KR20200052123A - blood vessel age diagnosis apparatus, blood vessel age diagnosis method, and computer program for executing the method - Google Patents

Abstract

According to the present invention, provided is a blood vessel age diagnosis method which comprises the steps of: allowing a blood vessel age diagnosis device to receive an image of photographing a subject from an image photographing device; allowing the blood vessel age diagnosis device to extract a three-dimensional binary blood vessel image; allowing the blood vessel age diagnosis device to extract blood vessel segmentation regions for each anatomically separated region from the three-dimensional binary vessel image; allowing the blood vessel age diagnosis device to extract a center line connecting center points of a first blood vessel section by using the blood vessel segmentation regions; allowing the blood vessel age diagnosis device to calculate a warping value based on the center line; and allowing the blood vessel age diagnosis device to infer a blood vessel age of the subject based on the warping value.

Description

A blood vessel age diagnosis apparatus, a blood vessel age diagnosis apparatus, and a computer program for executing the method

According to the present specification, a device for diagnosing vascular age, a method for diagnosing vascular age, and a computer program for executing the method.

Magnetic resonance imaging (MRI) imaging device is a device that photographs an object using a magnetic field, and it shows the bone, as well as the disc, joints, nerve ligaments, heart, and cerebral blood vessels in three dimensions from the desired angle. It is widely used for diagnosis. Magnetic resonance imaging has the advantage of being able to obtain various contrast ratios by adjusting several parameters, and by using this, clinical diagnosis is performed by obtaining images of multiple contrast ratios for the same site. Korean Patent Publication No. 2009-0075644 discloses a magnetic resonance imaging device that obtains a steady-state image of a patient by changing the spin phases of the patient's fat and moisture to generate a contrast ratio of the magnetic resonance image.

One embodiment of the present invention is to provide a blood vessel age diagnosis apparatus, method, and computer program for diagnosing blood vessel age of a subject based on an image of blood vessels of a subject, particularly cerebral blood vessels.

A method of diagnosing vascular age according to embodiments of the present invention includes the steps of: receiving, by an vascular age diagnostic apparatus, an image of an object taken from an imaging apparatus; Extracting a 3D binary blood vessel image from the image by the vessel age diagnosis apparatus; Extracting vascular segmentation regions for each anatomically separated region from the 3D binary vessel image by the vascular age diagnosis apparatus; Extracting a center line connecting the center points of the first blood vessel section by using the blood vessel segmentation regions; Calculating, by the vascular age diagnosis apparatus, a warping value of a blood vessel based on the center line; And the blood vessel age diagnosis apparatus inferring the blood vessel age of the subject based on the warpage value.

The step of extracting the 3D binary vascular image is performed by interpolating the x, y, and z components of each voxel of the image, correcting the z component of each voxel, and using the corrected result, each voxel of the image. By correcting the x, y, and z components of and changing the position of each voxel with the corrected x, y, and z components, it is possible to extract a 3D binary blood vessel image.

The step of inferring the vascular age may be characterized in that the vascular age of the object is inferred using a vascular age diagnosis model that is trained as a warping value of vascular segmentation regions for each region included in the image.

The step of extracting the centerline is a point to quantify the degree of bending (twist or tortuosity) of a blood vessel by tracking the center points connecting the start and end points set by user inputs on the 3D binary blood vessel image (on). It can be characterized by. In addition, the length of blood vessels (branch length) can be calculated as a feature (feature). Furthermore, more complicatedly, values calculated from the start point to the end point of the diameter or area of the cross section of the blood vessel passing through the center point may be calculated as feature points.

After the reasoning step, a step of evaluating the reasoning result of the blood vessel age by comparing the blood vessel age of the subject with the actual subject age may be included.

The present embodiment may further include the step of overlapping the center line extracted on the image and the 3D binary blood vessel image and outputting it through an input / output unit.

The computer program according to the embodiment of the present invention may be stored in a medium to execute any one of the methods of diagnosing vascular age according to the embodiment of the present invention using a computer.

In addition to this, another method for implementing the present invention, another system, and a computer-readable recording medium for recording a computer program for executing the method are further provided.

Other aspects, features and advantages other than those described above will become apparent from the following drawings, claims and detailed description of the invention.

