Method for predicting brain disease risk and method for learning brain disease risk analysis model

Abstract

A method for predicting the risk of brain diseases according to an embodiment of the present disclosure includes: obtaining a brain disease risk analysis model that has completed learning; obtaining target shape information of the cerebrovascular vessels of a target patient; obtaining target blood flow information of the cerebrovascular vessels of the target patient; obtaining target patient information of the target patient; and inputting the target shape information, the target blood flow information, and the target patient information into the brain disease risk analysis model, and obtaining a brain disease risk value output through the brain disease risk analysis model.

Description

Technical Field

[0001] The present disclosure relates to a method for predicting the risk of brain diseases and a method for training a brain disease risk analysis model. Specifically, the present disclosure relates to a method for predicting the risk of brain diseases based on the shape information, blood flow information, and / or patient information of cerebral blood vessels, and a method for training a brain disease risk analysis model for predicting the risk of brain diseases.

Background Art

[0002] Various imaging techniques such as Positron Emission Tomography (PET), Functional magnetic resonance imaging (fMRI), Magnetic Resonance Angiography (MRA), and Computed Tomography have been developed to visualize the three-dimensional structures of the brain and cerebral blood vessels. The development of such imaging techniques has enabled the measurement of cerebral blood flow throughout the brain, and the visualization of cerebral blood vessels and the measurement of cerebral blood flow have attracted attention for research on predicting or diagnosing the risk of brain diseases based on the shape information and blood flow information of cerebral blood vessels.

[0003] Conventionally, brain diseases related to cerebral blood vessels have been diagnosed or predicted depending on medical databases for patients' gender, diseases, underlying diseases, and ethnicity, and the patients' symptoms. However, the conventional methods have limitations in that they only predict the possibility of brain diseases based on the patients' symptoms and statistical data, and cannot diagnose or predict brain diseases by comprehensively considering the shape information and blood flow information of cerebral blood vessels.

[0004] Therefore, there is a real need for the development of new technologies for predicting or analyzing the risk of brain diseases by comprehensively considering the shape information, blood flow information, and / or health information of patients of cerebral blood vessels.

Summary of the Invention

Problems to be Solved by the Invention

[0005] One problem to be solved by the present disclosure is to provide a method for calculating the risk of brain diseases based on cerebrovascular shape information, cerebral blood flow information, and / or patient information, and a method for learning a brain disease risk analysis model.

[0006] The problems to be solved by the present disclosure are not limited to the problems described above, and problems not mentioned will be clearly understood by those of ordinary skill in the technical field to which the present disclosure pertains from the present specification and the accompanying drawings.

Means for Solving the Problems

[0007] This summary is provided to introduce a selection of concepts in a simplified form that will be further described in the detailed description below. This summary is not intended to identify the main features or essential features of the claimed subject matter, nor is it intended to help determine the scope of the claimed subject matter.

[0008] A method for predicting the risk of brain diseases according to an embodiment of the present disclosure may include: obtaining a learned brain disease risk analysis model; obtaining target shape information of the cerebrovascular of a target patient; obtaining target blood flow information of the cerebrovascular of the target patient; obtaining target patient information of the target patient; and inputting the target shape information, the target blood flow information, and the target patient information into the brain disease risk analysis model, and obtaining a brain disease risk value output through the brain disease risk analysis model.

[0009] The technical solution of the present disclosure is not limited to the technical solutions described above, and technical solutions not mentioned will be clearly understood by those of ordinary skill in the technical field to which the present disclosure pertains from the present specification and the accompanying drawings.

Advantages of the Invention

[0010] According to the learning method of the brain disease risk analysis model according to an embodiment of the present disclosure, it is possible to provide a brain disease risk analysis model that can automatically calculate or predict the brain disease risk based on the shape information data of cerebral blood vessels, the blood flow information of cerebral blood vessels, and / or the health information of a patient.

[0011] According to the brain disease risk prediction method according to an embodiment of the present disclosure, it is possible to calculate the personalized brain disease risk for a target patient in consideration of the shape information of the cerebral blood vessels of the target patient, the blood flow information related to the velocity and pressure of the blood flow of the target patient, and the patient information related to the age, gender, underlying diseases, race, etc. of the target patient.

[0012] The effects of the present disclosure are not limited to the effects described above, and the effects not mentioned will be clearly understood by those of ordinary skill in the technical field to which the present disclosure belongs from the present specification and the attached drawings.

Brief Description of the Drawings

[0013]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Mode for Carrying Out the Invention

[0014] The above-described technical problems, features, and advantages of the present disclosure will become more apparent through the following detailed description related to the accompanying drawings. However, the present disclosure can be modified in various ways and can have various embodiments. Hereinafter, specific embodiments will be illustrated in the drawings and will be described in detail.

[0015] Throughout the specification, the same reference numerals generally indicate the same components. Also, components having the same functions within the scope of the same concept shown in the drawings of each embodiment will be described using the same reference signs, and duplicate descriptions thereof will be omitted.

[0016] When it is determined that a specific description of a known function or configuration related to the present disclosure may unnecessarily obscure the gist of the present disclosure, the detailed description thereof will be omitted. Also, the numbers (for example, first, second, etc.) used in the description process of this specification are merely identification symbols for distinguishing one component from another component.

[0017] Also, the suffixes “module” and “unit” for the components used in the following embodiments are given or mixed only in consideration of the ease of preparing the specification, and do not have meanings or roles that are distinguished from each other by themselves.

[0018] In the following embodiments, the singular form includes the plural form unless the context clearly dictates otherwise.

[0019] In the following embodiments, terms such as "comprising" or "having" mean that the features or components described in the specification exist, and do not preclude in advance the possibility of adding one or more other features or components.

