Tumor identification method based on multi-modal information fusion

Abstract

The invention relates to the technical field of medical image processing and tumor diagnosis, in particular to a tumor recognition method based on multi-modal information fusion. The method comprises the following steps: acquiring tumor multi-modal information of a target object, wherein the multi-modal information comprises tumor image data, tumor genomics data and individual tumor related information; the tumor-related information comprises a vascular structure, a blood flow velocity, fluid viscosity and fluid pressure of the target object; determining a 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 the 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 vascular system according to the individual tumor related information; and according to the diffusion path of the tumor in the local tissue and the metastasis path of the tumor cells through the vascular system, identifying the potential metastasis site of the tumor.

Description

Technical Field

[0001] The present application relates to the technical field of medical image processing and tumor diagnosis, and in particular to a tumor recognition method based on multimodal information fusion. Background Art

[0002] Tumors are a serious threat to human health, and early identification of tumors is crucial. On the one hand, early detection of tumors allows patients to receive timely treatment, greatly improving cure and survival rates. For example, patients with early lung cancer have a high five-year survival rate through surgery and other treatment methods; on the other hand, accurate identification of tumor type, location, development stage and other information can help doctors develop personalized treatment plans, improve treatment effects and reduce unnecessary treatment damage.

[0003] The existing technologies for tumor identification use the following methods: Imaging examination: including X-ray, computed tomography (CT), magnetic resonance imaging (MRI), ultrasound, etc. For example, CT can clearly show the shape, size and location of the tumor; ultrasound is often used for preliminary screening of tumors in the thyroid, breast and other parts.

[0004] Pathological examination: Obtaining tumor tissue samples through biopsy and performing histological and cytological analysis is the "gold standard" for tumor diagnosis and can clarify the benign or malignant nature of the tumor and its specific type.

[0005] Tumor marker testing: Detection of tumor-related substances in the blood and body fluids, such as carcinoembryonic antigen (CEA) and alpha-fetoprotein (AFP), to assist in the early detection of tumors and disease monitoring. However, the results are for reference only and cannot be used alone to diagnose tumors.

[0006] However, the information from the above different modalities has its own limitations. For example, there may be noise, artifacts or incomplete information, which means that the accuracy of tumor identification still needs to be improved. Summary of the invention

[0007] Based on this, it is necessary to provide a tumor recognition method based on multimodal information fusion that can improve the accuracy of tumor recognition in response to the above technical problems.

[0008] In a first aspect, the present application provides a tumor recognition method based on multimodal information fusion, the method comprising: Obtaining 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; Determine tumor cell density, diffusion coefficient, and cell proliferation rate based on tumor imaging data and tumor genomics data; Predict the tumor's diffusion path in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate; Predict the metastatic pathways of tumor cells through the vascular system based on information about individual tumors; Identify potential metastatic sites of the tumor based on the tumor's spread pathways in local tissues and the tumor cell metastasis pathways through the vascular system.

[0009] In one embodiment, predicting the diffusion path of a tumor in a local tissue based on tumor cell density, diffusion coefficient, and cell proliferation rate includes: According to the tumor cell density, diffusion coefficient and cell proliferation rate, the reaction-diffusion equation was constructed; The finite element method is used to solve the reaction-diffusion equation and obtain the diffusion path of the tumor in the local tissue.

[0010] In one embodiment, the finite element method is used to solve the reaction-diffusion equation to obtain the diffusion path of the tumor in the local tissue, including: Divide the tissue region where the tumor is located into a finite element mesh; Discretize the reaction-diffusion equation to obtain a discretized linear equation system; The tumor cell density is used as the initial condition, and the diffusion coefficient and cell proliferation rate are used as dynamic parameters to input into the discretized linear equation system; An iterative algorithm is used to solve the discretized linear equations to obtain the spatial and temporal distribution of tumor cell density; According to the distribution of tumor cell density, the diffusion path of the tumor in the local tissue is extracted.

[0011] In one embodiment, extracting the diffusion path of the tumor in the local tissue according to the distribution of tumor cell density includes: According to the distribution of tumor cell density, the spatial gradient of tumor cell density is determined; According to the gradient direction of the spatial gradient, the diffusion direction of the tumor cells is determined; According to the diffusion direction, the expansion trajectory of the high-density area is identified to obtain the diffusion path; the high-density area is the area where the tumor cell density is greater than the density threshold.