According to an embodiment of the present invention, it is possible to infer the vascular age of a subject based on the distortion values of vascular stems extracted from an image.

In addition, according to an embodiment of the present invention, a blood vessel age recognition model may be trained from an image, and a blood vessel age of a subject may be inferred using the trained blood vessel age diagnosis model.

1 is a block diagram of a vascular age diagnosis system according to embodiments of the present invention.
2 is a block diagram of an apparatus for diagnosing vascular age according to embodiments of the present invention.
FIG. 3 is a diagram for explaining the structure of the storage unit of the vascular age diagnosis apparatus and the learning model apparatus, and data transmission and reception between the storage unit and the learning model apparatus.
4 is a block diagram of a data acquisition unit and a data recognition unit of the learning model device.
5 to 6 are flowcharts of a method for diagnosing vascular age according to embodiments of the present invention.
7 to 9 are exemplary views of a user interface provided through a vascular age diagnosis device.
10 is an exemplary diagram for inferring vascular age through a model trained from a warpage value of a blood vessel.

Hereinafter, exemplary embodiments according to the present invention will be described in detail with reference to the contents described in the accompanying drawings. Also, a method of configuring and using an electronic device according to an embodiment of the present invention will be described in detail with reference to contents described in the accompanying drawings. The same reference numerals or symbols in each drawing denote parts or components that perform substantially the same function.

Terms including ordinal numbers such as first and second may be used to describe various components, but the components are not limited by the terms. The terms are only used to distinguish one component from other components. For example, the first component may be referred to as a second component without departing from the scope of the present invention, and similarly, the second component may be referred to as a first component. The term and / or includes a combination of a plurality of related items or any one of a plurality of related items.

Terms used in the present specification are used to describe examples, and are not intended to limit and / or limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, terms such as include or have are intended to designate the existence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, and one or more other features, numbers, or steps. It should be understood that it does not preclude the existence or addition possibility of the operation, components, parts or combinations thereof.

Throughout the specification, when a part is connected to another part, this includes not only the case of being directly connected, but also the case of being electrically connected with another element in between. Also, when a part includes a certain component, this means that other components may be further included rather than excluding other components unless otherwise specified. In addition, terms such as “... unit” and “module” described in the specification mean a unit that processes at least one function or operation, which may be implemented in hardware or software, or a combination of hardware and software. .

In this specification, an image may mean multidimensional data composed of discrete image elements (eg, pixels in a 2D image and voxels in a 3D image). For example, the image may include an X-ray device, a CT device, an MRI device, an ultrasound diagnostic device, and a medical image of an object acquired by another medical imaging device.

In the present specification, the subject may include a person or an animal, or a part of a person or animal. For example, the subject may include blood vessels or devices such as the liver, heart, uterus, brain, breast, and abdomen. In addition, the object may include a phantom. Phantom refers to a material having a volume very close to the density and effective atomic number of an organism, and may include a spherical phantom having properties similar to a body.

In the present specification, the user is a medical professional, and may be a doctor, a nurse, a clinical pathologist, a medical imaging expert, or the like, but may be a technician repairing a medical device, but is not limited thereto.

In this specification, time-of-flight (TOF) magnetic resonance angiography (MRA) is an imaging technique that emphasizes the flow of blood flow in a subject without using a contrast agent, and is widely used to image intracranial cerebral artery vessels. This is the technique used. Since TOF MRA is obtained by shortening the repetition time, the stationary tissue decreases due to the saturation of the signal strength due to the application of repetitive radiofrequency (RF) pulses for a short time, but moves along the blood vessels. Since the water molecule to be out of the slice to which the RF pulse is applied does not experience repetitive RF pulses, the signal strength is relatively high. Therefore, there is an advantage in that it is easy to extract blood vessels based on signal strength from an image obtained using a TOF MRA.

1 is a block diagram of a vascular age diagnosis system according to embodiments of the present invention.

The vascular age diagnosis system 10 may include an imaging device 100 and a vascular age diagnosis device 200.

The vascular age diagnosis system 10 transmits an image of the object 1 acquired through the image taking device 100 to the vascular age diagnosis device 200, and the vascular age diagnosis device 200 transmits data obtained from the blood vessels. The blood vessel age of the subject 1 is determined based on the basis, and the blood vessel health of the subject is diagnosed based on the blood vessel age.

