated, so that doctors, patients and medical teams can intuitively understand the shape of the tumor;
[0037] The implementation principle of using the space transformation matrix to obtain the mapped three-dimensional coordinates:
[0038] Get the three-dimensional coordinates of the grid area (x sbu ,y sbu ,z sbu ) and the three-dimensional tumor boundary coordinate points (x, y, z) in the tumor shape area image before minimally invasive surgery calculate the mapped three-dimensional coordinate points. The specific algorithm formula is:
[0039]
[0040] Among them, the formula is used to calculate the three-dimensional coordinate points after mapping. After minimally invasive tumor, the transformed three-dimensional coordinate points are compared with the image data after minimally invasive tumor, which is helpful to evaluate the residual tumor after minimally invasive tumor;
[0041] The priority module is used to receive the three-dimensional shape of the residual tumor in the residual tumor shape module, decompose the three-dimensional shape of the residual tumor into a plurality of tetrahedrons, obtain the corresponding four vertex coordinates, calculate the volume of the tetrahedron according to the corresponding four vertex coordinates, and then sum up the three-dimensional shape volume of the residual tumor according to the calculated tetrahedron volumes, and at the same time organize the three-dimensional model to compare the three-dimensional shapes of all residual tumors in the aligned three-dimensional images in sequence according to the volumes, sort the three-dimensional shapes of the residual tumors from high to low priority according to the comparison results, and then prioritize the three-dimensional shapes of the residual tumors according to the results of the priority sorting. By prioritizing the residual tumors, medical resources (such as surgical time, drugs, and equipment) can be reasonably allocated, which can avoid waste of resources and concentrate limited resources on the most needed areas. At the same time, priority sorting can reduce excessive inspection of low-priority tumor areas, thereby saving time and reducing medical costs.
[0042] As a further improvement of this technical solution, the implementation principle of calculating the volume of the tetrahedron in the priority module is:
[0043] Get the corresponding four vertex coordinates (x 1 ,y 1 ,z 1 )、(x 2 ,y 2 ,z 2 )、(x 3 ,y 3 ,z 3 )、(x 4 ,y 4 ,z 4 ) Calculate the volume of the tetrahedron and obtain the calculated volume V of the tetrahedron i , the specific algorithm formula is:
[0044]
[0045] Among them, (x i1 ,y i1 ,z i1 )、(x i2 ,y i2 ,z i2 )、(x i3 ,y i3 ,z i3 )、(x i4 ,y i4 ,z i4) are the coordinates of the four vertices of the i-th tetrahedron, is a constant factor used to calculate the volume of a tetrahedron. 1 is a constant used to calculate the determinant of a matrix so that the formula can correctly represent the volume of the tetrahedron. This formula is used to calculate the volume of a tetrahedron. In medical imaging, the volume calculation of a tetrahedron can be used to accurately measure the volume of a tumor or other lesion.
[0046] The principle of summing up the three-dimensional shape and volume of the residual tumor is as follows:
[0047] Get the calculated tetrahedron volume V i Sum up the three-dimensional shape volume C of the residual tumor l , the specific algorithm formula is:
[0048]
[0049] Here, n refers to the nth tetrahedron. The formula is used to sum up the three-dimensional shape volume of the residual tumor. Tumors usually have irregular shapes. The tetrahedron decomposition method can effectively handle these complex geometric shapes, unlike some other volume calculation methods that are limited to regular geometric shapes.
[0050] Compared with the prior art, the present invention has the following beneficial effects:
[0051] 1. In the minimally invasive tumor treatment processing system for pathological data analysis, the overlap module is used to receive the intersection area and union area generated in the intersection and union module, divide the generated intersection area and union area into several grid areas using a grid division method, and obtain the corresponding three-dimensional coordinate points of the grid areas, and calculate the overlap using the intersection grid area and the corresponding union grid area. When the calculated overlap is lower than the standard overlap, it indicates that there is residual tumor in the grid area, which can more accurately identify the residual situation of the tumor and improve the sensitivity and accuracy of diagnosis.