[0012] In one embodiment, predicting the metastasis path of tumor cells through the vascular system based on tumor-related information, predicting the metastasis path of tumor cells through the vascular system based on individual tumor-related information, includes: According to the vascular structure, a geometric model of the vascular network is constructed; Initialize the parameters of the fluid dynamics equations according to the blood flow velocity, fluid viscosity and fluid pressure; The finite element method is used to solve the fluid dynamics equations and obtain the spatiotemporal distribution of the blood flow velocity field and pressure field; According to the distribution of blood flow velocity field and pressure field, the migration probability of tumor cells in the vascular system is determined; Based on the migration probability, the metastatic path of tumor cells through the vascular system is predicted.

[0013] In one embodiment, the fluid dynamics equations are solved to obtain the spatiotemporal distribution of the blood flow velocity field and the pressure field, including: Generate a finite element mesh of the fluid domain; The finite element method is used to discretize the fluid dynamics equations to obtain a discretized linear equation system; The blood flow velocity, fluid viscosity and fluid pressure are used as initial conditions and boundary conditions of the discretized linear equations; An iterative algorithm is used to solve the discretized linear equations to obtain the spatiotemporal distribution of blood flow velocity field and pressure field.

[0014] In one embodiment, determining the migration probability of tumor cells in a vascular system according to the distribution of blood flow velocity field and pressure field includes: Determine the migration speed of tumor cells in the vascular system based on the spatiotemporal distribution of the blood flow velocity field; Determine the migration direction of tumor cells in the vascular system based on the temporal and spatial distribution of the pressure field; A probability model is used to determine the metastasis probability of tumor cells migrating from their current location to other locations.

[0015] In one embodiment, predicting the metastatic path of tumor cells through the vascular system based on the migration probability includes: According to the migration probability of tumor cells in the vascular system, a high probability path of tumor cells in the vascular system is constructed; a high probability path is a path whose migration probability is greater than a probability threshold.

[0016] In one embodiment, identifying potential metastatic sites of a tumor based on the tumor's spreading path in local tissues and the tumor cell's metastatic path through the vascular system includes: According to the diffusion path of the tumor in the local tissue, determine the diffusion direction and range of tumor cells in the tissue; Determine the migration direction and migration target of tumor cells in the vascular system according to the metastatic path of tumor cells through the vascular system; Construct the metastatic pathway of tumor cells from local tissues to distant organs based on the direction of diffusion and migration; Based on the metastatic pathway, the potential metastatic sites of the tumor are predicted.

[0017] In one embodiment, determining the cell proliferation rate based on tumor imaging data and tumor genomics data includes: Screening key genes related to tumor proliferation from tumor genomics data; Construct a regression model of gene expression and proliferation rate: Based on the regression model, the cell proliferation rate was determined.

[0018] In a second aspect, the present application also provides a tumor identification device based on multi-modal information fusion, comprising: An acquisition module is used to acquire multimodal tumor information of a target object, wherein the multimodal information includes tumor imaging data, tumor genomics data, and individual tumor-related information; the tumor-related information includes the target object's vascular structure, blood flow velocity, fluid viscosity, and fluid pressure; A 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; The second analysis module is used to predict the diffusion path of the tumor in the local tissue based on the tumor cell density, diffusion coefficient and cell proliferation rate; A prediction module, which is used to predict the metastatic pathway of tumor cells through the vascular system based on information related to individual tumors; A comprehensive analysis module is used to identify potential metastatic sites of tumors based on the diffusion pathways of tumors in local tissues and the metastatic pathways of tumor cells through the vascular system.

[0019] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented: Obtaining 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; Determine tumor cell density, diffusion coefficient, and cell proliferation rate based on tumor imaging data and tumor genomics data; Predict the tumor's diffusion path in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate; Predict the metastatic pathways of tumor cells through the vascular system based on information related to individual tumors; Identify potential metastatic sites of the tumor based on the tumor's spread pathways in local tissues and the tumor cell metastasis pathways through the vascular system.

[0020] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented: Obtaining 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; Determine tumor cell density, diffusion coefficient, and cell proliferation rate based on tumor imaging data and tumor genomics data; Predict the tumor's diffusion path in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate; Predict the metastatic pathways of tumor cells through the vascular system based on information about individual tumors; Identify potential metastatic sites of the tumor based on the tumor's spread pathways in local tissues and the tumor cell metastasis pathways through the vascular system.