The imaging apparatus 100 is an apparatus for outputting an image of the object 1, and may be an X-ray apparatus, a CT apparatus, an MRI apparatus, an ultrasound diagnostic apparatus, and other medical imaging apparatus. The imaging apparatus 100 may acquire blood vessel images of all or part of the object 1. The imaging apparatus 100 may photograph an object and output a TOF image, diffusion weighted imaging (DWI), or apparent diffusion imaging (ADC).

The vascular age diagnosis apparatus 200 may extract input data necessary for vascular age diagnosis from the image of the subject, and infer the vascular age of the subject using the learned vascular age diagnosis model using the input data. The vascular age diagnosis apparatus 200 may calculate a degree of warping of a blood vessel using a blood vessel segmentation area and / or a blood vessel center line obtained from an input image, and determine a blood vessel age of the object based on the blood vessel age. The blood vessel age diagnosis apparatus 200 is a blood vessel database (not shown) stored in an external data server 300 and may determine blood vessel age using blood vessel map information of each object. The vascular age diagnosis apparatus 200 may convert vascular health of the diagnosed object into a predetermined standard and transmit the vascular health to an external blood vessel database (not shown). The vascular age diagnosis apparatus 200 may be a computing device including one or more processors and storage media.

2 is a block diagram of the vascular age diagnosis apparatus 200 according to embodiments of the present invention, and FIG. 3 is a structure of the vascular age diagnosis apparatus 250 and a storage unit 250 of the learning model device 300. ) Is a diagram for explaining data transmission and reception between the learning model device 300.

The vascular age diagnosis apparatus 200 may include a control unit 210, a communication unit 220, an input / output unit 240, and a storage unit 250.

The control unit 210 may be implemented with one or more processors, and configured to process instructions of a computer program by performing basic arithmetic, logic, and input / output operations. The command may be provided to the control unit 210 by the storage unit 250 and the communication unit 220. For example, the control unit 210 may be configured to execute a received command according to program code stored in a recording device such as the storage unit 250.

The communication unit 220 may provide a function for communicating with an external device through a network. For example, the request generated by the control unit 210 of the vascular age diagnosis apparatus 200 according to the program code stored in the recording device such as the storage unit 250 takes an external image through a network under the control of the communication unit 220 It may be delivered to the device 100, the database 300, or another user terminal. For example, a control signal or command received through the communication unit 220 may be transmitted to the control unit 210 or the storage unit 250, and the received image image may be stored by the storage unit 250.

The storage unit 250 is a computer-readable recording medium, and may include a non-destructive mass storage device such as random access memory (RAM), read only memory (ROM), and a disk drive. In addition, an operating system and at least one program code may be stored in the storage unit 250. These software components may be loaded from a computer-readable recording medium separate from the storage unit 250 using a drive mechanism. Such a separate computer-readable recording medium may include a computer-readable recording medium such as a floppy drive, disk, tape, DVD / CD-ROM drive, and memory card. In other embodiments, software components may be loaded into storage 250 via communication unit 220 rather than a computer-readable recording medium. For example, at least one program is a storage unit 250 based on a program (for example, the application described above) installed by files provided by a file distribution system that distributes installation files of developers or applications through a network. Can be loaded on.

The input / output unit 240 may receive input such as setting a segmentation section from a user, and display a screen showing a blood vessel image and a 3D binary blood vessel image. For example, the input / output unit 240 may include an operation panel for receiving user input, a display panel for displaying the screen, and the like.

Specifically, the input unit may include devices capable of receiving various types of user input, such as a keyboard, a physical button, a touch screen, a camera, or a microphone. Also, the output unit may include a display panel or a speaker. However, the present invention is not limited thereto, and the input / output unit 240 may include a configuration supporting various input / output.

The blood vessel age diagnosis apparatus 200 analyzes and processes an image of an object to diagnose blood vessel health, an image receiving unit 251, a blood vessel extracting unit 252, a center line acquiring unit 253, and a distortion calculating unit 254, The blood vessel age inference unit 255 and the evaluation unit 256 may be included.