[0020] In the drawings, for the sake of convenience of explanation, the sizes of components may be exaggerated or reduced. For example, the sizes and thicknesses of each configuration shown in the drawings are arbitrarily shown for the convenience of explanation, and the present disclosure is not necessarily limited to what is shown.

[0021] When a certain embodiment can be embodied differently, the order of a specific process may be performed differently from the described order. For example, two processes described consecutively may be performed substantially simultaneously, or may proceed in the order opposite to the described order.

[0022] In the following embodiments, when components or the like are connected, it includes not only the case where the components are directly connected, but also the case where components are interposed between the components and indirectly connected.

[0023] For example, when components or the like are electrically connected in this specification, it includes not only the case where the components or the like are directly electrically connected, but also the case where components or the like are interposed therebetween and indirectly electrically connected.

[0024] A method for predicting the risk of brain disease according to an embodiment of the present disclosure may include: obtaining a brain disease risk analysis model for which learning has been completed; obtaining target shape information of the cerebrovascular vessels of a target patient; obtaining target blood flow information of the cerebrovascular vessels of the target patient; obtaining target patient information of the target patient; and inputting the target shape information, the target blood flow information, and the target patient information into the brain disease risk analysis model, and obtaining a brain disease risk value output through the brain disease risk analysis model.

[0025] According to an embodiment of the present disclosure, the step of obtaining the target shape information further includes: obtaining the shape information of the standard blood vessel; and obtaining the difference between the target shape information and the shape information of the standard blood vessel. The step of obtaining the brain disease risk value can further include: inputting the difference between the target shape information and the shape information of the standard blood vessel into the brain disease risk analysis model, and obtaining the brain disease risk value output through the brain disease risk analysis model. According to an embodiment of the present application, the step of obtaining the target blood flow information further includes: obtaining the blood flow information of the standard blood vessel; and obtaining the difference between the target blood flow information and the blood flow information of the standard blood vessel. The step of obtaining the brain disease risk value can further include: inputting the difference between the target blood flow information and the blood flow information of the standard blood vessel into the brain disease risk analysis model, and obtaining the brain disease risk value output through the brain disease risk analysis model.

[0026] According to an embodiment of the present disclosure, the brain disease risk analysis model can be trained based on a first learning dataset related to the shape information of the cerebral blood vessels including the position information of the cerebral blood vessels or the diameter information of the cerebral blood vessels, a second learning dataset related to the blood flow information of the cerebral blood vessels including the blood flow velocity information of the cerebral blood vessels or the blood pressure information of the cerebral blood vessels, a third learning dataset related to the patient information including the age of the patient or the gender of the patient, and a fourth learning dataset related to the brain disease information.

[0027] According to an embodiment of the present disclosure, the brain disease risk analysis model is configured to obtain the first learning dataset, the second learning dataset, the third learning dataset, and the fourth learning dataset and output a brain disease risk prediction value. The brain disease risk analysis model can be trained to output the brain disease risk prediction value so as to approximate the brain disease information included in the fourth learning dataset.

[0028] According to one embodiment of the present disclosure, the brain disease risk analysis model may be trained based on a first training dataset composed of the difference between the shape information of a patient's cerebral blood vessels and the shape information of standard cerebral blood vessels, a second training dataset composed of the difference between the blood flow information of the patient's cerebral blood vessels and the blood flow information of the standard cerebral blood vessels, a third training dataset related to patient information including the patient's age or the patient's gender, and a fourth training dataset related to brain disease information.

[0029] According to one embodiment of the present disclosure, the brain disease risk analysis model obtains the first training dataset, the second training dataset, the third training dataset, and the fourth training dataset, and is configured to output a brain disease risk prediction value, and the brain disease risk analysis model may be trained to output the brain disease risk prediction value so as to approximate the brain disease information included in the fourth training dataset.

[0030] According to one embodiment of the present disclosure, a computer-readable recording medium recording a program for executing the method for predicting the brain disease risk may be provided.

[0031] Hereinafter, with reference to FIGS. 1 to 9, a method for predicting the brain disease risk, a method for training a brain disease risk analysis model, and a brain disease risk analysis apparatus (or a brain disease risk analysis server, hereinafter referred to as the "brain disease risk analysis apparatus") according to an embodiment of the present disclosure will be described.

[0032] FIG. 1 is a schematic diagram related to a brain disease risk analysis apparatus according to an embodiment of the present disclosure.

[0033] The brain disease risk analysis apparatus 1000 according to an embodiment of the present disclosure can train a brain disease risk analysis model. Further, the brain disease risk analysis apparatus 1000 can calculate or predict the brain disease risk from a medical image using the trained brain disease risk analysis model.

[0034] A brain disease risk analysis device 1000 according to an embodiment of the present disclosure may include a transceiver 1100, a memory 1200, and a processor 1300.

[0035] The transceiver 1100 can communicate with any external device including a medical imaging device (for example, an MRA device, an MRI device, a CT device, a PET device, etc.). For example, the brain disease risk analysis device 1000 can acquire a medical image captured by the medical imaging device through the transceiver 1100. Further, the brain disease risk analysis device 1000 can acquire any execution data for causing the executed brain disease risk analysis model through the transceiver 1100. Here, the execution data may be in the sense of including a structure information, a hierarchical information, an operation library of the brain disease risk analysis model, and a parameter set related to a weighted value included in the brain disease risk analysis model, and including any appropriate data for causing the brain disease risk analysis model to execute. Further, the brain disease risk analysis device 1000 can transmit or output the brain disease risk analysis result obtained through the brain disease risk analysis model to any external device including a user terminal through the transceiver 1100.