[0021] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor: Obtaining 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; Determine tumor cell density, diffusion coefficient, and cell proliferation rate based on tumor imaging data and tumor genomics data; Predict the tumor's diffusion path in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate; Predict the metastatic pathways of tumor cells through the vascular system based on information about individual tumors; Identify potential metastatic sites of the tumor based on the tumor's spread pathways in local tissues and the tumor cell metastasis pathways through the vascular system.

[0022] The above-mentioned tumor recognition method based on multimodal information fusion can comprehensively capture the biological characteristics and dynamic behavior of tumors by integrating tumor imaging data, genomic data and individual tumor-related information, and significantly improve the detection sensitivity of early tumors. Combining high-resolution imaging data and genomic data can detect tiny lesions that are difficult to detect with traditional methods, and achieve early diagnosis.

[0023] By integrating imaging data, genomic data, and individual tumor-related information, the limitations of traditional methods that rely on single-modality data are overcome, providing more comprehensive diagnostic information. By determining tumor cell density, diffusion coefficient, and cell proliferation rate, the growth and diffusion behavior of the tumor are quantified to improve the accuracy of diagnosis. Based on tumor cell density, diffusion coefficient, and cell proliferation rate, a tumor diffusion model is constructed to accurately predict the diffusion path of the tumor in local tissues.

[0024] Based on the information related to individual tumors (such as vascular structure, blood flow velocity, fluid viscosity and fluid pressure), a hemodynamic model is constructed to predict the metastatic path of tumor cells through the vascular system. By integrating the local diffusion path and the vascular metastasis path, the potential metastatic site of the tumor is identified, providing a scientific basis for clinical treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 is a flow chart of a tumor recognition method based on multi-modal information fusion in one embodiment; Figure 2 A schematic diagram of a flow chart of the steps of obtaining the diffusion path of a tumor in a local tissue in one embodiment; Figure 3 A schematic diagram of a flow chart of a step of extracting a tumor diffusion path in a local tissue in one embodiment; Figure 4 A schematic diagram of a process for predicting the metastatic pathway of tumor cells through the vascular system in one embodiment; Figure 5 It is a schematic flow chart of the steps of obtaining the spatiotemporal distribution of the blood flow velocity field and the pressure field in one embodiment; Figure 6 A schematic diagram of a flow chart of a step of determining the migration probability of tumor cells in a vascular system in one embodiment; Figure 7 A schematic diagram of a process for identifying potential metastatic sites of a tumor in one embodiment; Figure 8 FIG. 1 is a flow chart of the step of determining the cell proliferation rate in one embodiment. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0028] In an exemplary embodiment, Figure 1 As shown, a tumor recognition method based on multimodal information fusion is provided, comprising: S101, obtaining multimodal tumor information of a target object.

[0029] Among them, multimodal information includes 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.

[0030] Optional, tumor imaging data: information on tumor morphology, location, and surrounding tissues obtained through medical imaging technology (such as CT, MRI, or PET).

[0031] Tumor genomics data: tumor gene mutations, expression profiles and molecular characteristics obtained through gene sequencing technology.

[0032] Individual tumor-related information: including the target object’s vascular structure, blood flow velocity, fluid viscosity, fluid pressure and other information.

[0033] S102, determining tumor cell density, diffusion coefficient, and cell proliferation rate based on tumor imaging data and tumor genomics data.

[0034] Optionally, based on tumor imaging data and tumor genomics data, the biological characteristics of the tumor are determined by the following steps: analyzing the morphology and spatial distribution of the tumor through imaging data and calculating the density of tumor cells; determining the diffusion coefficient of tumor cells based on the gene expression spectrum in the genomics data; and calculating the proliferation rate of tumor cells in combination with the expression levels of genes related to tumor cell proliferation.

[0035] S103, predicting the diffusion path of the tumor in the local tissue based on the tumor cell density, diffusion coefficient and cell proliferation rate.

[0036] Optionally, a tumor diffusion model is constructed based on tumor cell density, diffusion coefficient and cell proliferation rate to predict the diffusion path of the tumor in the local tissue. Specifically, it includes: using partial differential equations (such as reaction-diffusion equations) to simulate the diffusion behavior of tumor cells in the local tissue; optimizing the parameters of the diffusion model in combination with the tissue boundary conditions in the tumor imaging data; and predicting the diffusion path of the tumor in the local tissue through numerical simulation methods.

[0037] S104, predicting the metastatic pathway of tumor cells through the vascular system based on the individual tumor related information.

