Preoperative adaptive matching of tumor morphology auxiliary planning method and system

Through medical imaging data processing and tumor morphology matching model optimization, an accurate tumor resection plan is designed, which solves the problem of inability to adapt to personalized tumor morphology in traditional methods, and improves the accuracy and efficiency of the surgery.

CN118280530BActive Publication Date: 2024-11-08THE FIRST MEDICAL CENT CHINESE PLA GENERAL HOSPITAL
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202410404562.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-07
Publication Date
2024-11-08
Estimated Expiration
2044-04-07

AI Technical Summary

Technical Problem

Traditional tumor surgical planning methods cannot adapt to the personalized tumor morphology of different patients, resulting in positioning errors and surgical damage, and it is especially difficult to deal with irregularly shaped tumors.

Method used

Tumor image data is obtained through medical imaging equipment, denoising and feature extraction is performed, and a pre-trained tumor morphology matching model is used to match similar tumor morphology models from the model library, optimize the model and design the resection scheme, and verify the resection scheme using computer simulation technology feasibility.

Benefits of technology

Accurate tumor resection surgery planning based on individual differences in patients is achieved, which improves surgical efficiency and accuracy and reduces surgical damage.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118280530B_ABST
    Figure CN118280530B_ABST
Patent Text Reader

Abstract

The present invention relates to artificial intelligence technology, and discloses an auxiliary planning method for adaptively matching tumor morphology before surgery, including: obtaining image data and tumor location information of a patient's tumor through medical imaging equipment, denoising and extracting features from the image data, matching in a tumor morphology model library according to the extracted tumor morphology features using a pre-trained tumor morphology matching model to obtain a first tumor morphology model, optimizing the model according to the tumor image data to obtain a second tumor morphology model, designing a tumor resection plan, using computer simulation technology to simulate tumor resection according to the second tumor morphology model and the resection plan, and verifying the feasibility of the resection plan and optimizing the resection plan according to the simulation results. The present invention also proposes an auxiliary planning system for adaptively matching tumor morphology before surgery. The present invention can perform accurate tumor resection surgery planning according to individual differences of patients.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a preoperative adaptive matching tumor morphology auxiliary planning method and system. Background Art

[0002] In tumor surgery, accurate tumor positioning and morphology matching are the key to improving surgical efficiency and reducing surgical damage. However, traditional tumor surgery planning methods are often based on fixed templates and cannot adapt to the personalized morphology of tumors in different patients. This may lead to positioning errors during surgery and poor imaging of surgical effects.

[0003] Although there are some tumor recognition and positioning technologies based on image processing, most of these technologies can only process tumors with regular shapes. It is difficult to accurately identify and locate irregular tumors, especially those with wide infiltration range and irregular shapes. At the same time, existing technologies also lack the function of adaptive matching of tumor morphology, and cannot accurately plan tumor resection surgery according to individual differences of patients. Summary of the invention

[0004] The present invention provides a preoperative auxiliary planning method and system for adaptively matching tumor morphology, the main purpose of which is to solve the problem that accurate tumor resection surgery planning cannot be performed according to individual differences of patients.

[0005] To achieve the above object, the present invention provides a preoperative adaptive matching tumor morphology auxiliary planning method, comprising:

[0006] Obtaining imaging data and tumor location information of the patient's tumor through medical imaging equipment;

[0007] Performing denoising processing on the image data to obtain preprocessed image data;

[0008] Extracting features from the preprocessed image data using a feature extraction algorithm to obtain tumor image features, wherein extracting features from the preprocessed image data using a feature extraction algorithm to obtain tumor image features includes:

[0009] The feature extraction algorithm is expressed using the following formula:

[0010]

[0011] Among them, L(x, y) nrepresents the nth local feature of the tumor image feature, (x, y) represents the coordinates of the pixel center point of the nth local feature, P is the gray value of the pixel center point, G represents the gray value of the neighborhood point of the pixel center point, S is a binary function, when (GP) is greater than or equal to 0, S returns 1, otherwise it returns 0, and M represents the number of neighborhood points of the pixel center point;

[0012] Using a pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features to obtain a first tumor morphology model, and optimizing the first tumor morphology model according to the tumor image data to obtain a second tumor morphology model;

[0013] designing a resection plan according to the second tumor morphology model and the tumor location information;

[0014] Computer simulation technology is used to perform tumor resection simulation based on the second tumor morphology model and the resection plan to obtain simulation results, and the feasibility of the resection plan is verified and the resection plan is optimized based on the simulation results.

