Tumor CT image segmentation processing method and system

Abstract

The invention relates to the technical field of tumor image analysis, in particular to a tumor CT image segmentation processing method and system, and the method comprises the following steps: adjusting image processing parameters based on a patient tumor CT image, carrying out the enhancement processing of the image, extracting various image feature information of a medical image, and generating an image enhancement processing result. According to the method, the image is enhanced by adjusting the contrast and brightness of the image, so that the visual quality of the image is improved, a clearer basis is provided for subsequent edge detection, the accuracy of edge detection is improved, the tumor region is segmented in combination with texture and color information, and the geometric features of the tumor are extracted; according to the method, the understanding of the tumor morphology is enhanced, the construction of a three-dimensional model and the calculation of the tumor volume are realized, the method is crucial to the stage evaluation of diseases and the formulation of treatment plans, and the data retrieval efficiency is improved by integrating tumor volume estimation data and key patient information and optimizing the classification and storage process of images.

Description

Technical Field

[0001] The present invention relates to the technical field of tumor image analysis, and particularly to a method and system for tumor CT image segmentation and processing. Background Art

[0002] The technical field of tumor image analysis involves analyzing and processing medical imaging data to extract information that is of great significance for tumor diagnosis and treatment. By using a variety of image processing techniques, including image processing, computer vision, and machine learning techniques, image data obtained from CT, MRI, and various medical imaging devices is analyzed, including image preprocessing, feature recognition, image segmentation, and pattern classification, aiming to assist doctors in identifying lesion areas in complex medical images, evaluating the type and stage of diseases, and helping medical professionals with diagnosis and monitoring.

[0003] Among them, the tumor image segmentation and processing method refers to using a variety of image processing steps to accurately separate tumor tissues and normal tissues from medical images, covering multiple technical steps such as image enhancement, edge detection, and region growing. Image enhancement is used to improve the image quality and make the tumor area clearer. Edge detection is used to determine the accurate boundary between the tumor and non-tumor tissues. The region growing technique gradually expands the identified tumor area according to preset criteria until the entire tumor area is completely divided, achieving accurate recognition and segmentation of the tumor area from CT images.

[0004] Traditional tumor CT image segmentation and processing methods have deficiencies in edge detection and accurate segmentation of tumor areas. In the case of highly heterogeneous tumor tissues and fuzzy tumor edges, it is difficult to accurately identify, resulting in low accuracy in tumor boundary recognition. In terms of three-dimensional reconstruction and volume calculation, there is often a lack of effective integration, and multi-angle image data cannot be fully utilized, limiting the accuracy of diagnosis and treatment planning, leading to misjudgment of tumor stages and inaccurate treatment plans in actual operations, affecting treatment effects and patient prognosis. There is a lack of accurate adjustment of image processing parameters, resulting in loss of image information, affecting the adjustment of treatment plans and the prediction of diseases. Summary of the Invention

[0005] In order to solve the technical problems existing in the prior art, embodiments of the present invention provide a method and system for tumor CT image segmentation and processing. The technical solutions are as follows:

[0006] On the one hand, a method for tumor CT image segmentation and processing is provided, and the method includes:

[0007] S1: Based on the CT image of the patient's tumor, adjust the image processing parameters to perform enhancement processing on the image, extract various image feature information of the medical image, and generate an image enhancement processing result;

[0008] S2: Based on the image enhancement processing result, perform edge detection on the image by analyzing the pixel intensity and texture features of multiple regions, extract the edge information of the image, and generate an edge information extraction record;

[0009] S3: Utilize the edge information extraction record, segment the tumor region in the image according to the texture features and color information of the image, adjust the segmentation parameters by analyzing the segmentation accuracy, extract the geometric features of the tumor, and generate a tumor segmentation image;

[0010] S4: According to the tumor segmentation image, perform three-dimensional reconstruction on the tumor based on the shape information and shooting angle of the tumor in multiple medical images of the target patient, calculate the volume of the tumor, and generate tumor volume estimation data;

[0011] S5: According to the tumor volume estimation data, classify the image based on the patient information, tumor type, location, volume, and shooting time of the target image, match the retrieval tags to optimize the retrieval efficiency, and generate an image classification and storage record.

[0012] As a further solution of the present invention, the image enhancement processing result is specifically an image enhancement parameter, a color feature extraction record, and a texture feature extraction result. The edge information extraction record includes a pixel intensity analysis result, texture feature analysis information, and image edge information. The tumor segmentation image specifically refers to a segmentation accuracy analysis result, a segmentation parameter adjustment record, and tumor geometric feature information. The tumor volume estimation data is specifically a tumor three-dimensional model, tumor volume calculation data, and image shooting angle information. The image classification and storage record includes a patient information extraction record, a tumor information data set, and a retrieval tag matching record.