[0036] The brain disease risk analysis device 1000 can be connected to a network through the transceiver unit 1100 to transmit and receive various data. The transceiver unit 1100 can mainly include a wired type and a wireless type. Since the wired type and the wireless type each have advantages and disadvantages, in some cases, both the wired type and the wireless type may be provided in the brain disease risk analysis device 1000. Here, in the case of the wireless type, a communication method of the WLAN (Wireless Local Area Network) series such as Wi-Fi can be mainly used. Or, in the case of the wireless type, cellular communication, for example, a communication method of the LTE or 5G series can be used. However, the wireless communication protocol is not limited to the above-mentioned examples, and any appropriate wireless type communication method can also be used. In the case of the wired type, LAN (Local Area Network) and USB (Universal Serial Bus) communication are typical examples, and other methods are also possible.

[0037] The memory 1200 can store various information. Various data can be stored in the memory 1200 temporarily or semi-permanently. Examples of the memory 1200 may include a hard disk drive (HDD), a solid state drive (SSD), a flash memory, a read-only memory (ROM), a random access memory (RAM), etc. The memory 1200 can be provided in a form built into the brain disease risk analysis device 1000 or in a removable form. In the memory 1200, various data necessary for the operation of the brain disease risk analysis device 1000 can be stored, including an operating program (OS: Operating System) for driving the brain disease risk analysis device 1000 and a program for operating each component of the brain disease risk analysis device 1000.

[0038] The processor 1300 can control the overall operation of the brain disease risk analysis device 1000. For example, the processor 1300 can perform operations such as acquiring pixel information of the original medical image described later, acquiring coordinate information from the pixel information, generating shape information of cerebral blood vessels, acquiring a predicted value of fluid flow (or blood flow information), training a brain disease risk analysis model, and / or predicting the brain disease risk, etc., to control the overall operation of the brain disease risk analysis device 1000. Specifically, the processor 1300 can load and execute a program for the overall operation of the brain disease risk analysis device 1000 from the memory 1200. The processor 1300 can be implemented by an AP (Application Processor), a CPU (Central Processing Unit), or a device similar thereto according to hardware, software, or a combination thereof. At this time, it can be provided in the form of an electronic circuit that processes electrical signals and performs a control function in terms of hardware, and can be provided in the form of a program or code that drives the hardware circuit in terms of software.

[0039] Hereinafter, with reference to FIGS. 2 and 3, various operations of the brain disease risk analysis device 1000 according to an embodiment of the present disclosure will be specifically described. FIG. 2 is a schematic diagram illustrating the operation of the brain disease risk analysis device 1000 according to an embodiment of the present disclosure. FIG. 3 is a drawing illustrating aspects of the operation of the brain disease risk analysis device 1000 according to an embodiment of the present disclosure.

[0040] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can acquire original medical images. Specifically, the brain disease risk analysis device 1000 can acquire original medical images (for example, MRA images, MRI images, and / or CT images, etc.) obtained through a medical image capturing device. At this time, the brain disease risk analysis device 1000 can acquire pixel information from the original medical images. Here, the pixel information can mean coordinate information composed of the x, y, and z coordinates of the original medical images and grey level information. On the other hand, the brain disease risk analysis device 1000 can acquire patient information stored in an external server (for example, a hospital server) or a database, and acquire pixel information of the original medical images from the original medical images included in the patient information.

[0041] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can acquire coordinate information corresponding to a blood vessel region from the pixel information of the original medical images. Specifically, the brain disease risk analysis device 1000 can acquire coordinate information related to a region of interest (for example, a region corresponding to a cerebral blood vessel) from pixel information whose grey level information included in the pixel information of the original medical images is smaller than or the same as a predetermined value. Here, the coordinate information can mean, as described above, the x, y, and z coordinates of the pixels constituting the original medical images. On the other hand, the coordinate information can be referred to as point information or point cloud in that it is related to information about a specific point of the original medical images.

[0042] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can generate shape information of cerebral blood vessels based on the coordinate information corresponding to the blood vessel region.

[0043] As an example, the brain disease risk analysis device 1000 can perform an operation of generating a surface model based on coordinate information corresponding to a blood vessel region. Specifically, the brain disease risk analysis device 1000 can generate a surface model indicating the shape information of cerebral blood vessels from the coordinate information corresponding to the region of interest by utilizing computer vision segmentation techniques. For example, the brain disease risk analysis device 1000 can be embodied to generate a surface model for the region of interest based on the coordinates constituting one surface among the coordinate information selected from pixel information. On the other hand, the original medical image can be composed of a set of 2D images and can include 3D coordinate information. At this time, the brain disease risk analysis device 1000 can generate a 3D surface model indicating the shape information of the region of interest (e.g., cerebral blood vessels) based on the 3D coordinate information.

[0044] As an example, the brain disease risk analysis device 1000 can perform an operation of generating a 1D model based on coordinate information corresponding to a blood vessel region. Specifically, the brain disease risk analysis device 1000 can generate a 1D model indicating the shape information of the region of interest (e.g., cerebral blood vessels) based on the coordinate information corresponding to the blood vessel region. Here, the 1D model refers to a model in which points corresponding to the coordinate information of the aforementioned region of interest are connected in an arbitrary manner.

[0045] As an example, the brain disease risk analysis device 1000 can generate a mesh model based on coordinate information corresponding to a blood vessel region. For example, the brain disease risk analysis device 1000 can generate a mesh model from the surface model generated based on the coordinate information.