[0015] Specifically, the image data includes the size, shape, contour, density and adhesion relationship of the patient's tumor with surrounding tissues.

[0016] Specifically, the denoising process is performed on the image data to obtain pre-processed image data, including:

[0017] Performing noise analysis on the image data to obtain a noise type;

[0018] Selecting a denoising algorithm according to the noise type;

[0019] Performing denoising processing on the image data using the denoising algorithm to obtain a preliminary denoised image;

[0020] If the preliminary denoising image does not meet the expected denoising effect, the denoising algorithm is replaced, and the step of denoising the image data using the denoising algorithm is re-executed to obtain the preliminary denoising image;

[0021] If the preliminary denoising image meets the expected denoising effect, the image data is saved to obtain pre-processed image data.

[0022] Specifically, before using the pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features, the method further includes:

[0023] Obtain tumor imaging data of different tumor patients to obtain a tumor imaging dataset containing a large number of tumor images;

[0024] Extracting features of the tumor images in the tumor image data set to obtain tumor features;

[0025] Using three-dimensional modeling software to sequentially establish a three-dimensional model of the tumor according to the tumor images in the tumor image data set to obtain a tumor morphology model data set;

[0026] The tumor morphology models in the tumor morphology model data set are labeled with corresponding tumor features, and the labeled tumor morphology models are stored in a pre-established database to obtain a tumor morphology model library.

[0027] Specifically, before using the pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features to obtain the first tumor morphology model, the method further includes:

[0028] Acquiring a tumor image dataset from a pre-established tumor morphology model library, and performing feature extraction on tumor images in the tumor image dataset to obtain a tumor image feature dataset;

[0029] Using a neural network model, matching tumor morphology models in the tumor morphology model library is performed in turn according to the tumor image features in the tumor image feature data set to obtain a matching result;

[0030] Calculating the matching accuracy rate according to the real tumor morphology model preset by the tumor image feature and the matching result;

[0031] If the matching accuracy is less than a preset accuracy threshold, the weight parameters and bias items of the neural network model are adjusted, and the step of using the neural network model to match the tumor morphology model in the tumor morphology model library according to the tumor image features in the tumor image feature data set in turn to obtain a matching result is re-executed;

[0032] If the matching accuracy is greater than or equal to a preset accuracy threshold, the model training is confirmed to be completed and a tumor morphology matching model is obtained.

[0033] Specifically, the method of using the pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features to obtain the first tumor morphology model includes:

[0034] Using the tumor morphology matching model to perform data conversion processing on the tumor image features to obtain a feature matrix;

[0035] Using the tumor morphology matching model to perform a convolution operation on the feature matrix through a convolution layer to obtain convolution features;

[0036] Using the tumor morphology matching model, performing a maximum pooling operation on the convolutional features through a pooling layer to obtain a pooling feature;

[0037] The tumor morphology matching model is used to match tumor morphology models in the tumor morphology model library through a fully connected layer according to the pooling features to obtain a first tumor morphology model.

[0038] Specifically, optimizing the first tumor morphology model according to the tumor image data to obtain a second tumor morphology model includes:

[0039] adjusting the structure of the first tumor morphology model according to the tumor image data to obtain an optimized model;

[0040] Determining whether the accuracy of the optimization model meets the expected requirements;

[0041] If the accuracy of the optimized model does not meet the expected requirements, re-performing the step of adjusting the structure of the first tumor morphology model according to the tumor image data to obtain the optimized model;

[0042] If the accuracy of the optimized model meets the expected requirements, the model optimization is confirmed to be completed and the second tumor morphology model is obtained.

[0043] Specifically, after optimizing the first tumor morphology model according to the tumor image data to obtain the second tumor morphology model, the second tumor morphology model is stored in the tumor morphology model library.