[0052] 2. In the minimally invasive tumor treatment processing system of the pathological data analysis, the residual tumor shape module is used to receive the three-dimensional coordinate points of the grid area in the overlap module and the three-dimensional tumor boundary coordinate points in the tumor image module, and convert the three-dimensional coordinate points of the grid area into the three-dimensional tumor boundary coordinate points in the tumor shape area image before minimally invasive surgery through the spatial transformation matrix to obtain the mapped three-dimensional coordinate points, and then connect all the mapped three-dimensional coordinate points to generate the three-dimensional shape of the residual tumor. By connecting the mapped three-dimensional coordinate points, a clear three-dimensional shape of the residual tumor can be generated, so that doctors, patients and medical teams can intuitively understand the shape of the tumor.
[0053] 3. In the minimally invasive tumor treatment processing system of the pathological data analysis, the priority module is used to receive the three-dimensional shape of the residual tumor in the residual tumor shape module, decompose the three-dimensional shape of the residual tumor into several tetrahedrons, and obtain the corresponding four vertex coordinates, calculate the volume of the tetrahedron according to the corresponding four vertex coordinates, and then sum up the three-dimensional shape volume of the residual tumor according to the calculated tetrahedron volume, and at the same time organize the three-dimensional model to compare the three-dimensional shapes of all residual tumors in the aligned three-dimensional images in turn according to the volume, and sort the three-dimensional shapes of the residual tumors from high to low priority according to the comparison results, and then prioritize the three-dimensional shapes of the residual tumors according to the results of the priority sorting. By prioritizing the residual tumors, medical resources (such as operation time, drugs, and equipment) can be reasonably allocated, which can avoid waste of resources and concentrate limited resources on the most needed places. At the same time, priority sorting can reduce excessive inspection of low-priority tumor areas, thereby saving time and reducing medical costs. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] Figure 1 is an overall block diagram of the present invention;
[0055] Figure 2 A block diagram of a data image unit of the present invention;
[0056] Figure 3 The model and scanning unit block diagram of the present invention;
[0057] Figure 4 It is a block diagram of the shape overlapping unit of the present invention;
[0058] Figure 5 It is a block diagram of the shape priority unit of the present invention.
[0059] The meaning of each number in the figure is:
[0060] 1. Data image unit; 11. Image conversion module; 12. Surface extraction module; 13. Tumor image module;
[0061] 2. Model and scanning unit; 21. Tumor and tissue model module; 22. Post-traumatic scanning module;
[0062] 3. Shape overlap unit; 31. Shape cutout module; 32. Intersection and union module; 33. Overlap degree module;
[0063] 4. Shape priority unit; 41. Residual tumor shape module; 42. Priority module. DETAILED DESCRIPTION
[0064] The following will be combined with the accompanying drawings in 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.
[0065] Example 1
[0066] The present invention provides a minimally invasive tumor treatment system based on pathological data analysis. Figure 1-Figure 5 , including a data image unit 1, a model and scanning unit 2, a shape overlapping unit 3, and a shape priority unit 4;
[0067] The data image unit 1 obtains the pathological data to generate a three-dimensional tissue model, and performs segmentation and surface extraction on the image data in the pathological data, and then takes three vertices of a triangle from the extracted surface data, and generates point cloud data by taking the three-dimensional tumor boundary coordinate points of the vertices, and generates three-dimensional tumor image data before minimally invasive surgery based on the point cloud data;
[0068] The model and scanning unit 2 is used to receive the data in the data image unit 1, input the three-dimensional tumor boundary coordinate points into the three-dimensional tissue model to connect and obtain the complete tumor shape, and generate a three-dimensional model of the tumor and its tissue, perform minimally invasive tumor treatment on the patient according to the three-dimensional model data of the tumor and tissue, and then perform CT and MRI scanning on the patient after minimally invasive tumor treatment;
[0069] The shape overlapping unit 3 is used to receive data from the data image unit 1 and the model and scanning unit 2, and to cut out the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery, align the three-dimensional tumor boundary coordinate points of the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery, generate an intersection area and a union area, divide the generated intersection area and the union area into a number of grid areas to calculate the overlap, and when the calculated overlap is lower than the standard overlap, it indicates that there is residual tumor in the grid area;
[0070] The shape priority unit 4 is used to receive data from the shape overlap unit 3 and the data image unit 1, obtain the mapped three-dimensional coordinate points through the space transformation matrix, connect the mapped three-dimensional coordinate points to generate the three-dimensional shape of the residual tumor, decompose the three-dimensional shape of the residual tumor into several tetrahedrons and sum up the three-dimensional shape volume of the residual tumor, and then prioritize the three-dimensional shape of the residual tumor from high to low according to the volume, and prioritize the three-dimensional shape of the residual tumor according to the result of the priority sorting.