[0046] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can obtain a predicted value of fluid flow based on coordinate information corresponding to a blood vessel region and shape information of cerebral blood vessels. Specifically, the brain disease risk analysis device 1000 can calculate a predicted value of fluid flow (or blood flow information) related to the velocity of blood flow and / or the pressure of blood flow based on the coordinate information corresponding to the blood vessel region and the shape information of cerebral blood vessels. As a result, the brain disease risk analysis device 1000 can obtain boundary conditions of cerebral blood vessels based on the shape information of cerebral blood vessels, and calculate or obtain blood flow information related to the velocity of blood flow and / or the pressure of blood flow based on the boundary conditions.

[0047] As an example, the brain disease risk analysis device 1000 can calculate a predicted value of fluid flow from shape data of cerebral blood vessels (for example, a mesh model and / or a surface model) by utilizing Computational Fluid Dynamics (CFD) or Computational Aided Engineering (CAE). As an example, the brain disease risk analysis device 1000 can calculate a predicted value of fluid flow from shape data of cerebral blood vessels (for example, a 1D model) through a mathematical formula assuming that the fluid flow is Poiseuille Flow.

[0048] On the other hand, the brain disease risk analysis device 1000 according to an embodiment of the present disclosure can perform operations of training a fluid flow model and predicting or calculating a fluid flow value through the fluid flow model.

[0049] As an example, the brain disease risk analysis device 1000 can generate a 1D model from a surface model by utilizing a computer vision skeleton technique, and generate a fluid flow model from the generated 1D model. Specifically, the brain disease risk analysis device 1000 trains a first neural network that generates a fluid flow model based on a 1D model by applying an artificial intelligence algorithm, and can obtain a fluid flow model from the 1D model through the trained first neural network.

[0050] As another example, the brain disease risk analysis device 1000 can generate a fluid flow model based on coordinate information. Specifically, the brain disease risk analysis device 1000 trains a second neural network that applies an artificial intelligence algorithm to generate a fluid flow model based on coordinate information, and can obtain a fluid flow model from the coordinate information through the trained second neural network.

[0051] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can perform an operation of training a brain disease risk analysis model. Specifically, the brain disease risk analysis device 1000 can train a brain disease risk analysis model based on the shape information of cerebral blood vessels, the predicted value of fluid flow (or blood flow information), and / or patient information (for example, the patient's age, gender, brain disease risk, etc.). As a result, the brain disease risk analysis device 1000 according to an embodiment of the present disclosure can obtain a patient-ordered type brain disease risk prediction result using the trained brain disease risk analysis model, and perform an operation of transmitting or outputting the patient-ordered type brain disease risk prediction result. The method of training the brain disease risk analysis model will be described in detail later in relation to FIGS. 4 to 9.

[0052] Hereinafter, with reference to FIGS. 4 and 5, a learning method of a brain disease risk analysis model according to an embodiment of the present application will be specifically described. FIG. 4 is a schematic diagram illustrating the operation of the brain disease risk analysis device 1000 for learning a brain disease risk analysis model according to an embodiment of the present disclosure. FIG. 5 is a drawing illustrating a state of learning a brain disease risk analysis model according to an embodiment of the present disclosure.

[0053] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can perform an operation of training a brain disease risk analysis model for calculating the brain disease risk from medical images. Specifically, the brain disease risk analysis device 1000 can perform an operation of learning a brain disease risk analysis model for calculating or quantifying the brain disease risk based on the shape information of cerebral blood vessels, the blood flow information of cerebral blood vessels, and patient information acquired from medical images. More specifically, the brain disease risk analysis device 1000 uses a first learning data set X related to the shape information (xn) of blood vessels including the position information of blood vessels and the diameter information of blood vessels, a second learning data set Y related to the blood flow information (vn) including the velocity information of blood flow and the pressure information of blood flow, a third learning data set R composed of patient information (rn) related to the age of the patient, the gender of the patient, and / or the presence or absence of brain disease, and a fourth learning data set related to the risk of brain disease (yn) for each patient to learn the brain disease risk analysis model and can acquire the learned brain disease risk analysis model.

[0054] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can acquire a first learning data set related to the shape information of cerebral blood vessels (for example, the position information of blood vessels and / or the diameter information of blood vessels), and can train a brain disease risk analysis model based on the first learning data set.

[0055] As an example, the first learning data set can be composed of the shape information of blood vessels of each patient. For example, the first learning data set can include the first shape information of the first patient, the second shape information of the second patient, and / or the Nth shape information of the Nth patient.

[0056] As another example, the first learning data set can be composed of the difference between the shape information of the blood vessels of the patient and the shape information of the standard blood vessels. For example, the first learning data set can be composed of the difference between the first shape information of the blood vessels of the first patient and the shape information of the standard blood vessels, the difference between the second shape information of the blood vessels of the second patient and the shape information of the standard blood vessels, and / or the difference between the Nth shape information of the blood vessels of the Nth patient and the shape information of the standard blood vessels. At this time, the shape information of the standard blood vessels can be the shape information of the standard blood vessels corresponding to the gender and age of the patient.

[0057] However, the above-described content is merely an example, and the brain disease risk analysis model may be configured to be trained using a first training dataset appropriately configured in any manner to achieve the purpose of training a brain disease risk analysis model for predicting the brain disease risk.

[0058] The brain disease risk analysis apparatus 1000 according to an embodiment of the present disclosure can acquire a second training dataset related to blood flow information of cerebral blood vessels (for example, blood flow velocity information, blood pressure information of blood flow), and train a brain disease risk analysis model based on the second training dataset.

[0059] As an example, the second training dataset may be composed of blood flow information of blood vessels of each patient. For example, the second training dataset may include first blood flow information of a first patient, second blood flow information of a second patient, and / or Nth blood flow information of an Nth patient.