[0038] Optionally, a vascular system model is constructed based on individual tumor-related information to predict the metastatic path of tumor cells through the vascular system. Specifically, it includes: building 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 their possible metastatic path; combining the expression level of invasion-related genes in genomic data to optimize the metastatic path prediction results.

[0039] S105, identifying the potential metastatic sites of the tumor based on the tumor's spreading pathway in local tissues and the tumor cell's metastatic pathway through the vascular system.

[0040] Optionally, based on the tumor diffusion path in the local tissue and the metastatic path of tumor cells through the vascular system, the tumor diffusion and metastatic behavior are comprehensively evaluated to identify the potential metastatic sites of the tumor. Specifically, it includes: fusing the local tissue diffusion path with the vascular system metastatic path; identifying high-risk potential metastatic sites based on the fusion results; generating a tumor metastasis risk assessment report to provide decision support for clinical treatment.

[0041] In an exemplary embodiment, the diffusion path of the tumor in the local tissue is predicted based on the tumor cell density, diffusion coefficient and cell proliferation rate, including: constructing a reaction-diffusion equation based on the tumor cell density, diffusion coefficient and cell proliferation rate; using the finite element method to solve the reaction-diffusion equation to obtain the diffusion path of the tumor in the local tissue.

[0042] Optionally, a reaction-diffusion equation is constructed based on the tumor cell density, diffusion coefficient, and cell proliferation rate. The reaction-diffusion equation is in the form of:

[0043] Where, u is the tumor cell density; D is the diffusion coefficient, which indicates the diffusion capacity of tumor cells in tissues; ρ is the cell proliferation rate, which indicates the proliferation ability of tumor cells; is the Laplace operator, which represents the diffusion behavior of tumor cells in space.

[0044] In an exemplary embodiment, Figure 2 As shown, the finite element method is used to solve the reaction-diffusion equation to obtain the diffusion path of the tumor in the local tissue, including: S201, dividing the tissue region where the tumor is located into finite element grids.

[0045] Optionally, the tissue region where the tumor is located is divided into finite element grids, and each grid unit can be in the form of a triangle, a quadrilateral or a tetrahedron, etc. The density of the grid division is adjusted according to the morphology of the tumor and the complexity of the local tissue to ensure the calculation accuracy.

[0046] S202, discretizing the reaction-diffusion equation to obtain a discretized linear equation system.

[0047] The reaction-diffusion equation is discretized, specifically including: using the Galerkin method or the least squares method to transform the reaction-diffusion equation into a weak form; interpolating the tumor cell density through basis functions to establish a discretized linear equation system.

[0048] S203, taking the tumor cell density as an initial condition, and the diffusion coefficient and the cell proliferation rate as dynamic parameters, and inputting them into the discretized linear equation system.

[0049] Optionally, an initial distribution of tumor cell density is used as an initial condition; the diffusion coefficient and the cell proliferation rate are used as dynamic parameters and input into the discretized linear equation system.

[0050] S204, using an iterative algorithm to solve the discretized linear equations to obtain the spatial and temporal distribution of tumor cell density.

[0051] An iterative algorithm (such as the conjugate gradient method or the GMRES method) is used to solve the discretized linear equations, specifically including: 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.

[0052] S205, extracting the diffusion path of the tumor in the local tissue according to the distribution of tumor cell density.

[0053] Optionally, extract high-value areas of tumor cell density as the frontier of tumor diffusion; combine tissue boundary conditions in tumor imaging data to optimize the prediction results of the diffusion path; generate a visualization image of the tumor diffusion path to provide a reference for clinical diagnosis.

[0054] In an exemplary embodiment, Figure 3 As shown, according to the distribution of tumor cell density, the diffusion path of the tumor in the local tissue is extracted, including: S301, determining a spatial gradient of tumor cell density according to the distribution of tumor cell density.

[0055] Optionally, the spatial gradient of tumor cell density is calculated according to its distribution in space, including: The central difference method or finite element method is used to calculate the gradient value of tumor cell density in space; The calculation formula of the gradient value is:

[0056] in, : indicates a function u The gradient of , which is a vector that describes the direction and rate at which the function changes fastest in space; u ： Represents tumor cell density and is a function of spatial coordinates , that is, at different locations in space There are different cell density values ​​at : It is the coordinate of three-dimensional space, which is used to determine the specific position in space; :yes u right x The partial derivative in the direction indicates that the tumor cell density is x rate of change in direction; :yes u right y The partial derivative in the direction indicates that the tumor cell density is y rate of change in direction; :yes u Yes z The partial derivative in the direction indicates that the tumor cell density is z The rate of change in direction.