[0071] The data image unit 1 includes an image conversion module 11, a surface extraction module 12 and a tumor image module 13;
[0072] The image conversion module 11 obtains the CT and MRI image data before minimally invasive surgery in the pathological data. Medical images (CT and MRI scans) are composed of many continuous two-dimensional slices. The CT and MRI image data before minimally invasive surgery are segmented to identify the tumor area. The segmented image data is then converted into a binary image of the slice using a binarization method. The binary image has a small amount of data and high calculation and storage efficiency, which can speed up image processing and improve calculation efficiency.
[0073] Pathological data include CT and MRI imaging data, tissue section images, histological features, and pathological diagnosis information;
[0074] The implementation principle of using the binarization method to convert the segmented image into a sliced binary image:
[0075] Collect segmented image data Fg tx , and extract the segmented image data Fg tx The pixel grayscale value at the position is determined according to the extracted pixel grayscale value I(x,y). The specific algorithm formula is:
[0076]
[0077] When the extracted pixel grayscale value I(x, y) is greater than or equal to the preset fixed grayscale value T, the tumor area is marked as 1;
[0078] When the extracted pixel grayscale value I(x,y) is less than the preset fixed grayscale value T, the tumor area is marked as 0;
[0079] The surface extraction module 12 is used to receive the binary image data of the slices in the image conversion module 11, and then stack the binary images of the slices in sequence by stacking slices to form three-dimensional volume data, and then extract the surface from the three-dimensional volume data by using Marching Cubes algorithm;
[0080] The tumor image module 13 is used to receive the surface data extracted in the surface extraction module 12 and the pathological data in the image conversion module 11, generate a three-dimensional tissue model according to the case data, use a three-dimensional grid to take three vertices of each triangle from the extracted surface data, the extracted surface data is composed of a plurality of triangles, each triangle has three vertices, each vertex represents a three-dimensional tumor boundary coordinate point on the tumor surface, generate point cloud data by taking the three-dimensional tumor boundary coordinate points of the vertices, and then use the point cloud processing technology to generate three-dimensional tumor image data before minimally invasive according to the point cloud data. The point cloud data can accurately represent objects in three-dimensional space, and can generate high-resolution three-dimensional tumor images, clearly showing the shape, boundary and structure of the tumor;
[0081] The model and scanning unit 2 includes a tumor and tissue model module 21 and a post-traumatic scanning module 22;
[0082] The tumor and tissue model module 21 is used to receive the three-dimensional tumor boundary coordinate points and the tissue three-dimensional model in the tumor image module 13, input the three-dimensional tumor boundary coordinate points into the tissue three-dimensional model, and the tissue three-dimensional model connects the three-dimensional tumor boundary coordinate points to obtain a complete tumor shape, and automatically generates a three-dimensional model of the tumor and its tissue, which can clearly show the tumor boundary and its relationship with the surrounding structures, and provide more detailed anatomical information;
[0083] The post-injury scanning module 22 is used to receive the tumor and tissue three-dimensional model data in the tumor and tissue model module 21, perform minimally invasive tumor surgery on the patient according to the tumor and tissue three-dimensional model data, and then perform CT and MRI scans on the patient after minimally invasive tumor surgery to obtain CT and MRI image data after minimally invasive surgery.