[0013] As a further solution of the present invention, based on the patient's tumor CT image, the steps of adjusting the image processing parameters to perform enhancement processing on the image and extracting various image feature information of the medical image to generate an image enhancement processing result are specifically as follows:

[0014] S101: Based on the patient's tumor CT image, perform enhancement processing on the image by adjusting the contrast and brightness parameters of the image, including adjusting the contrast and brightness, and generate an image preprocessing record;

[0015] S102: Based on the image preprocessing record, extract the color features of the image to generate color feature data;

[0016] S103: Based on the color feature data, analyze and extract the texture feature information of the target medical image to generate an image enhancement processing result.

[0017] As a further solution of the present invention, based on the image enhancement processing result, by analyzing the pixel intensity and texture features of multiple regions, edge detection is performed on the image, the edge information of the image is extracted, and the steps of generating the edge information extraction record are specifically as follows:

[0018] S201: Based on the image enhancement processing result, extract the pixel intensity information of multiple regions in the image to generate a local intensity analysis result;

[0019] S202: Based on the local intensity analysis result, perform edge detection on the image, identify the edge information in the image, and generate an edge feature analysis result;

[0020] S203: Based on the edge feature analysis result, by analyzing the continuity and clarity of the edge, adjust the edge extraction parameters, and record the extracted edge information to generate an edge information extraction record.

[0021] As a further solution of the present invention, the specific formula for identifying the edge information in the image is:

[0022]

[0023] where H i represents the calculation result of the gradient direction histogram corresponding to the i-th pixel point, i is an index variable, w i is the weight of the i-th pixel point, is the partial derivative of the function f at the x i point, is the partial derivative symbol, f represents the brightness function of the image, and x i represents the position of the i-th considered pixel point.

[0024] As a further solution of the present invention, using the edge information extraction record, according to the texture features and color information of the image, the tumor region in the image is segmented, and by analyzing the segmentation accuracy, the segmentation parameters are adjusted, and the geometric features of the tumor are extracted to generate the steps of the tumor segmentation image are specifically as follows:

[0025] S301: Using the edge information extraction record, segment the image according to the texture and color information in the image, identify the tumor and non-tumor regions, and generate a preliminary tumor segmentation image;

[0026] S302: According to the preliminary tumor segmentation image, by analyzing the segmentation accuracy of the image, adjust the segmentation parameters to optimize the segmentation accuracy, and generate a segmentation parameter optimization record;

[0027] S303: Based on the segmentation parameter optimization record, segment the image and extract the geometric feature information of the tumor, including shape and boundary, to generate a tumor segmentation image.

[0028] As a further solution of the present invention, the specific formula for optimizing the segmentation accuracy by adjusting the segmentation parameters is:

[0029]

[0030] where P new represents the new segmentation parameter value, P old represents the original segmentation parameter, α represents the learning rate, P best represents the currently known best segmentation parameter, D represents the difference degree between the current segmentation effect and the target effect, and D max represents the maximum acceptable difference degree.

[0031] As a further solution of the present invention, according to the tumor segmentation image, based on the shape information and shooting angle of the tumor in multiple medical images of the target patient, the steps of performing three-dimensional reconstruction on the tumor, calculating the volume of the tumor, and generating tumor volume estimation data are specifically as follows:

[0032] S401: Based on the tumor segmentation image, analyze multiple medical images of the target patient, extract the shooting angle and geometric shape information of the tumor, and generate shape and angle analysis data;

[0033] S402: Based on the shape and angle analysis data, construct a three-dimensional model of the patient's tumor, and map color features and texture features onto the model to generate a three-dimensional tumor model;

[0034] S403: Based on the three-dimensional tumor model, calculate the volume of the tumor according to the geometric shape information to generate tumor volume estimation data.

[0035] As a further solution of the present invention, according to the tumor volume estimation data, based on the patient information, tumor type, location, volume, and shooting time of the target image, the steps of classifying the image, matching retrieval tags to optimize the retrieval efficiency, and generating an image classification and storage record are specifically as follows:

[0036] S501: Based on the tumor volume estimation data, extract the tumor type, location, volume, and shooting time information of the target medical image, and combine the personal information of the patient to generate a classification information extraction record;

[0037] S502: Based on the classification information extraction record, classify the image according to the characteristic information of the tumor and the patient information to generate an image classification result;

[0038] S503: Use the image classification result to match retrieval tags for the medical image, optimize the retrieval efficiency, and generate an image classification and storage record.

