Tumor recognition method based on multi-modal information fusion

Abstract

The application relates to the technical field of medical image processing and tumor diagnosis, in particular to a tumor identification method based on information fusion of multi-modalities. The method comprises the following steps: acquiring tumor multi-modality information of a target object, wherein the multi-modality information comprises tumor image data, tumor genomics data and individual tumor related information; the tumor related information comprises blood vessel structure, blood flow velocity, fluid viscosity and fluid pressure of the target object; determining tumor cell density, a diffusion coefficient and a cell proliferation rate according to the tumor image data and the tumor genomics data; predicting a diffusion path of the tumor in local tissue according to the tumor cell density, the diffusion coefficient and the cell proliferation rate; predicting a metastasis path of tumor cells through a blood vessel system according to the individual tumor related information; and identifying a potential metastasis site of the tumor according to the diffusion path of the tumor in the local tissue and the metastasis path of the tumor cells through the blood vessel system.

Description

Technical Field

[0001] This application relates to the fields of medical image processing and tumor diagnosis technology, and in particular to a tumor identification method based on multimodal information fusion. Background Technology

[0002] Cancer poses a serious threat to human health, making early cancer identification crucial. On the one hand, early detection allows patients to receive timely treatment, significantly improving cure and survival rates. For example, early-stage lung cancer patients have a high five-year survival rate through surgery and other treatments. On the other hand, accurately identifying the type, location, and stage of cancer helps doctors develop personalized treatment plans, improve treatment outcomes, and reduce unnecessary treatment damage.

[0003] The methods used in existing technologies for tumor identification include the following:

[0004] Imaging examinations include X-rays, computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound. For example, CT can clearly show the shape, size, and location of tumors; ultrasound is often used for preliminary screening of tumors in the thyroid and breast.

[0005] Pathological examination: Obtaining tumor tissue samples through biopsy and performing histological and cytological analysis is the "gold standard" for tumor diagnosis, which can determine the benign or malignant nature of the tumor and its specific type.

[0006] Tumor marker testing: This involves detecting tumor-related substances in blood and body fluids, such as carcinoembryonic antigen (CEA) and alpha-fetoprotein (AFP), to aid in the early detection and monitoring of tumors. However, the results are for reference only and cannot be used to diagnose tumors on their own.

[0007] However, the information from the different modalities mentioned above has its limitations. For example, there may be problems such as noise, artifacts, or incomplete information, which means that the accuracy of tumor identification still needs to be improved. Summary of the Invention

[0008] Therefore, it is necessary to provide a tumor identification method based on multimodal information fusion that can improve the accuracy of tumor identification, addressing the aforementioned technical problems.

[0009] In a first aspect, this application provides a tumor identification method based on multimodal information fusion, the method comprising:

[0010] Acquire multimodal tumor information of the target object, including tumor imaging data, tumor genomics data, and individual tumor-related information; tumor-related information includes the target object's vascular structure, blood flow velocity, fluid viscosity, and fluid pressure;

[0011] Based on tumor imaging data and tumor genomics data, tumor cell density, diffusion coefficient, and cell proliferation rate are determined;

[0012] Predict the spread pathway of tumors in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate;

[0013] Based on individual tumor-related information, predict the metastatic pathway of tumor cells through the vascular system;

[0014] Potential metastatic sites of tumors can be identified based on the spread pathway of tumors in local tissues and the metastatic pathway of tumor cells through the vascular system.

[0015] In one embodiment, predicting the spread pathway of a tumor in a local tissue based on tumor cell density, diffusion coefficient, and cell proliferation rate includes:

[0016] Based on tumor cell density, diffusion coefficient, and cell proliferation rate, a reaction-diffusion equation was constructed.

[0017] The reaction-diffusion equation was solved using the finite element method to obtain the diffusion path of the tumor in local tissues.

[0018] In one embodiment, the finite element method is used to solve the reaction-diffusion equation to obtain the diffusion path of the tumor in local tissue, including:

[0019] The tissue region containing the tumor is divided into finite element meshes;

[0020] Discretize the reaction-diffusion equation to obtain a discretized system of linear equations.

[0021] Tumor cell density is used as the initial condition, and diffusion coefficient and cell proliferation rate are used as dynamic parameters, which are then input into a discretized system of linear equations.

[0022] An iterative algorithm is used to solve the discretized linear equations to obtain the spatial and temporal distribution of tumor cell density.

[0023] Based on the distribution of tumor cell density, the spread path of the tumor in local tissues is extracted.

[0024] In one embodiment, the spread path of the tumor in local tissue is extracted based on the distribution of tumor cell density, including:

[0025] Based on the distribution of tumor cell density, determine the spatial gradient of tumor cell density;

[0026] The direction of tumor cell spread is determined by the gradient direction of the spatial gradient.