[0044] Specifically, the using of computer simulation technology to perform tumor resection simulation according to the second tumor morphology model and the resection plan to obtain simulation results includes:

[0045] importing the second tumor morphology model into the computer simulation software;

[0046] Setting simulation parameters according to the second tumor morphology model and the resection plan;

[0047] The second tumor morphology model is subjected to simulated resection according to the resection plan and simulation parameters to obtain a simulation result.

[0048] In order to solve the above problems, the present invention also provides a preoperative adaptive matching tumor morphology auxiliary planning system, the system comprising:

[0049] A data acquisition module, used to obtain the imaging data and tumor location information of the patient's tumor through medical imaging equipment;

[0050] A data processing module is used to perform denoising on the image data to obtain pre-processed image data, and to extract features from the pre-processed image data using a feature extraction algorithm to obtain tumor image features, wherein the step of extracting features from the pre-processed image data using a feature extraction algorithm to obtain tumor image features includes:

[0051] The feature extraction algorithm is expressed using the following formula:

[0052] L(x,y) n =∑S(GP)*(M-1)

[0053] Among them, L(x, y) n represents the nth local feature of the tumor image feature, (x, y) represents the coordinates of the pixel center point of the nth local feature, P is the gray value of the pixel center point, G represents the gray value of the neighborhood point of the pixel center point, S is a binary function, when (GP) is greater than or equal to 0, S returns 1, otherwise it returns 0, and M represents the number of neighborhood points of the pixel center point;

[0054] a model matching module, configured to use a pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features to obtain a first tumor morphology model, and optimize the first tumor morphology model according to the tumor image data to obtain a second tumor morphology model;

[0055] A design optimization module is used to design a resection plan based on the second tumor morphology model and the tumor location information, use computer simulation technology to perform tumor resection simulation based on the second tumor morphology model and the resection plan, obtain simulation results, verify the feasibility of the resection plan based on the simulation results, and optimize the resection plan.

[0056] The embodiment of the present invention obtains the image data and tumor location information of the patient's tumor through medical imaging equipment, denoises the image data and extracts features, uses a pre-trained tumor morphology matching model to match the extracted tumor morphology features in the tumor morphology model library to obtain a first tumor morphology model, optimizes the model according to the tumor image data to obtain a second tumor morphology model, designs a tumor resection plan, uses computer simulation technology to simulate tumor resection according to the second tumor morphology model and the resection plan, verifies the feasibility of the resection plan according to the simulation results and optimizes the resection plan. Therefore, the auxiliary planning method and system for preoperative adaptive matching of tumor morphology proposed in the present invention can solve the problem of not being able to perform accurate tumor resection surgery planning according to individual differences of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1A schematic diagram of a flow chart of a method for assisting planning of preoperative adaptive matching of tumor morphology provided by an embodiment of the present invention;

[0058] Figure 2 A schematic diagram of a tumor morphology matching model training process provided by an embodiment of the present invention;

[0059] Figure 3 A schematic diagram of a process of matching a tumor morphology model according to tumor image features using a tumor morphology matching model provided by an embodiment of the present invention;

[0060] Figure 4 This is a functional module diagram of a preoperative adaptive matching tumor morphology auxiliary planning system provided by an embodiment of the present invention.

[0061] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0062] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0063] The embodiment of the present application provides a preoperative auxiliary planning method for adaptively matching tumor morphology. The execution subject of the auxiliary planning method for adaptively matching tumor morphology before surgery includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided in the embodiment of the present application. In other words, the auxiliary planning method for adaptively matching tumor morphology before surgery can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks (Content De li very Network, CDN), and big data and artificial intelligence platforms.

[0064] Reference Figure 1 FIG. 1 is a flow chart of a method for assisting planning of preoperative adaptively matching tumor morphology provided by an embodiment of the present invention. In this embodiment, the method for assisting planning of preoperative adaptively matching tumor morphology includes:

[0065] S1. Obtain imaging data and tumor location information of the patient's tumor through medical imaging equipment.

[0066] In the embodiments of the present invention, the medical imaging equipment includes but is not limited to X-ray machines, CT scanners, magnetic resonance imaging devices, and ultrasonic diagnostic devices. These devices can help doctors obtain image data of tumors, thereby diagnosing and treating tumors more accurately. It is necessary to select equipment based on the actual situation of the patient and the equipment conditions of the hospital, or use multiple devices in combination.