[0060] As another example, the second training dataset may be composed of the difference between the blood flow information of the blood vessels of a patient and the blood flow information of a standard blood vessel. For example, the second training dataset may be composed of the difference between the first blood flow information of the blood vessels of a first patient and the blood flow information of a standard blood vessel, the difference between the second blood flow information of the blood vessels of a second patient and the blood flow information of a standard blood vessel, and / or the difference between the Nth blood flow information of the blood vessels of an Nth patient and the blood flow information of a standard blood vessel. At this time, the standard blood vessel may be a standard blood vessel corresponding to the gender and age of the patient.

[0061] However, the above-described content is merely an example, and the brain disease risk analysis model may be configured to be trained using a second training dataset appropriately configured in any manner to achieve the purpose of training a brain disease risk analysis model for predicting the brain disease risk.

[0062] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can acquire a third learning dataset related to patient information and use the third learning dataset to train a brain disease risk analysis model. Specifically, the third learning dataset can be composed of patient data for at least one of the patient's age, the patient's gender, the patient's underlying diseases, or the patient's ethnicity. Specifically, the brain disease risk analysis device 1000 can acquire the third learning dataset based on patient information obtained from an external server (for example, a hospital server).

[0063] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can acquire a fourth learning dataset related to brain disease-specific risk information and use the fourth learning dataset to train a brain disease risk analysis model. For example, the brain disease risk analysis device 1000 can assign brain disease risk information (for example, a brain disease probability value) based on the position information included in the blood vessel shape information and the brain disease information included in the patient information. More specifically, the brain disease risk analysis device 1000 can detect a brain region related to a brain disease based on the brain disease information included in the patient information, and assign brain disease risk information (for example, a brain disease probability value) to the position information corresponding to the detected brain region to acquire the fourth learning dataset. As an example, the brain disease risk analysis device 1000 can calculate the probability value of a brain disease occurring for each brain region based on the patient information, and assign the probability value (for example, a value between 0 and 1) calculated using the position information included in the blood vessel shape information to each brain region. For example, in the case of a patient having a stroke disease, the brain disease risk analysis device 1000 can detect the brain region where the stroke actually occurred based on the patient information, and assign stroke risk information (for example, risk level 1) indicating that a stroke has occurred to the position information corresponding to the detected brain region. For example, the brain disease risk analysis device 1000 can detect the brain regions where no stroke has occurred based on the patient information, and assign stroke risk information (for example, risk level 0) indicating that no stroke has occurred to the position information corresponding to the detected brain regions.

[0064] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can train a brain disease risk analysis model based on a first learning dataset related to the shape information of cerebral blood vessels, a second learning dataset related to the blood flow information of cerebral blood vessels, a third learning dataset related to patient information, and / or a fourth learning dataset related to brain disease-specific risk information.

[0065] As an example, the brain disease risk analysis device 1000 can train a brain disease risk analysis model based on the cerebral blood vessel data of a patient (for example, the shape information of cerebral blood vessels or the blood flow information of cerebral blood vessels). This will be described more specifically with reference to FIG. 8.

[0066] As another example, the brain disease risk analysis device 1000 can train a brain disease risk analysis model based on the cerebral blood vessel data of a patient (for example, the shape information of cerebral blood vessels or the blood flow information of cerebral blood vessels) and the cerebral blood vessel data of a standard blood vessel (for example, the shape information of a standard cerebral blood vessel or the blood flow information of a standard cerebral blood vessel). Specifically, the brain disease risk analysis device 1000 can train a brain disease risk analysis model based on the difference between the cerebral blood vessel data of a patient (for example, the shape information of cerebral blood vessels or the blood flow information of cerebral blood vessels) and the cerebral blood vessel data of a standard blood vessel (for example, the shape information of a standard cerebral blood vessel or the blood flow information of a standard cerebral blood vessel). For example, the brain disease risk analysis device 1000 can train a brain disease risk analysis model using a learning dataset related to the difference between the shape information of the cerebral blood vessels of a patient and the shape information of a standard cerebral blood vessel. For example, the brain disease risk analysis device 1000 can train a brain disease risk analysis model using a learning dataset related to the difference between the blood flow information of the cerebral blood vessels of a patient and the blood flow information of a standard cerebral blood vessel. This will be described more specifically with reference to FIG. 9.

[0067] On the other hand, FIGS. 4 and 5 describe the brain disease risk analysis model as a single model. However, this is only for the convenience of explanation and should not be interpreted restrictively. It can be configured such that separate models are trained for each learning dataset, and the brain disease risk can be calculated integrally through a structure in which some layers of each model are shared.

[0068] For example, the first brain disease risk analysis model may be configured to obtain a fourth learning dataset related to the risk of each brain disease obtained from the first learning dataset and patient information, and output a predicted value of the brain disease risk. At this time, the first brain disease risk analysis model may be trained by updating any parameter included in the first brain disease risk analysis model based on the predicted value of the brain disease risk and the fourth learning dataset related to the risk of each brain disease.

[0069] For example, the second brain disease risk analysis model may be configured to obtain a fourth learning dataset related to the risk of each brain disease obtained from the second learning dataset and patient information, and output a predicted value of the brain disease risk. At this time, the second brain disease risk analysis model may be trained by updating any parameter included in the second brain disease risk analysis model based on the predicted value of the brain disease risk and the fourth learning dataset related to the risk of each brain disease.

[0070] At this time, a part of the layers of the first brain disease risk analysis model and a part of the layers of the second brain disease risk analysis model may be configured to have a structure shared with each other, and through this, the risk of brain diseases may be predicted by considering all the shape information and blood flow information of the cerebral blood vessels.