[0027] Based on the direction of diffusion, the expansion trajectory of high-density regions is identified to obtain the diffusion path; high-density regions are areas where the tumor cell density is greater than the density threshold.

[0028] In one embodiment, predicting the metastatic pathway of tumor cells through the vascular system based on tumor-related information includes:

[0029] Based on the vascular structure, construct a geometric model of the vascular network;

[0030] Initialize the parameters of the fluid dynamics equations based on blood flow velocity, fluid viscosity, and fluid pressure;

[0031] The finite element method was used to solve the fluid dynamics equations, and the spatiotemporal distributions of the blood flow velocity field and pressure field were obtained.

[0032] The migration probability of tumor cells in the vascular system is determined based on the distribution of blood flow velocity field and pressure field.

[0033] Based on migration probability, predict the metastatic pathway of tumor cells through the vascular system.

[0034] In one embodiment, the fluid dynamics equations are solved to obtain the spatiotemporal distributions of the blood flow velocity field and pressure field, including:

[0035] Generate a finite element mesh for the fluid domain;

[0036] The fluid dynamics equations are discretized using the finite element method, resulting in a discretized linear system of equations.

[0037] Blood flow velocity, fluid viscosity, and fluid pressure are used as the initial and boundary conditions for the discretized linear equation system.

[0038] An iterative algorithm is used to solve the discretized linear equations to obtain the spatiotemporal distributions of the blood flow velocity field and pressure field.

[0039] In one embodiment, the migration probability of tumor cells in the vascular system is determined based on the distribution of the blood flow velocity field and pressure field, including:

[0040] Based on the spatiotemporal distribution of the blood flow velocity field, the migration velocity of tumor cells in the vascular system is determined;

[0041] Based on the spatiotemporal distribution of the pressure field, the migration direction of tumor cells in the vascular system is determined;

[0042] A probabilistic model was used to determine the probability of tumor cells migrating from their current location to other locations.

[0043] In one embodiment, predicting the metastatic pathway of tumor cells through the vascular system based on migration probability includes:

[0044] Based on the migration probability of tumor cells in the vascular system, high-probability paths of tumor cells in the vascular system are constructed; high-probability paths are those with a migration probability greater than a probability threshold.

[0045] In one embodiment, potential metastatic sites of the tumor are identified based on the tumor's spread pathway in local tissues and the metastatic pathway of tumor cells through the vascular system, including:

[0046] Based on the spread path of the tumor in local tissues, determine the direction and extent of tumor cell spread in the tissue;

[0047] Based on the metastatic pathway of tumor cells through the vascular system, the migration direction and target of tumor cells in the vascular system can be determined.

[0048] Based on the direction of diffusion and migration, construct the metastasis pathway of tumor cells from local tissues to distant organs;

[0049] Based on the metastasis pathway, predict potential metastatic sites of the tumor.

[0050] In one embodiment, determining the cell proliferation rate based on tumor imaging data and tumor genomics data includes:

[0051] Screening for key genes associated with tumor proliferation from tumor genomics data;

[0052] Constructing a regression model for gene expression and proliferation rate:

[0053] The cell proliferation rate was determined based on the regression model.

[0054] Secondly, this application also provides a tumor recognition device based on multimodal information fusion, comprising:

[0055] The acquisition module is used to acquire multimodal tumor information of the target object. The multimodal information includes tumor imaging data, tumor genomics data, and individual tumor-related information. The tumor-related information includes the vascular structure, blood flow velocity, fluid viscosity, and fluid pressure of the target object.

[0056] The first analysis module is used to determine tumor cell density, diffusion coefficient, and cell proliferation rate based on tumor imaging data and tumor genomics data.

[0057] The second analysis module is used to predict the spread path of tumors in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate.

[0058] The prediction module is used to predict the metastasis pathway of tumor cells through the vascular system based on individual tumor-related information;

[0059] The comprehensive analysis module is used to identify potential metastatic sites of tumors based on the tumor's spread path in local tissues and the metastatic path of tumor cells through the vascular system.

[0060] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0061] Acquire multimodal tumor information of the target object, including tumor imaging data, tumor genomics data, and individual tumor-related information; tumor-related information includes the target object's vascular structure, blood flow velocity, fluid viscosity, and fluid pressure;

[0062] Based on tumor imaging data and tumor genomics data, tumor cell density, diffusion coefficient, and cell proliferation rate are determined;

[0063] Predict the spread pathway of tumors in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate;

[0064] Based on individual tumor-related information, predict the metastatic pathway of tumor cells through the vascular system;

[0065] Potential metastatic sites of tumors can be identified based on the spread pathway of tumors in local tissues and the metastatic pathway of tumor cells through the vascular system.