[0067] In the embodiment of the present invention, it is necessary to use the medical imaging device to scan the patient at multiple angles to obtain more comprehensive and accurate imaging data of the tumor.

[0068] In an embodiment of the present invention, the image data includes information such as the size, shape, contour, density, and adhesion relationship of the patient's tumor with surrounding tissues. This information can help doctors determine the nature, growth pattern, and spread of the tumor, thereby formulating a more accurate treatment plan. For example, CT scans can provide the CT value of the tumor, that is, the density difference between the diseased tissue and the normal tissue, thereby helping doctors determine the nature and degree of enhancement of the tumor. MRI can provide more detailed soft tissue information, which helps doctors better understand the relationship between the tumor and the surrounding tissues.

[0069] In the embodiment of the present invention, the image data of the patient's tumor and the tumor location information are obtained by medical imaging equipment, which provides necessary data for subsequent planning.

[0070] S2. Perform denoising processing on the image data to obtain preprocessed image data.

[0071] In the embodiment of the present invention, the denoising process is performed on the image data to obtain pre-processed image data, including:

[0072] Performing noise analysis on the image data to obtain a noise type;

[0073] Selecting a denoising algorithm according to the noise type;

[0074] Performing denoising processing on the image data using the denoising algorithm to obtain a preliminary denoised image;

[0075] If the preliminary denoising image does not meet the expected denoising effect, the denoising algorithm is replaced, and the step of denoising the image data using the denoising algorithm is re-executed to obtain the preliminary denoising image;

[0076] If the preliminary denoising image meets the expected denoising effect, the image data is saved to obtain pre-processed image data.

[0077] In the embodiment of the present invention, common noise types include Gaussian noise, salt and pepper noise, speckle noise, etc.

[0078] In the embodiment of the present invention, common denoising algorithms include median filtering, mean filtering, Gaussian filtering, bilateral filtering, etc.

[0079] In the embodiment of the present invention, by performing denoising processing on the image data, preprocessed image data is obtained, the quality of the image data is improved, the features of the image data are enhanced, and the speed of subsequent processing of the image data is increased.

[0080] S3. Using a feature extraction algorithm to extract features from the preprocessed image data to obtain tumor image features.

[0081] In the embodiment of the present invention, the feature extraction of the pre-processed image data may be performed by using LBP (Local Binary Pattern) algorithm.

[0082] In the embodiment of the present invention, by performing feature extraction on the pre-processed image data, the pre-processed image data can be simplified into a small number of features, thereby reducing the amount of storage and subsequent calculations, and by extracting local features of the image in the pre-processed image data, noise and interference in the image can be reduced. Through feature extraction, the original information in the image can be reorganized and expressed to make it more in line with human visual perception and cognitive habits, thereby improving the accuracy and comprehensibility of information expression.

[0083] In the embodiment of the present invention, the step of extracting features from the preprocessed image data using a feature extraction algorithm to obtain tumor image features includes:

[0084] The feature extraction algorithm is expressed using the following formula:

[0085] L(x,y) n =∑S(GP)*(M-1)

[0086] Among them, L(x, y) n represents the nth local feature of the tumor image feature, (x, y) represents the coordinates of the pixel center point of the nth local feature, P is the gray value of the pixel center point, G represents the gray value of the neighborhood point of the pixel center point, S is a binary function, when (GP) is greater than or equal to 0, S returns 1, otherwise it returns 0, and M represents the number of neighborhood points of the pixel center point.

[0087] In the embodiment of the present invention, a feature extraction algorithm is used to extract features from the pre-processed image data to obtain tumor image features, thereby improving the speed of subsequent data processing.

[0088] S4. Using a pre-trained tumor morphology matching model, according to the tumor image features, match the most similar tumor morphology model from a pre-established tumor morphology model library to obtain a first tumor morphology model, and optimizing the first tumor morphology model according to the tumor image data to obtain a second tumor morphology model.

[0089] In an embodiment of the present invention, the tumor morphology matching model is a neural network model, including but not limited to a CNN (Convolutional Neural Networks) model and an RNN (Recurrent Neural Network) model.

[0090] In the embodiment of the present invention, the pre-established tumor morphology model library is a database containing a large number of various tumor morphology models.