[0071] Hereinafter, with reference to FIGS. 6 to 9, a learning method of a brain disease risk analysis model and a prediction method of a brain disease risk according to an embodiment of the present disclosure will be described. FIG. 6 is a flowchart illustrating a learning method of a brain disease risk analysis model according to an embodiment of the present disclosure. In describing the learning method of the brain disease risk analysis model, some embodiments overlapping with the descriptions described in FIGS. 2 to 5 may be omitted. However, this is only for convenience of explanation and should not be construed restrictively.

[0072] The method for training a brain disease risk analysis model according to an embodiment of the present disclosure may include a step (S1100) of obtaining a first training dataset related to the shape information of cerebral blood vessels, a step (S1200) of obtaining a second training dataset related to the blood flow information of cerebral blood vessels, a step (S1300) of obtaining a third training dataset related to patient information, a step (S1400) of obtaining a fourth training dataset related to the risk for each brain disease, and a step (S1500) of training a brain disease risk analysis model using the first training dataset, the second training dataset, the third training dataset, or the fourth training dataset.

[0073] In the step (S1100) of obtaining the first training dataset related to the shape information of cerebral blood vessels, the brain disease risk analysis apparatus 1000 can obtain a first training dataset X related to the shape information of cerebral blood vessels (e.g., the position information of blood vessels, the diameter information of blood vessels). As described above, as an example, the first training dataset may be composed of the shape information of blood vessels of each patient. For example, the first training dataset X may include the first shape information of the first patient, the second shape information of the second patient, and / or the Nth shape information of the Nth patient. As another example, the first training dataset may be composed of the difference between the shape information of the blood vessels of a patient and the shape information of a standard blood vessel. For example, the first training dataset may be composed of the difference between the first shape information of the blood vessels of the first patient and the shape information of the standard blood vessel, the difference between the second shape information of the blood vessels of the second patient and the shape information of the standard blood vessel, and / or the difference between the Nth shape information of the blood vessels of the Nth patient and the shape information of the standard blood vessel.

[0074] In the step of obtaining the second learning dataset related to the blood flow information of the cerebral blood vessels (S1200), the cerebral disease risk analysis device 1000 can obtain a second learning dataset related to the blood flow information of the cerebral blood vessels (for example, blood flow velocity information, blood flow pressure information). As an example, the second learning dataset can be composed of the blood flow information of the blood vessels of each patient. For example, the second learning dataset can include the first blood flow information of the first patient, the second blood flow information of the second patient, and / or the Nth blood flow information of the Nth patient. As another example, the second learning dataset can be composed of the difference between the blood flow information of the blood vessels of the patient and the blood flow information of the standard blood vessels. For example, the second learning dataset can be composed of the difference between the first blood flow information of the blood vessels of the first patient and the blood flow information of the standard blood vessels, the difference between the second blood flow information of the blood vessels of the second patient and the blood flow information of the standard blood vessels, and / or the difference between the Nth blood flow information of the blood vessels of the Nth patient and the blood flow information of the standard blood vessels.

[0075] In the step of obtaining the third learning dataset related to the patient information (S1300), the cerebral disease risk analysis device 1000 can obtain a third learning dataset related to the patient information (for example, the age of the patient, the gender of the patient, the underlying diseases of the patient, and / or the race of the patient).

[0076] In the step of obtaining the fourth learning dataset related to the risk degree for each cerebral disease (S1400), the cerebral disease risk analysis device 1000 can obtain a fourth learning dataset related to the risk degree information for each cerebral disease. As described above, the fourth learning dataset can be related to the risk degree information for cerebral diseases (for example, cerebral disease probability values) assigned based on the position information included in the blood vessel shape information and the cerebral disease information included in the patient information. Specifically, the cerebral disease risk analysis device 1000 can calculate the probability values of cerebral diseases occurring for each cerebral region based on the patient information, and use the position information included in the blood vessel shape information to assign the calculated probability values (for example, values between 0 and 1) to each cerebral region to obtain the fourth learning dataset.

[0077] In the step (S1500) of training the brain disease risk analysis model using the first learning dataset, the second learning dataset, the third learning dataset, or the fourth learning dataset, the brain disease risk analysis apparatus 1000 can train the brain disease risk analysis model based on the first learning dataset related to the shape information of the cerebral blood vessels, the second learning dataset related to the blood flow information of the cerebral blood vessels, the third learning dataset related to patient information, and / or the fourth learning dataset related to the risk levels for different brain diseases. Specifically, the brain disease risk analysis apparatus 1000 can train the brain disease risk analysis model based on the predicted value output through the brain disease risk analysis model and the risk levels for different brain diseases included in the fourth learning dataset. For example, the brain disease risk analysis apparatus 1000 can update the parameters included in the brain disease risk analysis model so that the predicted value output through the brain disease risk analysis model approximates the risk levels for different brain diseases included in the fourth learning dataset, thereby training the brain disease risk analysis model.

[0078] Refer to FIG. 7. FIG. 7 is a drawing for explaining the aspect of training the brain disease risk analysis model and the aspect of verifying the brain disease risk analysis model according to an embodiment of the present disclosure. The brain disease risk analysis apparatus 1000 according to an embodiment of the present disclosure can perform an operation of preprocessing the learning data composed of the first learning dataset, the second learning dataset, the third learning dataset, and / or the fourth learning dataset. For example, the brain disease risk analysis apparatus 1000 can perform normalization on each data included in the learning data composed of the first learning dataset, the second learning dataset, the third learning dataset, and / or the fourth learning dataset.