[0057] S302, determining the diffusion direction of the tumor cells according to the gradient direction of the spatial gradient.

[0058] Optionally, the diffusion direction of the tumor cells is determined according to the gradient direction of the spatial gradient, including: the gradient direction points to the direction in which the tumor cell density increases fastest; the diffusion direction is consistent with the gradient direction, indicating the diffusion trend of the tumor cells.

[0059] S303, identifying the expansion trajectory of the high-density area according to the diffusion direction, and obtaining the diffusion path.

[0060] The high-density area is an area where the tumor cell density is greater than the density threshold.

[0061] Optionally, according to the diffusion direction, the expansion trajectory of the high-density area is identified to obtain the diffusion path. Specifically, the high-density area is defined as an area where the tumor cell density is greater than a preset density threshold; according to the diffusion direction, the expansion trajectory of the high-density area is tracked; and the expansion trajectory is used as the diffusion path of the tumor.

[0062] In an exemplary embodiment, Figure 4As shown, based on tumor-related information, the metastatic path of tumor cells through the vascular system is predicted. Based on individual tumor-related information, the metastatic path of tumor cells through the vascular system is predicted, including: S401, constructing a geometric model of the vascular network according to the vascular structure.

[0063] Optionally, a geometric model of the vascular network is constructed based on the vascular structure data in the individual tumor-related information, including: extracting the center line and radius of the blood vessels through medical imaging data (such as CT angiography or MRI angiography); using three-dimensional modeling technology (such as NURBS surface or voxelization method) to construct a geometric model of the vascular network.

[0064] S402, initializing parameters of the fluid dynamics equation according to the blood flow velocity, fluid viscosity and fluid pressure.

[0065] Optionally, the parameters of the fluid dynamics equation are initialized according to the blood flow velocity, fluid viscosity and fluid pressure data in the individual tumor related information, including: setting the initial conditions of the blood flow velocity; determining the viscosity coefficient and pressure gradient in the fluid dynamics equation according to the fluid viscosity and fluid pressure.

[0066] S403, using the finite element method to solve the fluid dynamics equations to obtain the spatiotemporal distribution of the blood flow velocity field and the pressure field.

[0067] Optionally, the finite element method is used to solve the fluid dynamics equations (such as the Navier-Stokes equations), specifically including: discretizing the vascular network geometric model into a finite element grid; converting the fluid dynamics equations into a weak form through the Galerkin method or the least squares method; using an iterative algorithm (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.

[0068] S404, determining the migration probability of tumor cells in the vascular system according to the distribution of the blood flow velocity field and the pressure field.

[0069] Optionally, the migration probability of tumor cells in the vascular system is calculated based on the distribution of blood flow velocity field and pressure field. Specifically, it 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 calculation results of migration probability in combination with the biological characteristics of tumor cells (such as adhesion and invasion ability).

[0070] S405, predicting the metastatic pathway of tumor cells through the vascular system based on the migration probability.

[0071] Optionally, the metastasis path of tumor cells through the vascular system is predicted based on the migration probability. Specifically, it includes: selecting the vascular branch with the highest migration probability as the main metastasis path of tumor cells; generating a metastasis path map of tumor cells based on the topological structure of the vascular network; generating a visualization image of the vascular metastasis path of tumor cells to provide a reference for clinical diagnosis.

[0072] In an exemplary embodiment, Figure 5 As shown, the finite element method is used to solve the fluid dynamics equations to obtain the spatiotemporal distribution of the blood flow velocity field and pressure field, including: S501, generating a finite element mesh of the fluid domain.

[0073] Optionally, a finite element mesh of the fluid domain is generated based on the geometric model of the vascular network. Specifically, it includes: The mesh density is adjusted according to the requirements of triangle complexity and calculation accuracy; ensuring that the mesh quality meets the stability requirements of numerical calculations.

[0074] S502, using a finite element method to discretize the fluid dynamics equations to obtain a discretized linear equation group.

[0075] Optionally, the finite element method is used to discretize the fluid dynamics equations (such as the Navier-Stokes equations), specifically including: converting the fluid dynamics equations into a weak form through the Galerkin method or the least squares method; interpolating the velocity and pressure using basis functions to establish a discretized linear equation system.

[0076] S503, taking the blood flow velocity, fluid viscosity and fluid pressure as initial conditions and boundary conditions of the discretized linear equations.