[0039] On the other hand, a tumor CT image segmentation processing system is provided. This system is applied to a tumor CT image segmentation processing method and includes:

[0040] An image preprocessing module, based on the patient's tumor CT image, enhances the image by adjusting the contrast and brightness of the image, combines feature extraction, records the color and texture features of the image, and generates an image enhancement processing result;

[0041] An edge detection module analyzes the pixel intensity and texture features of multiple regions in the image using the image enhancement processing result, identifies and extracts the edge information of the image through edge detection, and generates an edge information extraction record;

[0042] An image segmentation module uses the edge information extraction record, segments the tumor region in the image according to the texture features and color information of the image, combines the analysis of segmentation accuracy, optimizes the segmentation processing parameters, and extracts the geometric features of the tumor to generate a tumor segmentation image;

[0043] A three-dimensional modeling module analyzes multiple medical images of the patient based on the tumor segmentation image, performs three-dimensional reconstruction on the image and calculates the tumor volume according to the shooting angle of the image and the shape information of the tumor, and generates tumor volume estimation data;

[0044] A data classification module extracts patient information, tumor type, location, volume, and shooting time according to the tumor volume estimation data, classifies the image data, and matches retrieval tags to generate an image classification and storage record.

[0045] The beneficial effects brought by the technical solution provided by the embodiments of the present invention at least include:

[0046] By adjusting the contrast and brightness of the image to enhance the image, the visual quality of the image is improved, providing a clearer basis for subsequent edge detection, improving the accuracy of edge detection, segmenting the tumor region by combining texture and color information, realizing the extraction of tumor geometric features, enhancing the understanding of the tumor morphology, realizing the construction of a three-dimensional model and the calculation of tumor volume, which is crucial for the stage assessment of the disease and the formulation of treatment plans. By integrating tumor volume estimation data and key patient information, the classification and storage process of the image is optimized, and the efficiency of data retrieval is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0048] Figure 1 Schematic diagram of the working process of the present invention;

[0049] Figure 2 Refined flowchart of S1 of the present invention;

[0050] Figure 3 Refined flowchart of S2 of the present invention;

[0051] Figure 4 Refined flowchart of S3 of the present invention;

[0052] Figure 5 Refined flowchart of S4 of the present invention;

[0053] Figure 6 Refined flowchart of S5 of the present invention;

[0054] Figure 7 System flowchart of the present invention. Detailed implementation manners

[0055] The technical solutions in the present invention will be described below with reference to the accompanying drawings.

[0056] In the embodiments of the present invention, words such as "exemplarily", "for example", etc. are used to represent examples, illustrations or explanations. Any embodiment or design solution described as an "example" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, the use of the word "example" is intended to present concepts in a specific manner. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two.

[0057] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when their differences are not emphasized, the meanings they express are the same. "Of", "corresponding", and "corresponding" can sometimes be used interchangeably. It should be noted that when their differences are not emphasized, the meanings they express are the same.

[0058] In the embodiments of the present invention, sometimes subscripts such as W 1 may be written in a non-subscript form such as W1. When their differences are not emphasized, the meanings they express are the same.

[0059] To make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0060] The embodiments of the present invention provide a method for segmenting and processing tumor CT images, such as Figure 1The flowchart of the tumor CT image segmentation processing method shown. The processing flow of this method can include the following steps:

[0061] S1: Based on the patient's tumor CT image, adjust the image processing parameters to enhance the image, extract various image feature information of the medical image, and generate the image enhancement processing result;

[0062] S2: Based on the image enhancement processing result, perform edge detection on the image by analyzing the pixel intensity and texture features of multiple regions, extract the edge information of the image, and generate the edge information extraction record;

[0063] S3: Utilize the edge information extraction record, segment the tumor region in the image according to the texture feature and color information of the image, adjust the segmentation parameters by analyzing the segmentation accuracy, extract the geometric features of the tumor, and generate the tumor segmentation image;

[0064] S4: According to the tumor segmentation image, perform three-dimensional reconstruction of the tumor based on the shape information and shooting angle of the tumor in multiple medical images of the target patient, calculate the volume of the tumor, and generate the tumor volume estimation data;

[0065] S5: According to the tumor volume estimation data, classify the image according to the patient information, tumor type, location, volume, and shooting time of the target image, match the retrieval label to optimize the retrieval efficiency, and generate the image classification and storage record.

[0066] The image enhancement processing result specifically includes image enhancement parameters, color feature extraction record, texture feature extraction result. The edge information extraction record includes pixel intensity analysis result, texture feature analysis information, and image edge information. The tumor segmentation image specifically refers to the segmentation accuracy analysis result, segmentation parameter adjustment record, and tumor geometric feature information. The tumor volume estimation data specifically includes the tumor three-dimensional model, tumor volume calculation data, and image shooting angle information. The image classification and storage record includes patient information extraction record, tumor information dataset, and retrieval label matching record.