The image receiving unit 251 may receive one or more images of a part of the body of the object. The image receiving unit 251 receives an image of the object. Here, the image may be a Time of Flight (TOF) magnetic resonance angiography image, but is not limited thereto and may be an image or image in various formats.

The blood vessel extraction unit 252 may extract a 3D binary blood vessel image from the image. In the 3D binary blood vessel image, a voxel corresponding to a blood vessel may be set to 1, and a voxel not corresponding to a blood vessel may be set to 0. The blood vessel extracting unit 252 may extract a plurality of blood vessel segmentation regions for each anatomically separated region from the 3D binary blood vessel image. The blood vessel extraction unit 252 interpolates the x, y, and z components of each voxel of the image, corrects the z component of each voxel, and corrects the x, y, and z components of each voxel of the image using the corrected result. Then, by changing the position of each voxel with the corrected x, y, and z components, a 3D binary blood vessel image can be generated. For example, when the z component is rather large compared to the x and y components of any voxel, readability may decrease as the z component is expressed as long. By correcting the z component based on the x component and the y component using interpolation, it is possible to prevent the blood vessel shape from being elongated in any one axial direction.

More specifically, the blood vessel extraction unit 252 may skeletonize a 3D binary blood vessel image. The blood vessel extracting unit 252 may acquire blood vessel segmentation regions by skeletonizing blood vessel stems included in the 3D binary blood vessel image. Skeletonization may be performed by methods such as Hiliditch algorithm, Pavlidis algorithm, and Rosenfeld algorithm known to those skilled in the art.

The center line acquiring unit 253 may extract a set of center points of the first blood vessel section designated by the user using a 3D binary blood vessel image and blood vessel segmentation regions and set the center line. The center line acquiring unit 253 may extract a center line, which is a calculation target of the distortion value, by tracking the center points connecting the start and end points set by user inputs on the 3D binary blood vessel image.

The center line acquiring unit 253 may skeletonize the 3D binary blood vessel image and the blood vessel segmentation regions, and extract a set of center points of the first blood vessel section designated by the user and set the center line. The center line acquisition unit 253 may extract a set of center points by tracking the points connecting the start and end points specified by the user.

The warpage calculation unit 254 calculates the warpage value of the blood vessel based on the centerline obtained by the centerline acquisition unit 253. The distortion calculation unit 254 calculates the distortion value of the blood vessel based on the center line included in each of the blood vessel division regions. The warp calculation unit 254 calculates a warp value set from the image of the object. The distortion value can be calculated using the length of the center line and the shortest length of the center line. For example, the distortion value may be set to a value proportional to the length of the center line / the shortest length of the center line.

The vascular age inference unit 255 infers the vascular age of the object using a set of distortion values for each region. The vascular age inference unit 255 may infer the vascular age of the object using a vascular age diagnostic model that is trained as a distortion value of vascular segmentation regions for each region included in the image.

The blood vessel age inference unit 255 may infer the blood vessel age of the subject using a warping value corresponding to a preset blood vessel age. The vascular age inference unit 255 infers the vascular age of the subject against the set of warping values of the subject by inputting and learning the vascular age diagnosed by the diagnostician or the biological vascular age of the actual subject along with the set of warping values for each region. You may.

At this time, the deduced object blood vessel age may be output as a set of a plurality of values in the form of a probability map. For example, the vascular age can be inferred from (20%, 28 years old), (30%, 35 years old), (50%, 32 years old) from the measured subject's image. The vascular age inference unit 255 may determine the most likely 32 years old as the vascular age and finally output it. Alternatively, the vascular age inference unit 255 may determine the vascular age of the subject by weighting the probability by weighting the probability.

The evaluation unit 256 may compare the vascular age of the subject deduced through the vascular age diagnosis apparatus with the actual subject's age to evaluate the reasoning result of the vascular age. The evaluation unit 256 may change the vascular age diagnosis model or re-learn the vascular age diagnosis model when the reasoning result does not reach a preset evaluation criterion.