[0077] Optionally, the initial conditions and boundary conditions of the discretized linear equations are set according to the blood flow velocity, fluid viscosity and fluid pressure data in the individual tumor related information. The fluid domain can be discretized using quadrilateral or tetrahedral units; according to the geometric complexity of the blood vessels, the initial distribution of blood flow velocity is used as the initial condition; the fluid viscosity and fluid pressure are used as boundary conditions and input into the discretized linear equations.

[0078] S504, using an iterative algorithm to solve the discretized linear equations to obtain the spatiotemporal distribution of the blood flow velocity field and the pressure field.

[0079] Optionally, an iterative algorithm (such as a conjugate gradient method or a GMRES method) is used to solve the discretized linear equations, specifically including: setting an iterative convergence condition (such as a residual threshold or a maximum number of iterations); and obtaining the spatiotemporal distribution of the blood flow velocity field and the pressure field through iterative calculation.

[0080] In an exemplary embodiment, Figure 6 As shown, according to the distribution of blood flow velocity field and pressure field, the migration probability of tumor cells in the vascular system is determined, including: S601, determining the migration speed of tumor cells in the vascular system according to the spatiotemporal distribution of the blood flow velocity field.

[0081] Optionally, the migration speed of tumor cells in the vascular system is calculated based on the spatiotemporal distribution of the blood flow velocity field, which specifically includes: extracting the local velocity value in the blood flow velocity field; and adjusting the calculation result of the migration speed based on the biological characteristics of the tumor cells (such as size and shape).

[0082] S602, determining the migration direction of tumor cells in the vascular system according to the spatiotemporal distribution of the pressure field.

[0083] Optionally, the migration direction of the tumor cells in the vascular system is determined according to the temporal and spatial distribution of the pressure field, which specifically includes: extracting the pressure gradient in the pressure field; and determining the migration direction of the tumor cells according to the direction of the pressure gradient.

[0084] S603, using a probability model to determine the probability of tumor cells migrating from the current location to other locations.

[0085] Among them, predicting the metastasis path of tumor cells through the vascular system based on the migration probability includes: constructing a high-probability path of tumor cells in the vascular system based on the migration probability of tumor cells in the vascular system.

[0086] Among them, the high probability path is a path whose migration probability is greater than the probability threshold.

[0087] Optionally, the metastasis path of tumor cells through the vascular system is predicted based on the migration probability. Specifically, it includes: selecting the path with a migration probability greater than a preset probability threshold as a high-probability path; generating a high-probability path map of tumor cells in combination with the topological structure of the vascular network; generating a visualization image of the tumor cell vascular metastasis path to provide a reference for clinical diagnosis.

[0088] In an exemplary embodiment, Figure 7 As shown, based on the spread path of the tumor in the local tissue and the metastatic path of tumor cells through the vascular system, the potential metastatic sites of the tumor are identified, including: S701, determining the diffusion direction and diffusion range of tumor cells in the tissue according to the diffusion path of the tumor in the local tissue.

[0089] Optionally, the diffusion direction and diffusion range of tumor cells are extracted according to the diffusion path of the tumor in the local tissue, which specifically includes: determining the diffusion direction through the spatial gradient of the tumor cell density distribution; and determining the diffusion range according to the length and range of the diffusion path.

[0090] S702, determining the migration direction and migration target of the tumor cells in the vascular system according to the migration path of the tumor cells through the vascular system.

[0091] Optionally, the migration direction and migration target of tumor cells are extracted according to the metastatic path of tumor cells through the vascular system, including: determining the migration direction through the distribution of blood flow velocity field and pressure field; determining the migration target according to the end point of the high probability path.

[0092] S703, constructs the metastatic pathway of tumor cells from local tissues to distant organs based on the diffusion direction and migration direction.

[0093] Optionally, based on the diffusion direction and migration direction, a metastatic path of tumor cells from local tissues to distant organs is constructed, including: connecting the diffusion path in the local tissue with the metastatic path in the vascular system; combining the topological structure of the vascular network and the anatomical position of the distant organ to optimize the construction result of the metastatic path.

[0094] S704, predicting the potential metastatic site of the tumor based on the metastatic pathway.

[0095] Optionally, the potential metastatic site of the tumor can be predicted based on the metastatic pathway. Specifically, it includes: extracting the end point of the metastatic pathway as the potential metastatic site; optimizing the prediction results by combining the biological characteristics of distant organs (such as blood supply and tissue structure); generating a visual image of the potential metastatic site of the tumor to provide a reference for clinical diagnosis.