[0066] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0067] Acquire multimodal tumor information of the target object, including tumor imaging data, tumor genomics data, and individual tumor-related information; tumor-related information includes the target object's vascular structure, blood flow velocity, fluid viscosity, and fluid pressure;

[0068] Based on tumor imaging data and tumor genomics data, tumor cell density, diffusion coefficient, and cell proliferation rate are determined;

[0069] Predict the spread pathway of tumors in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate;

[0070] Based on individual tumor-related information, predict the metastatic pathway of tumor cells through the vascular system;

[0071] Potential metastatic sites of tumors can be identified based on the spread pathway of tumors in local tissues and the metastatic pathway of tumor cells through the vascular system.

[0072] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0073] Acquire multimodal tumor information of the target object, including tumor imaging data, tumor genomics data, and individual tumor-related information; tumor-related information includes the target object's vascular structure, blood flow velocity, fluid viscosity, and fluid pressure;

[0074] Based on tumor imaging data and tumor genomics data, tumor cell density, diffusion coefficient, and cell proliferation rate are determined;

[0075] Predict the spread pathway of tumors in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate;

[0076] Based on individual tumor-related information, predict the metastatic pathway of tumor cells through the vascular system;

[0077] Potential metastatic sites of tumors can be identified based on the spread pathway of tumors in local tissues and the metastatic pathway of tumor cells through the vascular system.

[0078] The aforementioned tumor identification method based on multimodal information fusion integrates tumor imaging data, genomic data, and individual tumor-related information to comprehensively capture the biological characteristics and dynamic behavior of tumors, significantly improving the detection sensitivity of early-stage tumors. Combining high-resolution imaging data and genomic data can detect minute lesions that are difficult to detect using traditional methods, enabling early diagnosis.

[0079] By fusing imaging data, genomic data, and individual tumor-related information, this approach overcomes the limitations of traditional methods that rely on single-modality data, providing more comprehensive diagnostic information. By determining tumor cell density, diffusion coefficient, and cell proliferation rate, it quantifies tumor growth and spread behavior, improving diagnostic accuracy. Based on tumor cell density, diffusion coefficient, and cell proliferation rate, a tumor spread model is constructed to accurately predict the spread pathway of tumors in local tissues.

[0080] Based on individual tumor-related information (such as vascular structure, blood flow velocity, fluid viscosity, and fluid pressure), a hemodynamic model is constructed to predict the metastatic pathways of tumor cells through the vascular system. By integrating local diffusion pathways and vascular metastasis pathways, potential metastatic sites of tumors are identified, providing a scientific basis for clinical treatment. Attached Figure Description

[0081] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the accompanying drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0082] Figure 1 This is a flowchart illustrating a tumor identification method based on multimodal information fusion in one embodiment;

[0083] Figure 2 This is a flowchart illustrating the steps involved in obtaining the spread path of a tumor in a local tissue, as shown in one embodiment.

[0084] Figure 3 This is a flowchart illustrating the steps for extracting the spread path of a tumor in a local tissue in one embodiment.

[0085] Figure 4 This is a flowchart illustrating the steps involved in predicting the metastatic pathway of tumor cells through the vascular system in one embodiment.

[0086] Figure 5 This is a flowchart illustrating the steps for obtaining the spatiotemporal distribution of the blood flow velocity field and pressure field in one embodiment.

[0087] Figure 6 This is a flowchart illustrating the steps for determining the migration probability of tumor cells in the vascular system in one embodiment.

[0088] Figure 7 This is a flowchart illustrating the steps for identifying potential metastatic sites of a tumor in one embodiment;

[0089] Figure 8 This is a flowchart illustrating the steps for determining the cell proliferation rate in one embodiment. Detailed Implementation

[0090] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0091] In one exemplary embodiment, such as Figure 1 As shown, a tumor identification method based on multimodal information fusion is provided, including:

[0092] S101, Obtain multimodal tumor information of the target object.

[0093] The multimodal information includes tumor imaging data, tumor genomics data, and individual tumor-related information; the tumor-related information includes the target's vascular structure, blood flow velocity, fluid viscosity, and fluid pressure.

[0094] Optional, tumor imaging data: information on tumor morphology, location, and surrounding tissues obtained through medical imaging techniques such as CT, MRI, or PET.

[0095] Tumor genomics data: tumor gene mutations, expression profiles, and molecular characteristics obtained through gene sequencing technology.

[0096] Individual tumor-related information includes information such as the target patient's vascular structure, blood flow velocity, fluid viscosity, and fluid pressure.

[0097] S102, based on tumor imaging data and tumor genomics data, determines tumor cell density, diffusion coefficient, and cell proliferation rate.

[0098] Optionally, based on tumor imaging data and tumor genomics data, the biological characteristics of the tumor are determined through the following steps: analyzing the morphology and spatial distribution of the tumor through imaging data to calculate the tumor cell density; determining the diffusion coefficient of tumor cells based on the gene expression profile in the genomics data; and calculating the proliferation rate of tumor cells by combining the expression levels of genes related to tumor cell proliferation.