[0084] The shape overlapping unit 3 includes a shape cutout module 31, an intersection and union module 32 and an overlap module 33;
[0085] The shape cutout module 31 is used to receive the three-dimensional tumor boundary coordinate points and the three-dimensional tumor image data before minimally invasive surgery in the tumor image module 13, receive the CT and MRI image data after minimally invasive surgery in the post-injury scanning module 22, and perform image segmentation according to the CT and MRI image data after minimally invasive surgery to generate the three-dimensional tumor image data after minimally invasive surgery, mark the three-dimensional tumor boundary coordinate points on the three-dimensional tumor image data after minimally invasive surgery and the three-dimensional tumor image data before minimally invasive surgery, and connect the marked coordinate points to form a shape on the three-dimensional tumor image data after minimally invasive surgery and the three-dimensional tumor image data before minimally invasive surgery, and use the cutout segmentation method to cut out the connected shape to cut out the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery, so as to accurately extract the tumor area from the image, ensure that only the tumor itself is focused on rather than the surrounding tissue, and provide higher segmentation accuracy;
[0086] The intersection and union module 32 is used to receive the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery in the shape cutout module 31, receive the tissue three-dimensional model and the three-dimensional tumor boundary coordinate points in the tumor image module 13, input the three-dimensional tumor boundary coordinate points, the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery into the tissue three-dimensional model, the tissue three-dimensional model amplifies the three-dimensional tumor boundary coordinate points of the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery to a preset value and stops, and then aligns the three-dimensional tumor boundary coordinate points of the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery to generate an intersection area and a union area;
[0087] The implementation principle of generating intersection areas and union areas:
[0088] First, collect the tumor shape area image W before minimally invasive surgery pre and the tumor shape area image W after minimally invasive surgery post Calculate the intersection area and the union area to obtain the calculated intersection area W overlap and the union area W union , the specific algorithm formula is:
[0089] W overlap =W pre ∩W post ;
[0090] W union =W pre ∪W post ;
[0091] Among them, this formula is used to generate the intersection area and the union area. By comparing the tumor shape before and after minimally invasive surgery, the minimally invasive resection range can be planned more accurately, thereby reducing mis-cutting and omissions;
[0092] The overlap module 33 is used to receive the intersection area and the union area generated in the intersection and union module 32, divide the generated intersection area and the union area into a plurality of grid areas using a grid division method, and obtain the corresponding three-dimensional coordinate points of the grid areas, and calculate the overlap using the intersection grid area and the corresponding union grid area. When the calculated overlap is lower than the standard overlap, it indicates that there is residual tumor in the grid area, which can more accurately identify the residual situation of the tumor and improve the sensitivity and accuracy of diagnosis;
[0093] The implementation principle of calculating the overlap in the overlap module 33 is as follows:
[0094] First collect the grid area W of the intersection overlap,sub and the mesh area W of the union union,sub Calculate the overlap and get the calculated overlap C d , the specific algorithm formula is:
[0095]
[0096] Among them, C di Refers to the calculated i-th overlap C d ,This formula is used to calculate the overlap, which is used to determine whether there is residual tumor in the grid area;
[0097] The shape priority unit 4 includes a residual tumor shape module 41 and a priority module 42;
[0098] The residual tumor shape module 41 is used to receive the three-dimensional coordinate points of the grid area in the overlap module 33 and the three-dimensional tumor boundary coordinate points in the tumor image module 13, and convert the three-dimensional coordinate points of the grid area into the three-dimensional tumor boundary coordinate points in the tumor shape area image before minimally invasive surgery through the space transformation matrix, and obtain the mapped three-dimensional coordinate points, and then connect all the mapped three-dimensional coordinate points to generate the three-dimensional shape of the residual tumor. By connecting the mapped three-dimensional coordinate points, a clear three-dimensional shape of the residual tumor can be generated, so that doctors, patients and medical teams can intuitively understand the shape of the tumor;
[0099] The implementation principle of using the space transformation matrix to obtain the mapped three-dimensional coordinates:
[0100] Get the three-dimensional coordinates of the grid area (x sbu ,y sbu ,z sbu ) and the three-dimensional tumor boundary coordinate points (x, y, z) in the tumor shape area image before minimally invasive surgery calculate the mapped three-dimensional coordinate points. The specific algorithm formula is:
[0101]
[0102] Among them, the formula is used to calculate the three-dimensional coordinate points after mapping. After minimally invasive tumor, the transformed three-dimensional coordinate points are compared with the image data after minimally invasive tumor, which is helpful to evaluate the residual tumor after minimally invasive tumor;
[0103] The priority module 42 is used to receive the three-dimensional shape of the residual tumor in the residual tumor shape module 41, decompose the three-dimensional shape of the residual tumor into a plurality of tetrahedrons, obtain the corresponding four vertex coordinates, calculate the volume of the tetrahedron according to the corresponding four vertex coordinates, and then sum up the three-dimensional shape volume of the residual tumor according to the calculated tetrahedron volume, and at the same time organize the three-dimensional model to compare the three-dimensional shapes of all residual tumors in the aligned three-dimensional images in turn according to the volume, sort the three-dimensional shapes of the residual tumors from high to low priority according to the comparison results, and then check the three-dimensional shapes of the residual tumors according to the priority sorting results. By sorting the priorities of the residual tumors, medical resources (such as operation time, drugs, and equipment) can be reasonably allocated, so as to avoid waste of resources and concentrate limited resources on the most needed places. At the same time, priority sorting can reduce excessive inspection of low-priority tumor areas, thereby saving time and reducing medical costs;
[0104] The implementation principle of calculating the volume of the tetrahedron in priority module 42 is:
[0105] Get the corresponding four vertex coordinates (x 1 ,y 1 ,z 1 )、(x 2,y 2 ,z 2 )、(x 3 ,y 3 ,z 3 )、(x 4 ,y 4 ,z 4 ) Calculate the volume of the tetrahedron and obtain the calculated volume V of the tetrahedron i , the specific algorithm formula is:
[0106]
[0107] Among them, (x i1 ,y i1 ,z i1 )、(x i2 ,y i2 ,z i2 )、(x i3 ,y i3 ,z i3 )、(x i4 ,y i4 ,z i4 ) are the coordinates of the four vertices of the i-th tetrahedron, is a constant factor used to calculate the volume of a tetrahedron. 1 is a constant used to calculate the determinant of a matrix so that the formula can correctly represent the volume of the tetrahedron. This formula is used to calculate the volume of a tetrahedron. In medical imaging, the volume calculation of a tetrahedron can be used to accurately measure the volume of a tumor or other lesion.
[0108] The principle of summing up the three-dimensional shape and volume of the residual tumor is as follows:
[0109] Get the calculated tetrahedron volume V i Sum up the three-dimensional shape volume C of the residual tumor l , the specific algorithm formula is:
[0110]
[0111] Here, n refers to the nth tetrahedron. The formula is used to sum up the three-dimensional shape volume of the residual tumor. Tumors usually have irregular shapes. The tetrahedron decomposition method can effectively handle these complex geometric shapes, unlike some other volume calculation methods that are limited to regular geometric shapes.
[0112] Usage process:
[0113] The overlap module 33 is used to receive the intersection area and the union area generated in the intersection and union module 32 and divide them into several grid areas, and obtain the corresponding three-dimensional coordinate points of the grid areas at the same time, and use the intersection grid area and the corresponding union grid area to calculate the overlap and determine the presence of residual tumors, and transmit the three-dimensional coordinate points of the grid areas to the residual tumor shape module 41. The residual tumor shape module 41 is used to receive the three-dimensional coordinate points of the grid areas in the overlap module 33 and the three-dimensional tumor boundary coordinate points in the tumor image module 13, and convert the three-dimensional coordinate points of the grid areas into three-dimensional tumor boundary coordinate points in the tumor shape area image before minimally invasive surgery through the space transformation matrix, and obtain the mapped three-dimensional coordinate points, and then all the mapped three-dimensional coordinate points are converted into the three-dimensional tumor boundary coordinate points in the tumor shape area image before minimally invasive surgery. The three-dimensional coordinate points are connected to generate a three-dimensional shape of the residual tumor, and the three-dimensional shape of the residual tumor is transmitted to the priority module 42. The priority module 42 is used to receive the three-dimensional shape of the residual tumor in the residual tumor shape module 41 and decompose it into a number of tetrahedrons, obtain the corresponding four vertex coordinates, calculate the volume of the tetrahedron according to the corresponding four vertex coordinates, and sum up the three-dimensional shape volume of the residual tumor. At the same time, the three-dimensional model is organized to compare the three-dimensional shapes of all residual tumors in the aligned three-dimensional images in turn according to the volume, and the three-dimensional shapes of the residual tumors are sorted from high to low priority according to the comparison results, and then the three-dimensional shapes of the residual tumors are prioritized according to the priority sorting results.