[0067] Please refer to Figure 2 , based on the patient's tumor CT image, the steps of adjusting the image processing parameters to enhance the image and extract various image feature information of the medical image to generate the image enhancement processing result are specifically as follows:

[0068] S101: Based on the patient's tumor CT image, enhance the image by adjusting the contrast and brightness parameters of the image, including adjusting contrast and brightness, and generate the image preprocessing record;

[0069] Perform preliminary processing on the image, modify the contrast and brightness parameters of the image, improve the image quality, make the tumor details in the image more clearly visible. By using image processing software such as Adobe Photoshop, ensure that the details of the dark and bright parts in the image are properly displayed. Use a high-pass filter to further enhance the edge features in the image and strengthen the contrast between the tumor and the surrounding tissues. The image parameters recorded during the process include the contrast values, brightness values, and filter setting parameters before and after adjustment. The target data is saved in the image preprocessing record for reference and reproduction in subsequent steps.

[0070] S102: Based on the image preprocessing record, extract the color feature of the image to generate color feature data;

[0071] Use color analysis tools, such as the color space conversion function in Matlab, to convert the image from the RGB color space to the HSV color space, visually analyze the color distribution and saturation in the image, identify and distinguish the tumor area from the normal tissue. In the HSV space, perform statistical analysis on the hue, saturation, and brightness, calculate the color histogram of each region, and generate color feature data. The data record includes the distribution of each main color component and its statistical characteristics, providing the basic data for the next step of texture feature extraction.

[0072] S103: Based on the color feature data, analyze and extract the texture feature information of the target medical image to generate the image enhancement processing result;

[0073] Further extract the texture information of the image through image processing tools and techniques. Use the gray-level co-occurrence matrix method to analyze the texture pattern in the image, capture the subtle changes in the texture from the local area of the image, such as the texture difference between the tumor tissue and the surrounding normal tissue. By setting different distance and angle parameters, calculate the GLCM matrix, and obtain statistical indicators such as the contrast, uniformity, entropy, and correlation of the texture from it. The target texture indicators provide quantitative texture features for the image enhancement processing result, making the tumor recognition and analysis more accurate, and further promoting the accurate diagnosis of the nature and boundary of the tumor.

[0074] Please refer to Figure 3 , based on the image enhancement processing result, by analyzing the pixel intensity and texture features of multiple regions, perform edge detection on the image to extract the edge information of the image. The specific steps for generating the edge information extraction record are as follows:

[0075] S201: Based on the image enhancement processing result, extract the pixel intensity information of multiple regions in the image to generate the local intensity analysis result;

[0076] Operate using image analysis software such as ImageJ to segment the enhanced image into multiple small regions. Each region usually contains hundreds to thousands of pixel points. For each region, calculate its average pixel intensity value, which is obtained by summing the brightness values of all pixels in the region and dividing by the total number of pixels. The standard deviation of the pixel intensity is also analyzed to evaluate the degree of variation of the pixel intensity within the region, which helps in the discrimination of edge intensity in subsequent steps. The average intensity and standard deviation data of each region are recorded and form the local intensity analysis result. The target data is extremely crucial for understanding the illumination and detail changes in the image region and provides precise input data for edge detection.

[0077] S202: Based on the local intensity analysis result, perform edge detection on the image to identify the edge information in the image and generate the edge feature analysis result;

[0078] The specific formula for identifying the edge information in the image is:

[0079]

[0080] where H i represents the calculation result of the gradient direction histogram corresponding to the i-th pixel point, i is the index variable, w i is the weight of the i-th pixel point, is the partial derivative of the function f at the x i point, is the partial derivative symbol, f represents the brightness function of the image, and x i represents the position of the i-th pixel point considered.

[0081] Formula:

[0082]

[0083] Detailed explanation of the formula and the derivation process of formula calculation:

[0084] The formula is used to calculate the gradient direction histogram of each pixel point in the image, and the result is used to detect the edges in the image and determine the boundary between tumor and non-tumor tissues;

[0085] Meaning of parameters and set values:

[0086] w i is the weight coefficient, representing the influence of the i-th pixel point. Assume w i = 0.5;

[0087] x i is the pixel point position in the image;

[0088] f(x i ) is the brightness value of the pixel point x i , assume f(xi ) = 120;

[0089] Represents the luminance gradient of pixel x i , assuming f(x i ) has luminances of 125 and 115 respectively in its adjacent pixels x i+1 and x i-1 ; substituting the parameters into the formula for calculation:

[0090] Substitute the parameters into the formula for calculation:

[0091] H i = 0.5·(5) 2 = 0.5·25 = 12.5;

[0092] The result H i = 12.5 indicates that the edge feature of pixel point x i in the image is relatively significant. The value reflects a relatively high luminance change rate starting from this point. Such high values in edge detection indicate possible boundary regions, ensuring more accurate segmentation of tumor and non - tumor tissues.