The vascular age diagnosis apparatus 200 may further include a vascular condition diagnosis unit (not shown) that diagnoses vascular health based on vascular age. The inferred or measured vascular age may be analyzed in consideration of the biometric information such as the age and gender of the subject. That is, the vascular state according to the vascular age of the subject may be classified as either normal or abnormal in consideration of the biological information such as the age and gender of the subject. In particular, the vascular condition diagnosis unit diagnoses the vascular condition of the object by communicating with a database storing an average value (distortion value, etc.) of the vessel according to the gender and age of the object based on a plurality of vascular images collected for inferring vascular age. Can be. The diagnosed vascular condition may include the risk of developing heart disease or the risk of developing cerebrovascular disease, the risk of developing cardiovascular disease, and the likelihood of stroke. The diagnosed vascular condition can be predicted based on the distortion value of each point of the vessel.

Through this, the vascular age diagnosis apparatus 200 of the embodiments of the present invention can diagnose the vascular age of the subject using an image. In particular, the blood vessel age diagnosis apparatus 200 may calculate warpage values for a plurality of blood vessel segmentation regions included in the blood vessel of the subject, and diagnose the blood vessel age by considering the warpage values of each blood vessel region.

The blood vessel age diagnosis apparatus 200 according to an embodiment of the present invention may be connected to the learning model apparatus 300 to diagnose blood vessel age. The vascular age diagnosis apparatus 200 may learn criteria for vascular age diagnosis by inputting distortion values calculated from images and images into the learning model apparatus 300.

Specifically, the learning model device 300 may include a data learning unit 310 and a data recognition unit 320. The operation of each component is as follows.

The data learning unit 310 may learn criteria for diagnosing vascular age. The data learning unit 310 may learn criteria to use what data to determine a predetermined blood vessel age and how to determine the blood vessel age using the data. The data learning unit 310 may learn criteria for vascular age diagnosis by acquiring an image, a warping value of a blood vessel, etc., which is data to be used for learning, and applying the obtained data to a vascular age diagnosis model to be described later.

The data recognition unit 320 may determine the blood vessel age of the object based on the input data. The data recognition unit 320 may diagnose and recognize a subject's vascular age from predetermined data using the learned vascular age diagnostic model. The data recognition unit 320 obtains predetermined data according to a preset criterion by learning, and uses the acquired data as an input value to use the vascular age diagnostic model to determine the vascular age of the object based on the predetermined data Can be. In addition, the result value output by the vascular age diagnostic model using the obtained data as an input value may be used to update the vascular age diagnostic model.

At least one of the data learning unit 310 and the data recognition unit 320 may be manufactured in the form of at least one hardware chip and mounted on the electronic device. For example, at least one of the data learning unit 310 and the data recognition unit 320 may be manufactured in the form of a dedicated hardware chip for artificial intelligence (AI), or an existing general-purpose processor (for example, a CPU) Alternatively, it may be manufactured as a part of an application processor or a graphics-only processor (for example, a GPU) and mounted on various electronic devices described above.

In this case, the data learning unit 310 and the data recognition unit 320 may be mounted on one electronic device, or may be mounted on separate electronic devices. For example, one of the data learning unit 310 and the data recognition unit 320 may be included in the electronic device, and the other one may be included in the server. In addition, the data learning unit 310 and the data recognition unit 320 may provide the model information constructed by the data learning unit 310 to the data recognition unit 320 through wired or wireless communication. 320) may be provided to the data learning unit 310 as additional learning data.

Meanwhile, at least one of the data learning unit 310 and the data recognition unit 320 may be implemented as a software module. When at least one of the data learning unit 310 and the data recognition unit 320 is implemented as a software module (or a program module including an instruction), the software module is a computer-readable, non-transitory readable It may be stored in a readable media (non-transitory computer readable media). In addition, in this case, at least one software module may be provided by an operating system (OS) or may be provided by a predetermined application. Alternatively, some of the at least one software module may be provided by an operating system (OS), and the other may be provided by a predetermined application.

As illustrated in FIG. 4, the data learning unit 310 includes a data acquisition unit 311, a pre-processing unit 312, a training data selection unit 313, a model learning unit 314, and a model evaluation unit 315. It can contain.