[0091] In the embodiment of the present invention, before using the pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features, the method further includes:

[0092] Obtain tumor imaging data of different tumor patients to obtain a tumor imaging dataset containing a large number of tumor images;

[0093] Extracting features of the tumor images in the tumor image data set to obtain tumor features;

[0094] Using three-dimensional modeling software to sequentially establish a three-dimensional model of the tumor according to the tumor images in the tumor image data set to obtain a tumor morphology model data set;

[0095] The tumor morphology models in the tumor morphology model data set are labeled with corresponding tumor features, and the labeled tumor morphology models are stored in a pre-established database to obtain a tumor morphology model library.

[0096] In the embodiment of the present invention, referring to Figure 2 As shown, a schematic diagram of a tumor morphology matching model training process is provided in an embodiment of the present invention. Before obtaining a first tumor morphology model by using a pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features, the method further includes:

[0097] S21, acquiring a tumor image dataset from a pre-established tumor morphology model library, and performing feature extraction on the tumor images in the tumor image dataset to obtain a tumor image feature dataset;

[0098] S22, using a neural network model to match tumor morphology models in the tumor morphology model library according to the tumor image features in the tumor image feature data set, and obtain a matching result;

[0099] S23, calculating the matching accuracy rate according to the real tumor morphology model preset according to the tumor image feature and the matching result;

[0100] S24, if the matching accuracy is less than a preset accuracy threshold, adjusting the weight parameters and bias items of the neural network model, and re-performing the above step of using the neural network model to match the tumor morphology model in the tumor morphology model library according to the tumor image features in the tumor image feature data set in turn, to obtain a matching result;

[0101] S25. If the matching accuracy is greater than or equal to a preset accuracy threshold, it is confirmed that the model training is completed and a tumor morphology matching model is obtained.

[0102] In detail, the preset accuracy threshold needs to be set according to the accuracy requirements of the model.

[0103] In the embodiment of the present invention, referring to Figure 3 As shown, a schematic diagram of a process of matching a tumor morphology model according to tumor image features using a tumor morphology matching model according to an embodiment of the present invention is provided, wherein the pre-trained tumor morphology matching model is used to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features to obtain a first tumor morphology model, including:

[0104] S31, performing data conversion processing on the tumor image features using the tumor morphology matching model to obtain a feature matrix;

[0105] S32, using the tumor morphology matching model to perform a convolution operation on the feature matrix through a convolution layer to obtain a convolution feature;

[0106] S33, using the tumor morphology matching model to perform a maximum pooling operation on the convolutional features through a pooling layer to obtain a pooling feature;

[0107] S34, using the tumor morphology matching model to perform tumor morphology model matching in the tumor morphology model library according to the pooling features through a fully connected layer to obtain a first tumor morphology model.

[0108] In the embodiment of the present invention, the data conversion process refers to representing the tumor image features in the form of word vectors, and then concatenating the word vectors to form a matrix.

[0109] In the embodiment of the present invention, the convolution operation refers to performing a convolution operation on the feature matrix and according to the convolution kernel parameters to extract some features in the feature matrix.

[0110] In the embodiment of the present invention, the convolution kernel parameter is an important concept in convolutional neural networks in deep learning. In convolutional neural networks, the convolution kernel is a parameter used to slide on the input data and perform convolution operations. Each convolution kernel has one or more weights, which are learned during the training process and used to determine the mode of convolution operation.

[0111] In the embodiment of the present invention, the maximum pooling operation can effectively reduce the dimension of the data while retaining the most important feature information. Since the maximum pooling operation only focuses on the maximum value in each pooling window and ignores other values, it can effectively reduce noise and redundant information and improve the generalization ability of the model.

[0112] In the embodiment of the present invention, the tumor morphology matching model has the ability to learn implicit features. During the training process, the tumor morphology matching model continuously updates the parameters of the model through the back propagation algorithm, so that the model can automatically learn the implicit features in the data. In the tumor morphology model matching problem in the tumor morphology model library, these implicit features may be abstract features related to the environment, such as a certain waveform shape, spectrum distribution, etc. Through deep neural networks, higher-level and more expressive features can be gradually extracted.