[0093] S203: Based on the edge feature analysis results, by analyzing the continuity and clarity of the edges, adjust the edge extraction parameters, and record the extracted edge information to generate an edge information extraction record;

[0094] Use the edge continuity algorithm to determine the continuity of the edges. If broken or blurred edges are detected, the algorithm will automatically adjust the edge extraction parameters, such as adjusting the low and high thresholds in the Canny algorithm, to improve the accuracy of edge detection. The adjusted parameters are recorded and used to repeat the edge detection process until the preset edge continuity and clarity standards are met. The result of this step, namely the edge information extraction record, details all the adjusted parameters and the improved edge information, providing an important basis for ensuring the clear separation of the tumor region and non - tumor region in subsequent image processing steps.

[0095] Please refer to Figure 4 , using the edge information extraction record, according to the texture features and color information of the image, segment the tumor region in the image, and by analyzing the accuracy of the segmentation, adjust the segmentation parameters, extract the geometric features of the tumor, and the steps to generate the tumor segmentation image are as follows:

[0096] S301: Using the edge information extraction record, segment the image according to the texture and color information in the image, identify the tumor and non - tumor regions, and generate a preliminary tumor segmentation image;

[0097] The region growing algorithm is used, which is a method for image segmentation based on similarity criteria. The region growing algorithm starts from seed points and adds the pixels adjacent to the seed points to the seed region. If the target pixel has similar texture or color features to the seed point, this process is iterated until all eligible pixels are included. During the implementation process, the position of the seed points is determined based on edge information. Usually, the pixels located inside the obvious tumor region are selected as seed points. Region growing is performed based on the gray value, color, and texture information of the pixels until the preset growth boundary, that is, the boundary line provided by edge detection, is reached. The preliminary tumor segmentation image generated by this process clearly shows the boundary between the tumor and non-tumor regions, providing an accurate basis for subsequent analysis.

[0098] S302: According to the preliminary tumor segmentation image, by analyzing the accuracy of the image segmentation, adjust the segmentation parameters to optimize the segmentation accuracy, and generate a record of optimized segmentation parameters;

[0099] The specific formula for adjusting the segmentation parameters to optimize the segmentation accuracy is:

[0100]

[0101] Among them, P new represents the new segmentation parameter value, P old represents the original segmentation parameter, α represents the learning rate, P best represents the currently known best segmentation parameter, D represents the difference degree between the current segmentation effect and the target effect, D max represents the maximum acceptable difference degree.

[0102] Formula:

[0103]

[0104] Detailed explanation of the formula and the derivation process of the formula calculation:

[0105] The formula is used to optimize the parameters in the image segmentation algorithm and adjust the segmentation parameters to improve the segmentation accuracy;

[0106] Meaning and setting values of the parameters:

[0107] P old is the original segmentation parameter, assumed to be 0.2;

[0108] P best is the currently best segmentation parameter, assumed to be 0.5;

[0109] α is the learning rate, assumed to be 0.05;

[0110] D is the difference degree between the current segmentation effect and the target effect, assumed to be 15;

[0111] Dmax is the maximum acceptable degree of difference, assumed to be 100;

[0112] Substitute the parameters into the formula for calculation:

[0113]

[0114] P new = 0.2 + 0.05·0.3·0.15;

[0115] P new = 0.2 + 0.00225;

[0116] P new = 0.20225;

[0117] The result 0.20225 is the new value after parameter adjustment. The calculation process is used to improve the ability to distinguish the differences in the image edge and texture, providing accurate tumor recognition and boundary definition for subsequent analysis.

[0118] S303: Based on the optimized record of the segmentation parameters, segment the image and extract the geometric feature information of the tumor, including the shape and boundary, to generate a tumor segmentation image;

[0119] Computational geometry techniques are used to define and quantify the shape parameters of the tumor, including using geometric moments and contour analysis methods to calculate the area, perimeter, and shape complexity of the tumor. Through object geometry calculations, the physical morphology of the tumor is accurately described, providing important information about the tumor growth pattern and spread path for clinical applications. The boundary extraction uses a gradient-based method to determine the tumor edge by identifying the regions with the most significant changes in color and texture information. The execution of the target process ensures that the tumor area can be accurately segmented from the CT image, and every geometric feature of the tumor is detailedly recorded, providing a reliable data basis for subsequent treatment planning and monitoring.