[0096] In an exemplary embodiment, Figure 8 As shown, based on tumor imaging data and tumor genomics data, cell proliferation rate is determined, including: S801, Screening of key genes associated with tumor proliferation from tumor genomics data.

[0097] Optionally, key genes related to tumor proliferation are screened from tumor genomics data, specifically including screening genes related to cell proliferation through gene function annotation database; screening significantly up-regulated or down-regulated genes by differential expression analysis; and determining key genes by combining literature and experimental data.

[0098] S802, construct a regression model of gene expression and proliferation rate.

[0099] Optionally, a regression model of gene expression and proliferation rate is constructed, specifically including: collecting gene expression data and cell proliferation rate data of tumor samples; constructing a regression model using multivariate linear regression or machine learning algorithm; and verifying the predictive performance of the model through cross-validation or independent data sets.

[0100] S803, determining the cell proliferation rate according to the regression model.

[0101] Optionally, the cell proliferation rate is determined based on the regression model, specifically including: inputting gene expression data of the target tumor sample; calculating the cell proliferation rate through the regression model; and optimizing the calculation result in combination with the spatial distribution information in the tumor imaging data.

[0102] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0103] According to the same inventive concept, the embodiment of the present application also provides a tumor identification device based on multimodal information fusion for implementing the above-mentioned tumor identification method based on multimodal information fusion. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodiments of the tumor identification device based on multimodal information fusion provided below can refer to the limitations of the tumor identification method based on multimodal information fusion above, and will not be repeated here.

[0104] In an exemplary embodiment, a tumor identification device based on multimodal information fusion is provided, comprising: An acquisition module is used to acquire multimodal tumor information of a target object, wherein the multimodal information includes tumor imaging data, tumor genomics data, and individual tumor-related information; the tumor-related information includes the target object's vascular structure, blood flow velocity, fluid viscosity, and fluid pressure; A 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; The second analysis module is used to predict the diffusion path of the tumor in the local tissue based on the tumor cell density, diffusion coefficient and cell proliferation rate; A prediction module, which is used to predict the metastatic pathway of tumor cells through the vascular system based on information related to individual tumors; A comprehensive analysis module is used to identify potential metastatic sites of tumors based on the diffusion pathways of tumors in local tissues and the metastatic pathways of tumor cells through the vascular system.

[0105] Each module in the above-mentioned tumor identification device based on multimodal information fusion can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0106] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented: Obtaining 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; Determine tumor cell density, diffusion coefficient, and cell proliferation rate based on tumor imaging data and tumor genomics data; Predict the tumor's diffusion path in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate; Predict the metastatic pathways of tumor cells through the vascular system based on information about individual tumors; Identify potential metastatic sites of the tumor based on the tumor's spread pathways in local tissues and the tumor cell metastasis pathways through the vascular system.

[0107] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: Obtaining 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; Determine tumor cell density, diffusion coefficient, and cell proliferation rate based on tumor imaging data and tumor genomics data; Predict the tumor's diffusion path in local tissues based on tumor cell density, diffusion coefficient, and cell proliferation rate; Predict the metastatic pathways of tumor cells through the vascular system based on information about individual tumors; Identify potential metastatic sites of the tumor based on the tumor's spread pathways in local tissues and the tumor cell metastasis pathways through the vascular system.

[0108] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps: Acquiring multimodal tumor information of the target object, wherein 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; Determining tumor cell density, diffusion coefficient, and cell proliferation rate based on the tumor imaging data and tumor genomics data; Predicting the diffusion path of the tumor in the local tissue according to the tumor cell density, diffusion coefficient and cell proliferation rate; predicting the metastatic pathway of tumor cells through the vascular system based on the individual tumor related information; Based on the spread path of the tumor in the local tissue and the metastasis path of tumor cells through the vascular system, the potential metastatic site of the tumor is identified.

[0109] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium 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. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0110] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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.

[0111] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A tumor recognition method based on multimodal information fusion, characterized in that: The method comprises: Acquiring multimodal tumor information of the target object, wherein 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; Determining tumor cell density, diffusion coefficient, and cell proliferation rate based on the tumor imaging data and tumor genomics data; Predicting the diffusion path of the tumor in the local tissue according to the tumor cell density, diffusion coefficient and cell proliferation rate; predicting the metastatic pathway of tumor cells through the vascular system based on the individual tumor related information; Based on the spread path of the tumor in the local tissue and the metastasis path of tumor cells through the vascular system, the potential metastatic site of the tumor is identified.