[0113] In detail, the performance of the tumor morphology matching model depends largely on the quality of available data and the representativeness of the samples. Therefore, collecting sufficient, representative and accurately annotated tumor imaging data and the corresponding tumor morphology model is the key to ensuring the accuracy of the model. In addition, timely updating and maintenance of the model is also an important factor in ensuring the effectiveness of the model under different operating environments and data changes.

[0114] In the embodiment of the present invention, the model structure of the first tumor morphology model is adjusted according to the tumor image data to increase the degree of consistency between the model and the tumor image data.

[0115] In an embodiment of the present invention, the step of optimizing the first tumor morphology model according to the tumor image data to obtain the second tumor morphology model includes:

[0116] adjusting the structure of the first tumor morphology model according to the tumor image data to obtain an optimized model;

[0117] Determining whether the accuracy of the optimization model meets the expected requirements;

[0118] If the accuracy of the optimized model does not meet the expected requirements, re-performing the step of adjusting the structure of the first tumor morphology model according to the tumor image data to obtain the optimized model;

[0119] If the accuracy of the optimized model meets the expected requirements, the model optimization is confirmed to be completed and the second tumor morphology model is obtained.

[0120] In the embodiment of the present invention, after the second tumor morphology model is obtained, the second tumor morphology model is stored in the tumor morphology model library.

[0121] In an embodiment of the present invention, a pre-trained tumor morphology matching model is used to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features to obtain a first tumor morphology model, and the first tumor morphology model is optimized according to the tumor image data to obtain a second tumor morphology model, thereby providing necessary data for subsequent operations.

[0122] S5, designing a resection plan according to the second tumor morphology model and the tumor location information;

[0123] In the embodiment of the present invention, the doctor needs to use his own medical knowledge to design a resection plan according to the second tumor morphology model and the tumor location information.

[0124] In detail, the resection plan includes the resection path, the selection of surgical instruments and the selection of surgical equipment.

[0125] S6. Perform tumor resection simulation according to the second tumor morphology model and the resection plan using computer simulation technology to obtain simulation results, verify the feasibility of the resection plan based on the simulation results, and optimize the resection plan.

[0126] In the embodiment of the present invention, the step of using computer simulation technology to perform tumor resection simulation according to the second tumor morphology model and the resection plan to obtain a simulation result includes:

[0127] importing the second tumor morphology model into the computer simulation software;

[0128] Setting simulation parameters according to the second tumor morphology model and the resection plan;

[0129] The second tumor morphology model is subjected to simulated resection according to the resection plan and simulation parameters to obtain a simulation result.

[0130] In the embodiment of the present invention, the computer simulation software includes Simulink, COMSOL Multiphysics and SolidWorks Simu l at ion. These computer simulation software are applied in many fields, and in the medical field, the process of surgery can be simulated.

[0131] like Figure 4 , is a functional module diagram of a preoperative adaptive matching of tumor morphology auxiliary planning system provided by an embodiment of the present invention.

[0132] The auxiliary planning system 100 for adaptively matching tumor morphology before surgery of the present invention can be installed in an electronic device. According to the functions to be implemented, the auxiliary planning system 100 for adaptively matching tumor morphology before surgery can include a data acquisition module 101, a data processing module 102, a model matching module 103, and a design optimization module 104. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0133] In this embodiment, the functions of each module / unit are as follows:

[0134] The data acquisition module 101 is used to obtain the image data and tumor location information of the patient's tumor through medical imaging equipment;

[0135] The data processing module 102 is used to perform denoising on the image data to obtain pre-processed image data, and extract features from the pre-processed image data using a feature extraction algorithm to obtain tumor image features, wherein the extracting features from the pre-processed image data using a feature extraction algorithm to obtain tumor image features includes:

[0136] The feature extraction algorithm is expressed using the following formula:

[0137]

[0138] Among them, L(x, y) n represents the nth local feature of the tumor image feature, (x, y) represents the coordinates of the pixel center point of the nth local feature, P is the gray value of the pixel center point, G represents the gray value of the neighborhood point of the pixel center point, S is a binary function, when (GP) is greater than or equal to 0, S returns 1, otherwise it returns 0, and M represents the number of neighborhood points of the pixel center point;

[0139] The model matching module 103 is used to use a pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features to obtain a first tumor morphology model, and optimize the first tumor morphology model according to the tumor image data to obtain a second tumor morphology model;

[0140] The design optimization module 104 is used to design a resection plan based on the second tumor morphology model and the tumor location information, perform tumor resection simulation based on the second tumor morphology model and the resection plan using computer simulation technology to obtain simulation results, verify the feasibility of the resection plan based on the simulation results, and optimize the resection plan.