[0120] Please refer to Figure 5 , according to the tumor segmentation image, based on the shape information and shooting angles of the tumor in multiple medical images of the target patient, the steps for three-dimensional reconstruction of the tumor and calculation of the tumor volume to generate tumor volume estimation data are as follows:

[0121] S401: Based on the tumor segmentation image, analyze multiple medical images of the target patient, extract the shooting angles and geometric shape information of the tumor, and generate shape and angle analysis data;

[0122] Extract key shape and angle information from multiple medical images. Using image registration technology, align CT images taken at different times to ensure that the tumor shape information extracted from each image is comparable. The technology relies on positioning markers in the images or prominent features of the tumor itself to ensure precise alignment between images. The aligned image data is processed by feature extraction algorithms, such as gradient-based shape descriptors. The object descriptors quantify the geometric features of the tumor in the image, such as boundary curvature, area, and volume. Each extracted feature is recorded and analyzed to evaluate the changes and development of the tumor shape. Integrate the shape and angle information of all images to generate shape and angle analysis data. The target data reflects the morphological changes of the tumor at different times and provides the necessary input information for subsequent 3D modeling.

[0123] S402: Based on the shape and angle analysis data, construct a 3D model of the patient's tumor and map color features and texture features onto the model to generate a 3D tumor model;

[0124] Apply the voxelization method to convert 2D image data into a point cloud in 3D space. Each point cloud represents the voxel of the tumor at a specific position, and its attributes include color and texture information, which are all inherited from the original CT image. Through 3D rendering techniques, such as volume rendering or surface rendering, convert the point cloud into a continuous 3D surface model. The model shows the external shape of the tumor and accurately maps the texture features of the tumor, enabling doctors to observe the tumor structure from multiple angles. The generated 3D tumor model is an all-round and dynamic expression of the tumor morphology and provides a powerful visual aid for clinical practice.

[0125] S403: Based on the 3D tumor model, calculate the volume of the tumor according to the geometric shape information to generate tumor volume estimation data;

[0126] Adopt a geometric volume calculation method to estimate the volume of the tumor. Utilize the geometric data of the 3D model, such as side lengths, areas, and voxel densities, and calculate the volume of the entire tumor through mathematical formulas. The calculation formulas may include 3D integrals or summing up voxels of complex shapes. Each voxel represents a small cube, and its volume is the cube of its side length. By summing up the volumes of all voxels, the total volume of the entire tumor is obtained. This calculation method not only provides a quantification of the tumor size but also reflects the internal volume distribution of the tumor. The generated tumor volume estimation data is of great clinical significance for evaluating the growth rate and treatment response of the tumor and helps doctors formulate more precise treatment plans.

[0127] Please refer to Figure 6 , according to the tumor volume estimation data, classify the images based on the patient information, tumor type, location, volume, and shooting time of the target image, and match retrieval tags to optimize the retrieval efficiency. The specific steps for generating the image classification and storage record are as follows:

[0128] S501: Based on the tumor volume estimation data, extract the tumor type, location, volume, and shooting time information of the target medical image, and combine with the patient's personal information to generate a classification information extraction record;

[0129] Use data mining techniques, such as decision tree algorithms, to analyze and classify image data. The decision tree selects the most effective attributes for branching by evaluating the information gain of various indicators, gradually refining the classification criteria, retrieves relevant data from the patient's personal information database, such as age, gender, and medical history, and associates it with the image data to enhance the accuracy of classification. Each piece of data is verified and cleaned to ensure the quality of the input data. Through the target steps, the system generates a complete classification information extraction record, which includes the detailed classification data of each image and its corresponding patient personal information, providing basic data for subsequent image processing and analysis.

[0130] S502: Based on the classification information extraction record, classify the images according to the tumor's characteristic information and patient information to generate an image classification result;

[0131] Use the support vector machine in machine learning technology for further classification processing of tumor images. The support vector machine performs class separation in a high-dimensional space by constructing one or more hyperplanes, including a training phase and a classification phase. In the training phase, a labeled image set is used to train the model, and in the classification phase, the model is used to classify new images. The SVM model evaluates the feature vectors of various images, such as texture, shape, and boundary information, and identifies the tumor type according to the set classification criteria. Each image will be labeled with a specific tumor type to generate a detailed image classification result. The target result not only includes the category of the tumor but also is accurate to specific features of the tumor, such as grade and prognostic indicators.

[0132] S503: Utilize the image classification result to match retrieval tags for the medical image, optimize the retrieval efficiency, and generate an image classification and storage record;

[0133] Adopt label matching technology to optimize the retrieval and storage process of medical images. The label system creates a set of descriptive tags for each image based on the image's classification result and medical records. The target tags cover tumor type, location, volume, and specific medical indicators. In the label generation process, natural language processing technology is applied to convert medical terms and classification information into keywords that are easy to retrieve. The image storage system dynamically adjusts the storage location and backup strategy of the image according to the relevance and search frequency of the tags. In this way, not only the management efficiency of the medical image library is improved, but also the speed and accuracy of the medical workflow are improved, enabling doctors and researchers to quickly find the required image materials.