[0114] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. A minimally invasive tumor treatment system based on pathological data analysis, characterized by: It includes a data image unit (1), a model and scanning unit (2), a shape overlapping unit (3), and a shape priority unit (4); The data image unit (1) acquires pathological data to generate a three-dimensional tissue model, and performs segmentation and surface extraction on image data in the pathological data, then takes three vertices of a triangle from the extracted surface data, generates point cloud data by taking the three-dimensional tumor boundary coordinate points of the vertices, and generates three-dimensional tumor image data before minimally invasive surgery based on the point cloud data; The model and scanning unit (2) is used to receive data from the data image unit (1), input the three-dimensional tumor boundary coordinate points into the three-dimensional tissue model to connect and obtain a complete tumor shape, and generate a three-dimensional model of the tumor and its tissue, perform minimally invasive tumor surgery on the patient according to the tumor and tissue three-dimensional model data, and then perform CT and MRI scanning on the patient after minimally invasive tumor surgery; The shape overlapping unit (3) is used to receive data from the data image unit (1) and the model and scanning unit (2), and to extract the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery, align the three-dimensional tumor boundary coordinate points of the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery, generate an intersection area and a union area, divide the generated intersection area and the union area into a plurality of grid areas to calculate the overlap, and when the calculated overlap is lower than the standard overlap, it indicates that there is residual tumor in the grid area; The shape priority unit (4) is used to receive data from the shape overlap unit (3) and the data image unit (1), obtain the mapped three-dimensional coordinate points through the space transformation matrix, connect the mapped three-dimensional coordinate points to generate the three-dimensional shape of the residual tumor, decompose the three-dimensional shape of the residual tumor into a plurality of tetrahedrons and sum up the three-dimensional shape volume of the residual tumor, and then prioritize the three-dimensional shape of the residual tumor from high to low according to the volume, and perform priority screening on the three-dimensional shape of the residual tumor according to the priority ranking result.
2. The minimally invasive tumor treatment processing system for pathological data analysis according to claim 1, characterized in that: The data image unit (1) comprises an image conversion module (11), a surface extraction module (12) and a tumor image module (13); The image conversion module (11) obtains CT and MRI image data before minimally invasive surgery in the pathological data, performs image segmentation, identifies the tumor area, and then uses a binarization method to convert the segmented image data into a binary image of the slice; The surface extraction module (12) is used to receive the binary image data of the slices in the image conversion module (11), and then stack the binary images of the slices in sequence using stacked slices to form three-dimensional volume data, and then extract the surface from the three-dimensional volume data using the Marching Cubes algorithm; The tumor image module (13) is used to receive the surface data extracted in the surface extraction module (12) and the pathological data in the image conversion module (11), generate a three-dimensional tissue model according to the case data, use a three-dimensional grid to extract three vertices of each triangle from the extracted surface data, generate point cloud data by taking the three-dimensional tumor boundary coordinate points of the vertices, and then use point cloud processing technology to generate three-dimensional tumor image data before minimally invasive surgery according to the point cloud data.
3. The minimally invasive tumor treatment processing system for pathological data analysis according to claim 2, characterized in that: The model and scanning unit (2) comprises a tumor and tissue model module (21) and a post-traumatic scanning module (22); The tumor and tissue model module (21) is used to receive the three-dimensional tumor boundary coordinate points and the tissue three-dimensional model in the tumor image module (13), and input them into the three-dimensional model. The tissue three-dimensional model connects the three-dimensional tumor boundary coordinate points to obtain a complete tumor shape, and generates a tumor and tissue three-dimensional model; The post-injury scanning module (22) is used to receive the tumor and tissue three-dimensional model data in the tumor and tissue model module (21), perform minimally invasive surgery on the patient according to the tumor and tissue three-dimensional model data, and then perform CT and MRI scans on the patient after minimally invasive surgery.