[0099] S103 predicts the spread path of tumors in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate.

[0100] Optionally, a tumor diffusion model can be constructed based on tumor cell density, diffusion coefficient, and cell proliferation rate to predict the diffusion path of the tumor in local tissues. Specifically, this includes: simulating the diffusion behavior of tumor cells in local tissues using partial differential equations (such as reaction-diffusion equations); optimizing the parameters of the diffusion model by incorporating tissue boundary conditions from tumor imaging data; and predicting the diffusion path of the tumor in local tissues through numerical simulation methods.

[0101] S104 predicts the metastatic pathway of tumor cells through the vascular system based on individual tumor-related information.

[0102] Optionally, based on individual tumor-related information, a vascular system model can be constructed to predict the metastatic pathways of tumor cells through the vascular system. Specifically, this includes: establishing a hemodynamic model based on vascular structure, blood flow velocity, fluid viscosity, and fluid pressure information; simulating the migration behavior of tumor cells in the vascular system to predict possible metastatic pathways; and optimizing the metastatic pathway prediction results by combining the expression levels of invasion-related genes from genomic data.

[0103] S105 identifies potential metastatic sites of tumors based on the tumor's spread pathway in local tissues and the metastatic pathway of tumor cells through the vascular system.

[0104] Optionally, based on the tumor's spread pathway in local tissues and the metastatic pathway of tumor cells through the vascular system, the spread and metastatic behavior of the tumor can be comprehensively assessed to identify potential metastatic sites. Specifically, this includes: performing a fusion analysis of local tissue spread pathways and vascular system metastatic pathways; identifying high-risk potential metastatic sites based on the fusion results; and generating a tumor metastasis risk assessment report to provide decision support for clinical treatment.

[0105] In an exemplary embodiment, predicting the diffusion path of a tumor in a local tissue based on tumor cell density, diffusion coefficient, and cell proliferation rate includes: constructing a reaction-diffusion equation based on tumor cell density, diffusion coefficient, and cell proliferation rate; and solving the reaction-diffusion equation using the finite element method to obtain the diffusion path of the tumor in the local tissue.

[0106] Optionally, a reaction-diffusion equation can be constructed based on tumor cell density, diffusion coefficient, and cell proliferation rate. The reaction-diffusion equation takes the following form:

[0107] Where u is the tumor cell density;

[0108] D is the diffusion coefficient, which represents the ability of tumor cells to spread in tissues;

[0109] ρ is the cell proliferation rate, representing the proliferative capacity of tumor cells;

[0110] is the Laplace operator, representing the diffusion behavior of tumor cells in space.

[0111] In one exemplary embodiment, such as Figure 2 As shown, the reaction-diffusion equation is solved using the finite element method to obtain the diffusion path of the tumor in local tissues, including:

[0112] S201 divides the tissue region containing the tumor into a finite element mesh.

[0113] Optionally, the tissue region containing the tumor can be divided into finite element meshes, where each mesh element can be a triangle, quadrilateral, or tetrahedron. The mesh density is adjusted based on the tumor's morphology and the complexity of the local tissue to ensure computational accuracy.

[0114] S202, the reaction-diffusion equation is discretized to obtain a discretized linear equation system.

[0115] Discretization of the reaction-diffusion equations includes: using the Galerkin method or least squares method to transform the reaction-diffusion equations into a weak form; and establishing a discretized linear equation system by interpolating the tumor cell density using basis functions.

[0116] S203 uses tumor cell density as the initial condition and diffusion coefficient and cell proliferation rate as dynamic parameters, inputting them into a discretized linear equation system.

[0117] Optionally, the initial distribution of tumor cell density can be used as the initial condition; the diffusion coefficient and cell proliferation rate can be used as dynamic parameters and input into the discretized linear equation system.

[0118] S204 uses an iterative algorithm to solve a set of discretized linear equations to obtain the spatial and temporal distribution of tumor cell density.

[0119] Iterative algorithms (such as the conjugate gradient method or the GMRES method) are used to solve the discretized linear equation system. Specifically, this includes setting iterative convergence conditions (such as residual threshold or maximum number of iterations); and obtaining the spatial and temporal distribution of tumor cell density through iterative calculation.

[0120] S205 extracts the spread path of tumors in local tissues based on the distribution of tumor cell density.

[0121] Optionally, high-density areas of tumor cells can be extracted as the front line of tumor spread; the prediction results of the spread path can be optimized by combining the tissue boundary conditions in the tumor imaging data; and a visualized image of the tumor spread path can be generated to provide a reference for clinical diagnosis.

[0122] In one exemplary embodiment, such as Figure 3 As shown, based on the distribution of tumor cell density, the spread path of the tumor in the local tissue is extracted, including:

[0123] S301, based on the distribution of tumor cell density, determines the spatial gradient of tumor cell density.