2. The method according to claim 1, characterized in that The step of predicting the diffusion path of the tumor in the local tissue according to the tumor cell density, diffusion coefficient and cell proliferation rate comprises: constructing a reaction-diffusion equation according to the tumor cell density, diffusion coefficient and cell proliferation rate; The reaction-diffusion equation is solved by using the finite element method to obtain the diffusion path of the tumor in the local tissue.

3. The method according to claim 2, characterized in that The finite element method is used to solve the reaction-diffusion equation to obtain the diffusion path of the tumor in the local tissue, including: Divide the tissue region where the tumor is located into a finite element mesh; Discretize the reaction-diffusion equation to obtain a discretized linear equation system; Using the tumor cell density as an initial condition, and the diffusion coefficient and cell proliferation rate as dynamic parameters, and inputting them into the discretized linear equation system; Using an iterative algorithm to solve the discretized linear equations to obtain the spatial and temporal distribution of tumor cell density; The diffusion path of the tumor in the local tissue is extracted according to the distribution of the tumor cell density.

4. The method according to claim 3, characterized in that The step of extracting the diffusion path of the tumor in the local tissue according to the distribution of the tumor cell density includes: Determining a spatial gradient of tumor cell density according to the distribution of tumor cell density; According to the gradient direction of the spatial gradient, the diffusion direction of the tumor cells is determined; According to the diffusion direction, identifying the expansion trajectory of the high-density area to obtain the diffusion path; The high-density area is an area where the tumor cell density is greater than a density threshold.

5. The method according to claim 1, characterized in that Predicting the metastasis path of tumor cells through the vascular system according to the tumor-related information Predicting the metastasis path of tumor cells through the vascular system according to the individual tumor-related information includes: constructing a geometric model of the vascular network according to the vascular structure; Initializing parameters of a fluid dynamics equation according to the blood flow velocity, fluid viscosity, and fluid pressure; The fluid dynamics equation is solved by using the finite element method to obtain the spatiotemporal distribution of the blood flow velocity field and the pressure field; determining the migration probability of tumor cells in the vascular system according to the distribution of the blood flow velocity field and the pressure field; Based on the migration probability, the metastatic path of tumor cells through the vascular system is predicted.

6. The method according to claim 5, characterized in that The finite element method is used to solve the fluid dynamics equation to obtain the spatiotemporal distribution of the blood flow velocity field and the pressure field, including: Generate a finite element mesh of the fluid domain; The finite element method is used to discretize the fluid dynamics equations to obtain a discretized linear equation system; The blood flow velocity, fluid viscosity and fluid pressure are used as initial conditions and boundary conditions of the discretized linear equations; An iterative algorithm is used to solve the discretized linear equations to obtain the spatiotemporal distribution of blood flow velocity field and pressure field.

7. The method according to claim 5, characterized in that Determining the migration probability of tumor cells in the vascular system according to the distribution of the blood flow velocity field and the pressure field includes: Determine the migration speed of tumor cells in the vascular system based on the spatiotemporal distribution of the blood flow velocity field; Determine the migration direction of tumor cells in the vascular system based on the temporal and spatial distribution of the pressure field; A probability model is used to determine the metastasis probability of tumor cells migrating from their current location to other locations.

8. The method according to claim 5, characterized in that The step of predicting the metastasis path of tumor cells through the vascular system according to the migration probability comprises: According to the migration probability of tumor cells in the vascular system, a high probability path of tumor cells in the vascular system is constructed; the high probability path is a path whose migration probability is greater than a probability threshold.

9. The method according to claim 1, characterized in that: The method of identifying the potential metastatic site of the tumor according to the diffusion path of the tumor in the local tissue and the metastatic path of the tumor cells through the vascular system includes: Determining the diffusion direction and diffusion range of tumor cells in the tissue according to the diffusion path of the tumor in the local tissue; Determining the migration direction and migration target of the tumor cells in the vascular system according to the migration path of the tumor cells through the vascular system; According to the diffusion direction and migration direction, construct a metastatic path of tumor cells from local tissues to distant organs; Based on the metastasis pathway, the potential metastatic site of the tumor is predicted.

10. The method according to claim 1, characterized in that Determining the cell proliferation rate based on the tumor imaging data and the tumor genomics data includes: Screening key genes related to tumor proliferation from tumor genomics data; Construct a regression model of gene expression and proliferation rate: According to the regression model, the cell proliferation rate is determined.

Images