[0134] Please refer to Figure 7, a tumor CT image segmentation processing system, which is used to execute the above-mentioned tumor CT image segmentation processing method. The system includes:

[0135] The image preprocessing module is based on the patient's tumor CT image. By adjusting the contrast and brightness of the image, the image is enhanced, combined with feature extraction, the color and texture features of the image are recorded, and the image enhancement processing result is generated.

[0136] The edge detection module uses the image enhancement processing result to analyze the pixel intensity and texture features of multiple regions in the image. Through edge detection, the edge information of the image is identified and extracted, and the edge information extraction record is generated.

[0137] The image segmentation module uses the edge information extraction record. According to the texture features and color information of the image, the tumor region in the image is segmented. Combining the analysis of the segmentation accuracy, the segmentation processing parameters are optimized, and the geometric features of the tumor are extracted to generate the tumor segmentation image.

[0138] The three-dimensional modeling module is based on the tumor segmentation image. It analyzes multiple medical images of the patient. According to the shooting angle of the image and the shape information of the tumor, the image is three-dimensionally reconstructed and the tumor volume is calculated to generate the tumor volume estimation data.

[0139] The data classification module extracts patient information, tumor type, location, volume, and shooting time according to the tumor volume estimation data, classifies the image data, and matches the retrieval tags to generate the image classification and storage record.

[0140] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions according to the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, or magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0141] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. Additionally, the character " / " in this document generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be understood specifically with reference to the context before and after.

[0142] In the present invention, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or its similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0143] It should be understood that in various embodiments of the present invention, the magnitudes of the sequence numbers of the above processes do not imply the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0144] Those of ordinary skill in the art will realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0145] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described devices, apparatuses, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0146] In several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0147] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0148] In addition, the functional units in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0149] When the above-mentioned function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0150] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A tumor CT image segmentation processing method, characterized in that: The method comprises: S1: Based on the patient's tumor CT image, adjust the image processing parameters to enhance the image, extract various image feature information of the medical image, and generate the image enhancement processing result; S2: Based on the image enhancement processing result, by analyzing the pixel intensity and texture features of multiple regions, edge detection is performed on the image, edge information of the image is extracted, and an edge information extraction record is generated; S3: using the edge information extraction record, segmenting the tumor area in the image according to the texture features and color information of the image, adjusting the segmentation parameters by analyzing the accuracy of the segmentation, extracting the geometric features of the tumor, and generating a tumor segmentation image; S4: performing three-dimensional reconstruction of the tumor according to the tumor segmentation image and the shape information and shooting angles of the tumor in multiple medical images of the target patient, and calculating the volume of the tumor to generate tumor volume estimation data; S5: Based on the tumor volume estimation data, the images are classified according to the patient information, tumor type, location, volume and shooting time of the target image, the retrieval tags are matched to optimize the retrieval efficiency, and image classification and storage records are generated.

2. The tumor CT image segmentation processing method according to claim 1, characterized in that: The image enhancement processing results specifically include image enhancement parameters, color feature extraction records, and texture feature extraction results. The edge information extraction records include pixel intensity analysis results, texture feature analysis information, and image edge information. The tumor segmentation image specifically refers to segmentation accuracy analysis results, segmentation parameter adjustment records, and tumor geometric feature information. The tumor volume estimation data specifically includes a tumor three-dimensional model, tumor volume calculation data, and image shooting angle information. The image classification and storage records include patient information extraction records, tumor information data sets, and retrieval label matching records.

3. The tumor CT image segmentation processing method according to claim 1, characterized in that: Based on the patient's tumor CT image, the image processing parameters are adjusted to enhance the image, and various image feature information of the medical image is extracted to generate the image enhancement processing results. The specific steps are as follows: S101: Based on the patient's tumor CT image, the image is enhanced by adjusting the contrast and brightness parameters of the image, including adjusting the contrast and brightness, and generating an image preprocessing record; S102: extracting color features of the image based on the image preprocessing record to generate color feature data; S103: Based on the color feature data, analyze and extract texture feature information of the target medical image to generate an image enhancement processing result.