4. The minimally invasive tumor treatment processing system for pathological data analysis according to claim 3, characterized in that: The shape overlapping unit (3) comprises a shape cutout module (31), an intersection and union module (32) and an overlap degree module (33); The shape cutout module (31) is used to receive the three-dimensional tumor boundary coordinate points and the three-dimensional tumor image data before minimally invasive surgery in the tumor image module (13), receive the CT and MRI image data after minimally invasive surgery in the post-injury scanning module (22), and perform image segmentation based on the CT and MRI image data after minimally invasive surgery to generate the three-dimensional tumor image data after minimally invasive surgery, mark the three-dimensional tumor boundary coordinate points on the three-dimensional tumor image data after minimally invasive surgery and the three-dimensional tumor image data before minimally invasive surgery, and connect the marked coordinate points to form a shape, and use a cutout segmentation method to cut out the connected shape to obtain the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery; The intersection and union module (32) is used to receive the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery in the shape cutout module (31), receive the tissue three-dimensional model and the three-dimensional tumor boundary coordinate points in the tumor image module (13), input the three-dimensional tumor boundary coordinate points, the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery into the tissue three-dimensional model, the tissue three-dimensional model amplifies the three-dimensional tumor boundary coordinate points of the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery to a preset value and stops, and then aligns the three-dimensional tumor boundary coordinate points of the tumor shape area image before minimally invasive surgery and the tumor shape area image after minimally invasive surgery to generate an intersection area and a union area.
5. The minimally invasive tumor treatment processing system for pathological data analysis according to claim 4, characterized in that: The overlap module (33) is used to receive the intersection area and the union area generated in the intersection and union module (32), divide the generated intersection area and the union area into a plurality of grid areas using a grid division method, and obtain the corresponding three-dimensional coordinate points of the grid areas, and calculate the overlap using the intersection grid area and the corresponding union grid area. When the calculated overlap is lower than the standard overlap, it indicates that there is residual tumor in the grid area.
6. The minimally invasive tumor treatment processing system for pathological data analysis according to claim 5, characterized in that: The implementation principle of calculating the overlap degree in the overlap degree module (33) is: First collect the grid area W of the intersection overlap,sub and the mesh area W of the union union,sub Calculate the overlap and get the calculated overlap C d , the specific algorithm formula is: Among them, C di Refers to the calculated i-th overlap C d .
7. The minimally invasive tumor treatment processing system for pathological data analysis according to claim 4, characterized in that: The shape priority unit (4) comprises a residual tumor shape module (41) and a priority module (42); The residual tumor shape module (41) is used to receive the three-dimensional coordinate points of the grid area in the overlap module (33) and the three-dimensional tumor boundary coordinate points in the tumor image module (13), convert the three-dimensional coordinate points of the grid area into the three-dimensional tumor boundary coordinate points in the tumor shape area image before minimally invasive surgery through a spatial transformation matrix, obtain the mapped three-dimensional coordinate points, and then connect all the mapped three-dimensional coordinate points to generate the three-dimensional shape of the residual tumor; The priority module (42) is used to receive the three-dimensional shape of the residual tumor in the residual tumor shape module (41), decompose the three-dimensional shape of the residual tumor into a plurality of tetrahedrons, obtain the corresponding four vertex coordinates, calculate the volume of the tetrahedron according to the corresponding four vertex coordinates and sum up the three-dimensional shape volume of the residual tumor, and at the same time organize the three-dimensional model to compare the three-dimensional shapes of the residual tumor in the aligned three-dimensional images in sequence according to the volume, and according to the comparison results, prioritize the three-dimensional shapes of the residual tumor from high to low according to the volume, and then prioritize the three-dimensional shapes of the residual tumor according to the priority sorting results.
8. The minimally invasive tumor treatment processing system for pathological data analysis according to claim 7, characterized in that: The implementation principle of calculating the volume of the tetrahedron in the priority module (42) is: Get the corresponding four vertex coordinates (x1, y1, z1), (x2, y2, z2), (x3, y3, z3), (x4, y4, z4) to calculate the volume of the tetrahedron and get the calculated volume V of the tetrahedron i , the specific algorithm formula is: Among them, (x i1 ,y i1 ,z i1 )、(x i2 ,y i2 ,z i2 )、(x i3 ,y i3 ,z i3 )、(x i4 ,y i4 ,z i4 ) are the coordinates of the four vertices of the i-th tetrahedron, is the constant factor used to calculate the volume of the tetrahedron, 1 is a constant.