[0079] Subsequently, the brain disease risk analysis apparatus 1000 according to an embodiment of the present application can perform an operation of dividing the preprocessed learning data into a training dataset and a test dataset or a validation dataset. For example, the brain disease risk analysis apparatus 1000 can randomly divide the preprocessed learning data into a training dataset and a validation dataset. At this time, the brain disease risk analysis apparatus 1000 can train a brain disease risk analysis model using the training dataset through any machine learning algorithm (ML Algorithm). At this time, the brain disease risk analysis apparatus 1000 can test or verify the performance of the brain disease risk analysis model learned using the validation dataset. For example, the brain disease risk analysis apparatus 1000 calculates the performance of the learned brain disease risk analysis model, compares the calculated performance with a predetermined target performance, and performs the learning of the above-described brain disease risk analysis model until the calculated performance becomes greater than the predetermined target performance, thereby obtaining an optimized brain disease risk analysis model.

[0080] FIG. 8 is a diagram for explaining a state of training a brain disease risk analysis model according to an embodiment of the present disclosure and a state of predicting a brain disease risk using the learned brain disease risk analysis model.

[0081] The brain disease risk analysis device 1000 according to an embodiment of the present disclosure can train a brain disease risk analysis model f based on a first learning dataset X related to the shape information of the patient's blood vessels obtained from the patient's blood vessel image, a second learning dataset V related to the patient's blood flow information, a third learning dataset R related to the patient information, and a fourth learning dataset Y1 related to the risk information for each brain disease obtained from the patient information. Specifically, the brain disease risk analysis device 1000 can input the first learning dataset X, the second learning dataset V, the third learning dataset R, and the fourth learning dataset Y1 into the brain disease risk analysis model f, and obtain the predicted value of the brain disease risk output through the brain disease risk analysis model f. At this time, the brain disease risk analysis device 1000 can update any parameter included in the brain disease risk analysis model based on the predicted value of the brain disease risk and the risk information for each brain disease. Specifically, the brain disease risk analysis device 1000 compares the predicted value of the brain disease risk with the risk information Y1 for each brain disease, and updates any parameter included in the brain disease risk analysis model f so that the predicted value of the brain disease risk is approximately output to the risk information for each brain disease, and trains the brain disease risk analysis model f to obtain a trained brain disease risk analysis model f″.

[0082] Furthermore, the brain disease risk analysis device 1000 according to an embodiment of the present disclosure can predict the brain disease risk using the trained brain disease risk analysis model f″. Specifically, the brain disease risk analysis device 1000 acquires target shape information TX obtained from the vascular image of a target patient to be analyzed, target blood flow information TV of the target patient, and target patient information TR of the target patient, and can be configured to input the target shape information TX, the target blood flow information TV, and the target patient information TR into the trained brain disease risk analysis model f″. At this time, the brain disease risk analysis device 1000 can acquire the brain disease risk value Y output through the trained brain disease risk analysis model f″. At this time, as described above, the trained brain disease risk analysis model f″ is trained to output the risk information Y1 for each brain disease based on the shape information X of the blood vessel, the blood flow information V of the blood vessel, and the patient information R. Therefore, the brain disease risk value Y can be output based on the target shape information TX of the target patient, the target blood flow information TV of the target patient, and the target patient information TR of the target patient.

[0083] FIG. 9 is a drawing for explaining an aspect of training a brain disease risk analysis model according to another embodiment of the present disclosure and an aspect of predicting the brain disease risk using the learned brain disease risk analysis model.

[0084] The brain disease risk analysis device 1000 according to another embodiment of the present disclosure can train a brain disease risk analysis model f based on a first training dataset X2 composed of the difference between the shape information of the patient's blood vessels obtained from the patient's blood vessel image and the shape information of the standard blood vessels, a second training dataset V2 composed of the difference between the blood flow information of the patient and the blood flow information of the standard blood vessels, a third training dataset R related to patient information, and a fourth training dataset Y1 related to the risk information for each brain disease obtained from the patient information. Specifically, the brain disease risk analysis device 1000 can input the first training dataset X2, the second training dataset V2, the third training dataset R, and the fourth training dataset Y1 into the brain disease risk analysis model f, and obtain the predicted value of the brain disease risk output through the brain disease risk analysis model f. At this time, the brain disease risk analysis device 1000 can update any parameter included in the brain disease risk analysis model based on the predicted value of the brain disease risk and the risk information Y1 for each brain disease. Specifically, the brain disease risk analysis device 1000 compares the predicted value of the brain disease risk with the risk information Y1 for each brain disease, and updates any parameter included in the brain disease risk analysis model f so that the predicted value of the brain disease risk is approximately output to the risk information Y1 for each brain disease, thereby training the brain disease risk analysis model f and obtaining the trained brain disease risk analysis model f″.

[0085] Furthermore, the brain disease risk analysis apparatus 1000 according to another embodiment of the present disclosure can predict the brain disease risk using the learned brain disease risk analysis model f″. Specifically, the brain disease risk analysis apparatus 1000 acquires the difference TX2 between the target shape information obtained from the vascular image of the target patient to be analyzed and the shape information of the standard blood vessel, the difference TV2 between the target blood flow information of the target patient and the blood flow information of the standard blood vessel, and the target patient information TR of the target patient, and can be configured to input the difference TX2 between the target shape information and the shape information of the standard blood vessel, the difference TV2 between the target blood flow information and the blood flow information of the standard blood vessel, and the target patient information TR into the learned brain disease risk analysis model f″. At this time, the brain disease risk analysis apparatus 1000 can acquire the brain disease risk value Y output through the learned brain disease risk analysis model f″. At this time, as described above, the learned brain disease risk analysis model f″ is trained to output the risk information for each brain disease based on the first training dataset composed of the difference X2 between the shape information of the blood vessel and the shape information of the standard blood vessel, the second training dataset composed of the difference V2 between the blood flow information of the blood vessel and the blood flow information of the standard blood vessel, and the patient information R. Therefore, the brain disease risk value Y can be output based on the difference TX2 between the target shape information and the shape information of the standard blood vessel, the difference TV2 between the target blood flow information and the blood flow information of the standard blood vessel, and the target patient information TR.