The data acquisition unit 311 may acquire data necessary for determining blood vessel age. The data acquisition unit 311 may acquire data necessary for learning to determine blood vessel age. The data acquisition unit 311 may acquire images, images, and text related to blood vessels from the blood vessel age diagnosis apparatus 200 or the image taking apparatus 100. Specifically, the data acquisition unit 311 may acquire distortion values of parts of an image or a blood vessel. The data acquisition unit 311 is not limited to the above description, and an external electronic device

The pre-processing unit 312 may pre-process the acquired data so that the acquired data can be used for learning to determine blood vessel age. The pre-processing unit 312 may process the acquired data in a preset format so that the model learning unit 314, which will be described later, can use the acquired data for learning to determine blood vessel age. The pre-processing unit 312 may extract a blood vessel region corresponding to blood vessels included in the image. To visualize the vascular region in three dimensions, the x, y, and z-axis components of each voxel can be corrected in the vascular region. At this time, x, y, and z-axis components may be corrected using an interpolation method. Through this correction process, the visibility of the vascular region may be increased. In addition, the pre-processing unit 312 may extract a centerline connecting the center points of the blood vessels by skeletonizing an image corresponding to the blood vessel region rendered in 3D. Also, the pre-processing unit 312 may calculate distortion values of center lines extracted from the image.

The learning data selector 313 may select data necessary for learning from pre-processed data. The selected data may be provided to the model learning unit 314. The learning data selector 313 may select data necessary for learning from pre-processed data according to preset criteria for determining blood vessel health such as vascular age. In addition, the learning data selection unit 313 may select data according to criteria set by learning by the model learning unit 314 to be described later. The learning data selection unit 313 may select all or part of a distortion value of centerlines of a plurality of branches included in the image.

The model learning unit 314 may learn criteria on how to determine vascular health such as vascular age based on the training data. In addition, the model learning unit 314 may also learn the criteria for which vascular stem distortion data, etc., should be used to determine vascular health such as vascular age.

In addition, the model learning unit 314 may train a vascular age diagnosis model used for vascular health determination such as vascular age using learning data. In this case, the vascular age diagnosis model may be a pre-built model. For example, the vascular age diagnosis model may be a pre-built model receiving basic learning data (eg, DWI, ADC, TOF images, etc.).

The vascular age diagnosis model may be constructed in consideration of the application field of the recognition model, the purpose of learning, or computer performance of the device. The vascular age diagnosis model may be, for example, a model based on a neural network. For example, a model such as a deep neural network (DNN), a recurrent neural network (RNN), or a bidirectional recurrent deep neural network (BRDNN) may be used as a vascular age diagnosis model, but is not limited thereto.

According to various embodiments of the present disclosure, when there are a plurality of pre-built vascular age diagnostic models, the model learning unit 314 may learn the vascular age diagnostic model for learning the vascular age diagnostic model having a high relationship between the input learning data and the basic learning data. You can decide. In this case, the basic learning data may be pre-classified for each type of data, and the vascular age diagnosis model may be pre-built for each type of data. For example, the basic training data is classified into various criteria such as the region where the training data is generated, the time when the training data is generated, the size of the training data, the genre of the training data, the creator of the training data, and the type of object in the training data. It may be.

In addition, the model learning unit 314 may train a vascular age diagnosis model using, for example, a learning algorithm including an error back-propagation or a gradient descent method.

In addition, the model learning unit 314 may train the vascular age diagnosis model, for example, through supervised learning using learning data as an input value. In addition, the model learning unit 314, for example, by learning the type of data necessary for vascular health determination, such as vascular age, without any guidance, unsupervised to discover criteria for vascular health determination, such as vascular age. Through unsupervised learning, a vascular age diagnostic model can be trained. In addition, the model learning unit 314 may train a vascular age diagnosis model through reinforcement learning using feedback on whether a result of vascular health determination such as vascular age according to learning is correct, for example. have.

In addition, when the vascular age diagnosis model is learned, the model learning unit 314 may store the learned vascular age diagnosis model. In this case, the model learning unit 314 may store the trained vascular age diagnosis model in the memory of the electronic device including the data recognition unit 320. Alternatively, the model learning unit 314 may store the trained vascular age diagnosis model in the memory of the electronic device including the data recognition unit 320 to be described later. Alternatively, the model learning unit 314 may store the trained vascular age diagnosis model in a memory of a server connected to a wired or wireless network with the electronic device.

In this case, the memory in which the learned vascular age diagnosis model is stored may store, for example, commands or data related to at least one other component of the electronic device. Also, the memory may store software and / or programs. The program may include, for example, a kernel, middleware, application programming interface (API), and / or application program (or "application").