[0124] Optionally, the spatial gradient of tumor cell density can be calculated based on its spatial distribution. This specifically includes:

[0125] The central difference method or the finite element method are used to calculate the gradient value of tumor cell density in space;

[0126] The formula for calculating the gradient value is:

[0127] in, : Represents a function u The gradient of a function is a vector that describes the direction and rate of change of the function in space. u：Representing tumor cell density, it is a function of spatial coordinates. That is, in different locations in space Different cell density values ​​exist at different locations; : These are coordinates in three-dimensional space, used to determine a specific location in space;

[0128] :yes u right x The partial derivative in the direction represents the tumor cell density at... x Rate of change in direction;

[0129] :yes u right y The partial derivative in the direction represents the tumor cell density at... y Rate of change in direction;

[0130] :yes u For z The partial derivative in the direction represents the tumor cell density at... z Rate of change in direction.

[0131] S302 determines the direction of tumor cell spread based on the gradient direction of the spatial gradient.

[0132] Optionally, the direction of tumor cell spread can be determined based on the gradient direction of the spatial gradient. Specifically, this includes: the gradient direction pointing towards the direction of the fastest increase in tumor cell density; and the spread direction being consistent with the gradient direction, characterizing the spread trend of tumor cells.

[0133] S303: Based on the diffusion direction, identify the expansion trajectory of the high-density region and obtain the diffusion path.

[0134] High-density regions are those where the tumor cell density is greater than the density threshold.

[0135] Optionally, based on the direction of diffusion, the expansion trajectory of the high-density region can be identified to obtain the diffusion path. Specifically, this includes: defining the high-density region as a region where the tumor cell density is greater than a preset density threshold; tracking the expansion trajectory of the high-density region according to the direction of diffusion; and using the expansion trajectory as the diffusion path of the tumor.

[0136] In one exemplary embodiment, such as Figure 4 As shown, based on tumor-related information, the metastatic pathway of tumor cells through the vascular system is predicted. This includes predicting the metastatic pathway of tumor cells through the vascular system based on individual tumor-related information.

[0137] S401, construct a geometric model of the vascular network based on the vascular structure.

[0138] Optionally, a geometric model of the vascular network can be constructed based on vascular structure data from individual tumor-related information. Specifically, this includes: extracting the centerline and radius of blood vessels from medical imaging data (such as CT angiography or MRI angiography); and constructing a geometric model of the vascular network using 3D modeling techniques (such as NURBS surfaces or voxelization methods).

[0139] S402 initializes the parameters of the fluid dynamics equations based on blood flow velocity, fluid viscosity, and fluid pressure.

[0140] Optionally, the parameters of the fluid dynamics equations can be initialized based on blood flow velocity, fluid viscosity, and fluid pressure data from individual tumor-related information. Specifically, this includes setting initial conditions for blood flow velocity; and determining the viscosity coefficient and pressure gradient in the fluid dynamics equations based on fluid viscosity and fluid pressure.

[0141] S403 uses the finite element method to solve the fluid dynamics equations and obtain the spatiotemporal distribution of the blood flow velocity field and pressure field.

[0142] Optionally, the finite element method can be used to solve the fluid dynamics equations (such as the Navier-Stokes equations). Specifically, this includes: discretizing the vascular network geometric model into a finite element mesh; transforming the fluid dynamics equations into a weak form using the Galerkin method or the least squares method; and using iterative algorithms (such as the conjugate gradient method or the GMRES method) to solve the discretized linear equations to obtain the spatiotemporal distribution of the blood flow velocity field and pressure field.

[0143] S404 determines the migration probability of tumor cells in the vascular system based on the distribution of blood flow velocity field and pressure field.

[0144] Optionally, the migration probability of tumor cells in the vascular system can be calculated based on the distribution of the blood flow velocity field and the pressure field. Specifically, this includes: determining the migration direction of tumor cells based on the blood flow velocity field; calculating the migration probability of tumor cells in different vascular branches based on the pressure field; and optimizing the calculated migration probability by combining the biological characteristics of tumor cells (such as adhesion and invasiveness).

[0145] S405 predicts the metastatic pathway of tumor cells through the vascular system based on migration probability.

[0146] Optionally, based on migration probability, the metastatic pathways of tumor cells through the vascular system can be predicted. Specifically, this includes: selecting the vascular branch with the highest migration probability as the primary metastatic pathway for tumor cells; generating a metastatic pathway map of tumor cells by combining the topology of the vascular network; and generating a visualized image of the vascular metastatic pathway of tumor cells to provide a reference for clinical diagnosis.

[0147] In one exemplary embodiment, such as Figure 5As shown, the finite element method is used to solve the fluid dynamics equations, obtaining the spatiotemporal distributions of the blood flow velocity and pressure fields, including:

[0148] S501 generates the finite element mesh for the fluid domain.