[0141] In detail, each module described in the auxiliary planning system 100 for adaptively matching tumor morphology before surgery in the embodiment of the present invention is used in the same manner as described above. Figures 1 to 3 The same technical means are used as the auxiliary planning method for preoperative adaptive matching of tumor morphology described in, and can produce the same technical effects, so they will not be repeated here.

[0142] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0143] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0144] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0145] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0146] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0147] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0148] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems stated in a system claim can also be implemented by one unit or system through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0149] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A preoperative adaptive matching tumor morphology auxiliary planning method, characterized in that: The method comprises: Obtaining imaging data and tumor location information of the patient's tumor through medical imaging equipment; Performing denoising processing on the image data to obtain preprocessed image data; Extracting features from the preprocessed image data using a feature extraction algorithm to obtain tumor image features, wherein extracting features from the preprocessed image data using a feature extraction algorithm to obtain tumor image features includes: The feature extraction algorithm is expressed using the following formula: L(x,y) n =∑S(G-P)*(M-1) Among them, L(x, y) n represents the nth local feature of the tumor image feature, (x, y) represents the coordinates of the pixel center point of the nth local feature, P is the gray value of the pixel center point, G represents the gray value of the neighborhood point of the pixel center point, S is a binary function, when (GP) is greater than or equal to 0, S returns 1, otherwise it returns 0, and M represents the number of neighborhood points of the pixel center point; Using a pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features to obtain a first tumor morphology model, and optimizing the first tumor morphology model according to the tumor image data to obtain a second tumor morphology model; designing a resection plan according to the second tumor morphology model and the tumor location information; Performing a tumor resection simulation according to the second tumor morphology model and the resection plan using computer simulation technology to obtain a simulation result, and verifying the feasibility of the resection plan and optimizing the resection plan according to the simulation result; Before obtaining the first tumor morphology model by using the pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features, the method further includes: Acquiring a tumor image dataset from a pre-established tumor morphology model library, and performing feature extraction on tumor images in the tumor image dataset to obtain a tumor image feature dataset; Using a neural network model, matching tumor morphology models in the tumor morphology model library is performed in turn according to the tumor image features in the tumor image feature data set to obtain a matching result; Calculating the matching accuracy rate according to the real tumor morphology model preset by the tumor image feature and the matching result; If the matching accuracy is less than a preset accuracy threshold, the weight parameters and bias items of the neural network model are adjusted, and the step of using the neural network model to match the tumor morphology model in the tumor morphology model library according to the tumor image features in the tumor image feature data set in turn to obtain a matching result is re-executed; If the matching accuracy is greater than or equal to a preset accuracy threshold, the model training is confirmed to be completed and a tumor morphology matching model is obtained; The method of using the pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features to obtain a first tumor morphology model includes: Using the tumor morphology matching model to perform data conversion processing on the tumor image features to obtain a feature matrix; Using the tumor morphology matching model to perform a convolution operation on the feature matrix through a convolution layer to obtain convolution features; Using the tumor morphology matching model, performing a maximum pooling operation on the convolutional features through a pooling layer to obtain a pooling feature; The tumor morphology matching model is used to match tumor morphology models in the tumor morphology model library through a fully connected layer according to the pooling features to obtain a first tumor morphology model.

2. The preoperative adaptive matching tumor morphology auxiliary planning method according to claim 1, characterized in that: The image data includes the size, shape, contour, density and adhesion relationship of the patient's tumor with surrounding tissues.

3. The preoperative adaptive matching tumor morphology auxiliary planning method according to claim 1, characterized in that: The denoising process is performed on the image data to obtain pre-processed image data, including: Performing noise analysis on the image data to obtain a noise type; Selecting a denoising algorithm according to the noise type; Performing denoising processing on the image data using the denoising algorithm to obtain a preliminary denoised image; If the preliminary denoising image does not meet the expected denoising effect, the denoising algorithm is replaced, and the step of denoising the image data using the denoising algorithm is re-executed to obtain the preliminary denoising image; If the preliminary denoising image meets the expected denoising effect, the image data is saved to obtain pre-processed image data.