The model evaluation unit 315 may input the evaluation data into the vascular age diagnosis model and, if the recognition result output from the evaluation data does not satisfy a predetermined criterion, may cause the model learning unit 314 to learn again. In this case, the evaluation data may be preset data for evaluating the vascular age diagnosis model.

For example, the model evaluation unit 315, when the recognition result of the vascular age, etc. among the recognition results of the learned vascular age diagnosis model for the evaluation data, the number or ratio of evaluation data that is not accurate exceeds a preset threshold. It can be evaluated that the predetermined criteria are not satisfied.

On the other hand, when there are a plurality of trained vascular age diagnostic models, the model evaluator 315 evaluates whether or not a predetermined criterion is satisfied for each trained video recognition model, and determines a final vascular age for a model that satisfies a predetermined criterion. You can decide as a model. In this case, when there are a plurality of models satisfying a predetermined criterion, the model evaluator 315 may determine, as a final vascular age diagnosis model, any one or a predetermined number of models preset in order of highest evaluation score.

Meanwhile, at least one of the data acquisition unit 311, the pre-processing unit 312, the training data selection unit 313, the model learning unit 314, and the model evaluation unit 315 in the data learning unit 310 is at least one. It can be manufactured in the form of a hardware chip and mounted on an electronic device. For example, at least one of the data acquisition unit 311, the pre-processing unit 312, the training data selection unit 313, the model learning unit 314, and the model evaluation unit 315 is artificial intelligence (AI). It may be manufactured in the form of a dedicated hardware chip for, or may be manufactured as part of an existing general-purpose processor (for example, a CPU or application processor) or a graphics-only processor (for example, a GPU) and mounted on various electronic devices described above.

In addition, the data acquisition unit 311, the pre-processing unit 312, the training data selection unit 313, the model learning unit 314 and the model evaluation unit 315 may be mounted on one electronic device, or separately Each may be mounted on electronic devices. In addition, at least one of the data acquisition unit 311, the pre-processing unit 312, the training data selection unit 313, the model learning unit 314, and the model evaluation unit 315 may be implemented as a software module. At least one of the data acquisition unit 311, the pre-processing unit 312, the training data selection unit 313, the model learning unit 314, and the model evaluation unit 315 includes a software module (or an instruction). Program module), the software module may be stored in a computer-readable readable non-transitory computer readable media.

Next, the structure of the data recognition unit 320 will be described.

The data recognition unit 320 according to an embodiment of the present invention includes a data acquisition unit 321, a pre-processing unit 322, a recognition data selection unit 323, a recognition result providing unit 324, and a model update unit 325. It can contain.

The data acquisition unit 321 may acquire data necessary for determining vascular health such as vascular age, and the pre-processing unit 322 may acquire data so that data obtained for vascular health determination such as vascular age may be used Can be pretreated. The pre-processing unit 322 may process the acquired data in a preset format so that the recognition result providing unit 324, which will be described later, can use the acquired data to determine vascular health such as vascular age. The pre-processing unit 322 may extract a blood vessel region corresponding to blood vessels included in the image. To visualize the vascular region in three dimensions, the x, y, and z-axis components of each voxel can be corrected in the vascular region. At this time, x, y, and z-axis components may be corrected using an interpolation method. Through this correction process, the visibility of the vascular region may be increased. In addition, the pre-processing unit 322 may extract the center line connecting the center points of the blood vessels by skeletonizing an image corresponding to the blood vessel region rendered in 3D. Also, the pre-processing unit 322 may calculate distortion values of center lines extracted from the image. Before the pre-processing unit 322 calculates the warpage value, the center line connecting the center points of the blood vessel may be tracked.

The recognition data selection unit 323 may select data necessary for determining vascular health, such as vascular age, among pre-processed data. The selected data may be provided to the recognition result providing unit 324. The recognition data selector 323 may select some or all of the pre-processed data according to preset criteria for determining blood vessel health such as blood vessel age. In addition, the recognition data selection unit 323 may select data according to a preset criterion by learning by the model learning unit 314.