[0149] Optionally, a finite element mesh for the fluid domain is generated based on the geometric model of the vascular network. Specifically, this includes:

[0150] The mesh density was adjusted based on triangular complexity and computational accuracy requirements to ensure that the mesh quality met the stability requirements of numerical computation.

[0151] S502 uses the finite element method to discretize the fluid dynamics equations, resulting in a discretized linear equation set.

[0152] Optionally, the fluid dynamics equations (such as the Navier-Stokes equations) can be discretized using the finite element method. Specifically, this includes: transforming the fluid dynamics equations into a weak form using the Galerkin method or the least squares method; and using basis functions to interpolate velocity and pressure to establish a discretized linear equation set.

[0153] S503 uses blood flow velocity, fluid viscosity, and fluid pressure as the initial and boundary conditions for the discretized linear equations.

[0154] Optionally, initial and boundary conditions for the discretized linear equation system can be set based on blood flow velocity, fluid viscosity, and fluid pressure data from individual tumor-related information. Quadrilateral or tetrahedral elements can be used to discretize the fluid domain; based on the geometric complexity of the blood vessels, the initial distribution of blood flow velocity can be used as the initial condition; fluid viscosity and fluid pressure can be used as boundary conditions, and these can be input into the discretized linear equation system.

[0155] S504 uses an iterative algorithm to solve the discretized linear equations to obtain the spatiotemporal distribution of the blood flow velocity field and pressure field.

[0156] Optionally, an iterative algorithm (such as the conjugate gradient method or the GMRES method) can be used to solve the discretized linear equation system. Specifically, this includes setting iterative convergence conditions (such as residual threshold or maximum number of iterations); and obtaining the spatiotemporal distribution of the blood flow velocity field and pressure field through iterative calculation.

[0157] In one exemplary embodiment, such as Figure 6 As shown, the migration probability of tumor cells in the vascular system is determined based on the distribution of blood flow velocity and pressure fields, including:

[0158] S601 determines the migration speed of tumor cells in the vascular system based on the spatiotemporal distribution of the blood flow velocity field.

[0159] Optionally, the migration velocity of tumor cells in the vascular system can be calculated based on the spatiotemporal distribution of the blood flow velocity field. Specifically, this includes: extracting local velocity values ​​from the blood flow velocity field; and adjusting the calculated migration velocity based on the biological characteristics of the tumor cells (such as size and shape).

[0160] S602 determines the migration direction of tumor cells in the vascular system based on the spatiotemporal distribution of the pressure field.

[0161] Optionally, the migration direction of tumor cells in the vascular system can be determined based on the spatiotemporal distribution of the pressure field. Specifically, this includes: extracting the pressure gradient in the pressure field; and determining the migration direction of tumor cells based on the direction of the pressure gradient.

[0162] S603 uses a probabilistic model to determine the probability of metastasis of tumor cells from their current location to other locations.

[0163] Among them, predicting the metastasis path of tumor cells through the vascular system based on migration probability includes: constructing high-probability paths of tumor cells in the vascular system based on the migration probability of tumor cells in the vascular system.

[0164] Among them, high-probability paths are those with a migration probability greater than a probability threshold.

[0165] Optionally, based on migration probability, the metastatic pathways of tumor cells through the vascular system can be predicted. Specifically, this includes: selecting pathways with migration probabilities greater than a preset probability threshold as high-probability pathways; generating a high-probability pathway map of tumor cells by combining the topology of the vascular network; and generating a visualized image of the tumor cell vascular metastasis pathway to provide a reference for clinical diagnosis.

[0166] In one exemplary embodiment, such as Figure 7 As shown, potential metastatic sites of tumors are identified based on the tumor's spread pathway in local tissues and the metastatic pathway of tumor cells through the vascular system, including:

[0167] S701 determines the direction and extent of tumor cell spread within the tissue based on the tumor's spread path in the local tissue.

[0168] Optionally, the direction and extent of tumor cell diffusion can be extracted based on the tumor's diffusion path within the local tissue. Specifically, this includes: determining the diffusion direction using the spatial gradient of tumor cell density distribution; and determining the diffusion extent based on the length and range of the diffusion path.

[0169] S702 determines the migration direction and target of tumor cells in the vascular system based on the metastasis pathway of tumor cells through the vascular system.

[0170] Optionally, the migration direction and target of tumor cells can be extracted based on the metastatic pathway of tumor cells through the vascular system. Specifically, this includes: determining the migration direction by analyzing the distribution of blood flow velocity and pressure fields; and determining the migration target based on the endpoints of high-probability paths.

[0171] S703 constructs a metastasis pathway for tumor cells from local tissues to distant organs based on the direction of diffusion and migration.