4. The tumor CT image segmentation processing method according to claim 1, characterized in that: Based on the image enhancement processing result, the steps of performing edge detection on the image and extracting edge information of the image by analyzing the pixel intensity and texture features of multiple regions and generating edge information extraction records are specifically as follows: S201: extracting pixel intensity information of multiple regions in the image based on the image enhancement processing result, and generating a local intensity analysis result; S202: Based on the local intensity analysis result, edge detection is performed on the image to identify edge information in the image and generate an edge feature analysis result; S203: Based on the edge feature analysis result, by analyzing the continuity and clarity of the edge, adjusting edge extraction parameters, and recording the extracted edge information to generate an edge information extraction record.

5. The tumor CT image segmentation processing method according to claim 4, characterized in that: The specific formula for identifying edge information in the image is: Among them, H i Represents the calculation result of the gradient direction histogram corresponding to the i-th pixel point, i is the index variable, w i is the weight of the i-th pixel, is in x i The partial derivative of the function f at a point, is the symbol of partial derivative, f represents the brightness function of the image, x i Represents the position of the i-th considered pixel.

6. The tumor CT image segmentation processing method according to claim 1, characterized in that: The steps of using the edge information extraction record to segment the tumor area in the image according to the texture features and color information of the image, adjusting the segmentation parameters by analyzing the accuracy of the segmentation, extracting the geometric features of the tumor, and generating the tumor segmentation image are as follows: S301: using the edge information extraction record, segmenting the image according to texture and color information in the image, identifying tumor and non-tumor areas, and generating a preliminary tumor segmentation image; S302: According to the preliminary tumor segmentation image, by analyzing the accuracy of image segmentation, adjusting segmentation parameters to optimize segmentation accuracy, and generating a segmentation parameter optimization record; S303: Based on the segmentation parameter optimization record, segment the image and extract the geometric feature information of the tumor, including shape and boundary, to generate a tumor segmentation image.

7. The tumor CT image segmentation processing method according to claim 6, characterized in that: The specific formula for adjusting the segmentation parameters to optimize the segmentation accuracy is: Among them, P new represents the new segmentation parameter value, P old represents the original segmentation parameter, α represents the learning rate, and P best represents the currently known best segmentation parameter, D represents the difference between the current segmentation effect and the target effect, and D max Represents the maximum acceptable difference.

8. The tumor CT image segmentation processing method according to claim 1, characterized in that: According to the tumor segmentation image, according to the shape information and shooting angles of the tumor in multiple medical images of the target patient, the tumor is three-dimensionally reconstructed, and the volume of the tumor is calculated. The steps of generating tumor volume estimation data are specifically as follows: S401: Based on the tumor segmentation image, analyzing multiple medical images of the target patient, extracting shooting angles and geometric shape information of the tumor, and generating shape and angle analysis data; S402: constructing a three-dimensional model of the patient's tumor based on the shape and angle analysis data, and mapping color features and texture features onto the model to generate a three-dimensional tumor model; S403: Based on the three-dimensional tumor model and according to the geometric shape information, the volume of the tumor is calculated to generate tumor volume estimation data.

9. The tumor CT image segmentation processing method according to claim 1, characterized in that: According to the tumor volume estimation data, the images are classified according to the patient information, tumor type, location, volume and shooting time of the target image, and the retrieval efficiency is optimized by matching the retrieval tags. The steps of generating image classification and storage records are specifically as follows: S501: extracting the tumor type, location, volume and shooting time information of the target medical image based on the tumor volume estimation data, and generating a classification information extraction record in combination with the patient's personal information; S502: extracting records based on the classification information, classifying images according to characteristic information of the tumor and patient information, and generating image classification results; S503: Using the image classification results, matching retrieval tags for medical images, optimizing retrieval efficiency, and generating image classification and storage records.

10. A tumor CT image segmentation processing system, characterized in that: According to the tumor CT image segmentation processing method according to any one of claims 1 to 9, the system comprises: The image preprocessing module is based on the patient's tumor CT image. It enhances the image by adjusting the contrast and brightness of the image, combines feature extraction, records the color and texture features of the image, and generates image enhancement processing results. The edge detection module uses the image enhancement processing result to analyze the pixel intensity and texture characteristics of multiple areas in the image, identifies and extracts the edge information of the image through edge detection, and generates an edge information extraction record; The image segmentation module uses the edge information extraction record to segment the tumor area in the image according to the texture features and color information of the image, optimizes the segmentation processing parameters in combination with the segmentation accuracy analysis, and extracts the geometric features of the tumor to generate a tumor segmentation image; The three-dimensional modeling module analyzes multiple medical images of the patient based on the tumor segmentation image, performs three-dimensional reconstruction on the image and calculates the tumor volume according to the shooting angle of the image and the shape information of the tumor, and generates tumor volume estimation data; The data classification module extracts patient information, tumor type, location, volume and shooting time according to the tumor volume estimation data, classifies the image data, matches the retrieval tags, and generates image classification and storage records.

Images