[0086] The learned brain disease risk analysis model f″ can provide the effect of outputting the patient-ordered type brain disease risk individualized for the target patient in consideration of the shape information of the cerebral blood vessels of the target patient, the blood flow information related to the blood flow velocity and pressure of the target patient, and the patient information related to the age, gender, underlying diseases, race, etc. of the target patient.

[0087] On the other hand, in FIG. 8, the content of predicting the brain disease risk based on the cerebral blood vessel data of the patient (for example, the shape information of the cerebral blood vessels or the blood flow information of the cerebral blood vessels) is mainly described, and in FIG. 9, the content of predicting the brain disease risk based on the difference between the cerebral blood vessel data of the patient and the standard cerebral blood vessel data is mainly described. However, this is only an example and can be selected, combined, or changed by any appropriate method.

[0088] According to the learning method of the brain disease risk analysis model according to an embodiment of the present disclosure, it is possible to provide a brain disease risk analysis model that can automatically calculate or predict the brain disease risk based on the shape information of cerebral blood vessels, the blood flow information of cerebral blood vessels, and / or patient information.

[0089] The various operations of the brain disease risk analysis device 1000 described above can be stored in the memory 1200 of the brain disease risk analysis device 1000, and the processor 1300 of the brain disease risk analysis device 1000 can be provided to perform the operations stored in the memory 1200.

[0090] The features, structures, effects, etc. described in the embodiments above are included in at least one embodiment of the present disclosure and are not necessarily limited to only one embodiment. Thus, the features, structures, effects, etc. exemplified in each embodiment can be combined or modified and implemented in other embodiments by those with ordinary knowledge in the field to which the embodiments belong. Therefore, the content related to such combinations and modifications should be interpreted as being included in the scope of the present disclosure.

[0091] Also, although the embodiments have been mainly described above, this is merely an example and does not limit the present disclosure. It will be understood by those with ordinary knowledge in the field to which the present disclosure belongs that various modifications and applications not exemplified above are possible without departing from the essential characteristics of this embodiment. That is, each component specifically shown in the embodiment can be implemented with modifications. And the differences related to such modifications and applications should be interpreted as being included in the scope of the present disclosure defined by the appended claims.

Industrial Applicability

[0092] The method for predicting the brain disease risk, the learning method of the brain disease risk analysis model, and the brain disease risk analysis device as described above can be applied to the medical service field that provides medical services.

Claims

1. In a method for predicting the risk of brain diseases by a brain disease risk analysis device, obtaining a trained brain disease risk analysis model; obtaining target shape information of the cerebral blood vessels of a target patient; obtaining target blood flow information of the cerebral blood vessels of a target patient; obtaining target patient information of a target patient; and inputting the target shape information, the target blood flow information, and the target patient information into the brain disease risk analysis model, and obtaining a brain disease risk value output through the brain disease risk analysis model; A method for predicting the risk of brain diseases, comprising:

2. The step of obtaining the target shape information further includes: obtaining shape information of a standard blood vessel; and obtaining the difference between the target shape information and the shape information of the standard blood vessel; The step of obtaining the brain disease risk value further includes: inputting the difference between the target shape information and the shape information of the standard blood vessel into the brain disease risk analysis model, and obtaining a brain disease risk value output through the brain disease risk analysis model; The method for predicting the risk of brain diseases according to claim 1.

3. The step of obtaining the target blood flow information further includes: obtaining blood flow information of a standard blood vessel; and obtaining the difference between the target blood flow information and the blood flow information of the standard blood vessel; The step of obtaining the brain disease risk value further includes: inputting the difference between the target blood flow information and the blood flow information of the standard blood vessel into the brain disease risk analysis model, and obtaining a brain disease risk value output through the brain disease risk analysis model; The method for predicting the risk of brain diseases according to claim 2.

4. The brain disease risk analysis model is Based on a first learning dataset related to shape information of cerebral blood vessels including position information of cerebral blood vessels or diameter information of cerebral blood vessels, a second learning dataset related to blood flow information of cerebral blood vessels including blood flow velocity information of cerebral blood vessels or blood flow pressure information of cerebral blood vessels, a third learning dataset related to patient information including the age of the patient or the gender of the patient, and a fourth learning dataset related to brain disease information. The method for predicting the risk of brain diseases according to claim 1, which is trained.

5. The brain disease risk analysis model is configured to obtain the first learning dataset, the second learning dataset, the third learning dataset, and the fourth learning dataset and output a predicted value of the brain disease risk, The brain disease risk analysis model is The method for predicting the risk of brain disease according to claim 4, which is trained to output the brain disease risk prediction value so as to approximate the brain disease information included in the fourth learning data set.

6. The brain disease risk analysis model is The method for predicting the risk of brain disease according to claim 3, which is trained based on a first learning data set composed of the difference between the shape information of the patient's cerebral blood vessels and the shape information of the standard cerebral blood vessels, a second learning data set composed of the difference between the blood flow information of the patient's cerebral blood vessels and the blood flow information of the standard cerebral blood vessels, a third learning data set related to patient information including the patient's age or the patient's gender, and a fourth learning data set related to brain disease information.

7. The brain disease risk analysis model is configured to obtain the first learning data set, the second learning data set, the third learning data set, and the fourth learning data set and output a brain disease risk prediction value. The brain disease risk analysis model is The method for predicting the risk of brain disease according to claim 6, which is trained to output the brain disease risk prediction value so as to approximate the brain disease information included in the fourth learning data set.

8. A computer-readable recording medium recording a program for causing a computer to execute the method according to any one of claims 1 to 7.

Images