[0172] Optionally, based on the direction of diffusion and migration, a metastatic pathway for tumor cells from local tissues to distant organs can be constructed. Specifically, this includes: connecting the diffusion pathways in local tissues with the metastatic pathways in the vascular system; and optimizing the constructed metastatic pathways by combining the topology of the vascular network and the anatomical location of distant organs.

[0173] S704 predicts potential metastatic sites of tumors based on metastatic pathways.

[0174] Optionally, based on the metastatic pathway, potential metastatic sites of the tumor can be predicted. This includes: extracting the endpoints of the metastatic pathway as potential metastatic sites; optimizing the prediction results by combining the biological characteristics of distant organs (such as blood supply and tissue structure); and generating visualized images of potential metastatic sites to provide a reference for clinical diagnosis.

[0175] In one exemplary embodiment, such as Figure 8 As shown, cell proliferation rate is determined based on tumor imaging data and tumor genomics data, including:

[0176] S801, screening key genes associated with tumor proliferation from tumor genomics data.

[0177] Optionally, key genes associated with tumor proliferation can be screened from tumor genomics data. Specifically, this includes screening for genes associated with cell proliferation through gene function annotation databases; screening for significantly upregulated or downregulated genes using differential expression analysis; and identifying key genes by combining literature and experimental data.

[0178] S802, construct a regression model of gene expression and proliferation rate.

[0179] Optionally, a regression model for gene expression and proliferation rate can be constructed, which specifically includes: collecting gene expression data and cell proliferation rate data of tumor samples; constructing a regression model using multiple linear regression or machine learning algorithms; and validating the predictive performance of the model through cross-validation or independent datasets.

[0180] S803, based on a regression model, determines the cell proliferation rate.

[0181] Optionally, the cell proliferation rate can be determined based on a regression model, specifically including: inputting gene expression data of the target tumor sample; calculating the cell proliferation rate through a regression model; and optimizing the calculation results by combining spatial distribution information in tumor imaging data.

[0182] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0183] Based on the same inventive concept, this application also provides a tumor identification device based on multimodal information fusion for implementing the tumor identification method based on multimodal information fusion described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations of one or more embodiments of the tumor identification device based on multimodal information fusion provided below can be found in the limitations of the tumor identification method based on multimodal information fusion described above, and will not be repeated here.

[0184] In one exemplary embodiment, a tumor identification device based on multimodal information fusion is provided, comprising:

[0185] The acquisition module is used to acquire multimodal tumor information of the target object. The multimodal information includes tumor imaging data, tumor genomics data, and individual tumor-related information. The tumor-related information includes the vascular structure, blood flow velocity, fluid viscosity, and fluid pressure of the target object.

[0186] The first analysis module is used to determine tumor cell density, diffusion coefficient, and cell proliferation rate based on tumor imaging data and tumor genomics data.

[0187] The second analysis module is used to predict the spread path of tumors in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate.

[0188] The prediction module is used to predict the metastasis pathway of tumor cells through the vascular system based on individual tumor-related information;

[0189] The comprehensive analysis module is used to identify potential metastatic sites of tumors based on the tumor's spread path in local tissues and the metastatic path of tumor cells through the vascular system.

[0190] The modules in the tumor recognition device based on multimodal information fusion described above can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0191] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0192] Acquire multimodal tumor information of the target object, including tumor imaging data, tumor genomics data, and individual tumor-related information; tumor-related information includes the target object's vascular structure, blood flow velocity, fluid viscosity, and fluid pressure;

[0193] Based on tumor imaging data and tumor genomics data, tumor cell density, diffusion coefficient, and cell proliferation rate are determined;

[0194] Predict the spread pathway of tumors in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate;

[0195] Based on individual tumor-related information, predict the metastatic pathway of tumor cells through the vascular system;

[0196] Potential metastatic sites of tumors can be identified based on the spread pathway of tumors in local tissues and the metastatic pathway of tumor cells through the vascular system.

[0197] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0198] Acquire multimodal tumor information of the target object, including tumor imaging data, tumor genomics data, and individual tumor-related information; tumor-related information includes the target object's vascular structure, blood flow velocity, fluid viscosity, and fluid pressure;

[0199] Based on tumor imaging data and tumor genomics data, tumor cell density, diffusion coefficient, and cell proliferation rate are determined;

[0200] Predict the spread pathway of tumors in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate;

[0201] Based on individual tumor-related information, predict the metastatic pathway of tumor cells through the vascular system;

[0202] Potential metastatic sites of tumors can be identified based on the spread pathway of tumors in local tissues and the metastatic pathway of tumor cells through the vascular system.

[0203] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0204] The tumor multimodal information of the target object is acquired, including tumor imaging data, tumor genomics data, and individual tumor-related information; the tumor-related information includes the vascular structure, blood flow velocity, fluid viscosity, and fluid pressure of the target object.