4. The preoperative adaptive matching tumor morphology auxiliary planning method according to claim 1, characterized in that: Before using the pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features, the method further includes: Obtain tumor imaging data of different tumor patients to obtain a tumor imaging dataset containing a large number of tumor images; Extracting features of the tumor images in the tumor image data set to obtain tumor features; Using three-dimensional modeling software to sequentially establish a three-dimensional model of the tumor according to the tumor images in the tumor image data set to obtain a tumor morphology model data set; The tumor morphology models in the tumor morphology model data set are labeled with corresponding tumor features, and the labeled tumor morphology models are stored in a pre-established database to obtain a tumor morphology model library.

5. The preoperative adaptive matching tumor morphology auxiliary planning method according to claim 1, characterized in that: The step of optimizing the first tumor morphology model according to the tumor image data to obtain a second tumor morphology model includes: adjusting the structure of the first tumor morphology model according to the tumor image data to obtain an optimized model; Determining whether the accuracy of the optimization model meets the expected requirements; If the accuracy of the optimized model does not meet the expected requirements, re-performing the step of adjusting the structure of the first tumor morphology model according to the tumor image data to obtain the optimized model; If the accuracy of the optimized model meets the expected requirements, the model optimization is confirmed to be completed and the second tumor morphology model is obtained.

6. The preoperative adaptive matching tumor morphology auxiliary planning method according to claim 1, characterized in that: After optimizing the first tumor morphology model according to the tumor image data to obtain the second tumor morphology model, the second tumor morphology model is stored in the tumor morphology model library.

7. The preoperative adaptive matching tumor morphology auxiliary planning method according to claim 1, characterized in that: The step of using computer simulation technology to perform tumor resection simulation according to the second tumor morphology model and the resection plan to obtain a simulation result includes: importing the second tumor morphology model into the computer simulation software; Setting simulation parameters according to the second tumor morphology model and the resection plan; The second tumor morphology model is subjected to simulated resection according to the resection plan and simulation parameters to obtain a simulation result.

8. A preoperative adaptive matching tumor morphology auxiliary planning system, the system implements the method according to any one of claims 1 to 7, characterized in that: The system comprises: A data acquisition module, used to obtain the imaging data and tumor location information of the patient's tumor through medical imaging equipment; A data processing module is used to perform denoising on the image data to obtain pre-processed image data, and to extract features from the pre-processed image data using a feature extraction algorithm to obtain tumor image features, wherein the step of extracting features from the pre-processed image data using a feature extraction algorithm to obtain tumor image features includes: The feature extraction algorithm is expressed using the following formula: L(x,y) n =∑S(G-P)*(M-1) Among them, L(x, y) n represents the nth local feature of the tumor image feature, (x, y) represents the coordinates of the pixel center point of the nth local feature, P is the gray value of the pixel center point, G represents the gray value of the neighborhood point of the pixel center point, S is a binary function, when (GP) is greater than or equal to 0, S returns 1, otherwise it returns 0, and M represents the number of neighborhood points of the pixel center point; a model matching module, configured to use a pre-trained tumor morphology matching model to match the most similar tumor morphology model from a pre-established tumor morphology model library according to the tumor image features to obtain a first tumor morphology model, and optimize the first tumor morphology model according to the tumor image data to obtain a second tumor morphology model; A design optimization module is used to design a resection plan based on the second tumor morphology model and the tumor location information, use computer simulation technology to perform tumor resection simulation based on the second tumor morphology model and the resection plan, obtain simulation results, verify the feasibility of the resection plan based on the simulation results, and optimize the resection plan.

Citation Information

Patent Citations

  • Tumor thermal ablation operation intelligent simulation model training system and training method

    CN112562859A

  • System for determining tumor target contour set after drug therapy and storage medium

    CN114283921A

  • Tumor identification method and system based on artificial intelligence

    CN116862840A

  • Bone tumor focus positioning method and device based on precise recognition

    CN117218200A

  • Shape-detecting method and system thereof for a image of a tumor

    TW200731954A