[0205] Based on the tumor imaging data and tumor genomics data, the tumor cell density, diffusion coefficient, and cell proliferation rate were determined.

[0206] Based on the tumor cell density, diffusion coefficient, and cell proliferation rate, predict the spread path of the tumor in local tissues;

[0207] Based on the individual tumor-related information, predict the metastatic pathway of tumor cells through the vascular system;

[0208] Potential metastatic sites of the tumor can be identified based on the tumor's spread pathway in local tissues and the metastatic pathway of tumor cells through the vascular system.

[0209] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchain. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logic devices based on quantum computing, artificial intelligence (AI) processors, etc., and are not limited to these.

[0210] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0211] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A tumor recognition method based on multi-modal information fusion, characterized by: obtaining tumor multi-modal information of a target object, the multi-modal information including tumor image data, tumor genomics data, and individual tumor related information; the individual tumor related information including blood vessel structure, blood flow velocity, fluid viscosity, and fluid pressure of the target object; determining tumor cell density, diffusion coefficient, and cell proliferation rate according to the tumor image data and the tumor genomics data; dividing a tumor-located tissue region into a finite element grid; discretizing a reaction-diffusion equation to obtain a discretized linear equation set; inputting the tumor cell density as an initial condition, and inputting the diffusion coefficient and the cell proliferation rate as dynamic parameters into the discretized linear equation set; solving the discretized linear equation set by using an iterative algorithm to obtain the distribution of the tumor cell density in space and time; determining the diffusion direction of the tumor cells according to the distribution of the tumor cell density in space and time, and extracting the diffusion path of the tumor in the local tissue; constructing a geometric model of a blood vessel network based on the blood vessel structure, initializing a fluid dynamics equation in combination with the blood flow velocity, the fluid viscosity, and the fluid pressure, and solving the fluid dynamics equation by using a finite element method to obtain the spatiotemporal distribution of the blood flow velocity field and the pressure field; determining the migration probability of the tumor cells in the blood vessel system according to the spatiotemporal distribution of the blood flow velocity field and the pressure field, and predicting the metastasis path of the tumor cells through the blood vessel system; determining the migration velocity of the tumor cells in the blood vessel system according to the spatiotemporal distribution of the blood flow velocity field; determining the migration direction of the tumor cells in the blood vessel system according to the spatiotemporal distribution of the pressure field; determining the metastasis probability of the tumor cells from the current position to other positions by using a probability model; determining the spatial gradient of the tumor cell density according to the distribution of the tumor cell density in space and time; determining the diffusion direction of the tumor cells according to the gradient direction of the spatial gradient; identifying the expansion trajectory of a high-density region according to the diffusion direction to obtain a diffusion path; the high-density region is a region where the tumor cell density is greater than a preset density threshold; and matching the diffusion direction of the diffusion path with the migration direction of the metastasis path to construct a continuous metastasis path of the tumor cells from the local tissue to a distant organ.

2. The multi-modal based information fusion based tumor identification method according to claim 1, characterized in that, solving the fluid dynamics equation by using a finite element method to obtain the spatiotemporal distribution of the blood flow velocity field and the pressure field, including: generating a finite element grid of a fluid domain; discretizing the fluid dynamics equation by using a finite element method to obtain a discretized linear equation set; inputting the blood flow velocity, the fluid viscosity, and the fluid pressure as initial conditions and boundary conditions of the discretized linear equation set; solving the discretized linear equation set by using an iterative algorithm to obtain the spatiotemporal distribution of the blood flow velocity field and the pressure field.

3. The multi-modal based information fusion based tumor identification method according to claim 1, characterized in that, predicting the metastasis path of the tumor cells through the blood vessel system according to the migration probability, including: constructing a high-probability path of the tumor cells in the blood vessel system according to the migration probability of the tumor cells in the blood vessel system; the high-probability path is a path with a migration probability greater than a probability threshold.

4. The multi-modal based information fusion based tumor identification method according to claim 1, characterized in that, matching the diffusion direction of the diffusion path with the migration direction of the metastasis path to construct a continuous metastasis path of the tumor cells from the local tissue to a distant organ, including: According to the diffusion path of the tumor in the local tissue, the diffusion direction and diffusion range of the tumor cells in the tissue are determined; According to the metastasis path of tumor cells through the vascular system, the migration direction and migration target of tumor cells in the vascular system are determined; According to the diffusion direction and migration direction, the metastasis path of tumor cells from local tissue to distant organs is constructed; According to the metastasis path, the potential metastatic sites of the tumor are predicted.

5. The multi-modal based information fusion based tumor identification method, according to claim 1, characterized in that, According to the tumor image data and the tumor genomics data, the cell proliferation rate is determined, including: Screening key genes related to tumor proliferation from tumor genomics data; Constructing a regression model of gene expression and proliferation rate: According to the regression model, the cell proliferation rate is determined.

Images