The recognition result providing unit 324 may determine vascular health such as vascular age by applying the selected data to the vascular age diagnosis model. The recognition result providing unit 324 may provide a recognition result according to the purpose of recognizing data. The recognition result providing unit 324 may apply the selected data to the vascular age diagnosis model by using the data selected by the recognition data selection unit 323 as an input value. In addition, the recognition result is determined by the vascular age diagnostic model, and may include vascular health conditions such as the vascular age of the subject.

The model update unit 325 may allow the vascular age diagnosis model to be updated based on the evaluation of the recognition result provided by the recognition result providing unit 324. For example, the model update unit 325 may provide the model learning unit 324 with the recognition result provided by the recognition result providing unit 324, so that the model learning unit 324 updates the vascular age diagnosis model. Can be.

On the other hand, at least one of the data acquisition unit 321, the pre-processing unit 322, the recognition data selection unit 323, the recognition result providing unit 324 and the model update unit 325 in the data recognition unit 320, at least It can be manufactured in the form of one hardware chip and mounted on an electronic device. For example, at least one of the data acquisition unit 321, the pre-processing unit 322, the recognition data selection unit 323, the recognition result providing unit 324, and the model update unit 325 is artificial intelligence (AI). It may be manufactured in the form of a dedicated hardware chip for), or it may be manufactured as a part of an existing general-purpose processor (for example, a CPU or application processor) or a graphics-only processor (for example, a GPU) and mounted on various electronic devices described above.

4 and 5 are flowcharts of a method for diagnosing vascular health according to embodiments of the present invention.

As shown in FIG. 4, the method for diagnosing vascular health receives an image captured from the image capturing apparatus 100.

In S120, the blood vessel age diagnosis apparatus 200 may extract a 3D binary blood vessel image from the image. In the 3D binary blood vessel image, a voxel corresponding to a blood vessel may be set to 1, and a voxel not corresponding to a blood vessel may be set to 0. The vascular age diagnosis apparatus 200 may extract a plurality of vascular segmentation regions for each anatomically separated region from a 3D binary vascular image.

In S130, the vascular age diagnosis apparatus 200 may extract a set of center points of a first blood vessel section designated by a user using a 3D binary blood vessel image and blood vessel segmentation regions and set the center line.

)을 찾고, 끝점과 가장 가까운 중심선 상의 점(

)을 찾는다.

에서 출발하여

으로 향하는 중심선 상의 점들을 자동으로 순차적으로 추적한다. 점과 점을 잇는 벡터의 방향 코사인 (directional cosine)을 근거로 하여 자동적으로 중심점들을 추적할 수 있다. 이 점들을 따라 거리를 합하면 혈관 분할 영역에 대한 혈관 길이(branch length)로 근사화 하여 다음과 같이 산출할 수 있다. In S140, the vascular age diagnosis apparatus 200 may calculate the distortion from the center line of the vascular segmentation region (eg, the right middle cerebral artery region) numerically. The starting and ending points of the corresponding blood vessel may be given for the blood vessel segmentation region. At this time, the start point and the end point of the blood vessel segmentation region may be automatically found, or the user may manually view and view the MRI image using the interface. Given a starting point and an ending point, the point on the centerline closest to the starting point (

), And the point on the centerline closest to the endpoint (

).

From

The points on the centerline heading to are automatically tracked sequentially. The center points can be automatically tracked based on the directional cosine of the vector connecting the points. Summing the distances along these points can be approximated to the vessel length for the blood vessel segmentation region and calculated as follows.

는 3차원 상의 점

와 점

간의 직선 거리 (즉, Euclidean distance)이다. 해당 혈관 영역에서 혈관의 뒤틀림(tortuosity) 값을 다음과 같이 산출한다. At this time,

Is a 3D point

And dot

It is the straight line distance (ie Euclidean distance). The tortoosity value of the blood vessel in the corresponding blood vessel region is calculated as follows.

The blood vessel age diagnosis apparatus 200 calculates the warpage values of blood vessels based on the center line included in each of all blood vessel division regions on this principle. In general, it can be seen that there is a positive correlation between blood vessel warpage value and blood vessel age. The vascular age diagnosis apparatus 200 calculates a set of distortion values for each vascular segment from an image of the object.

In S150, the vascular age diagnosis apparatus 200 infers a subject's vascular age using a set of distortion values for each region. The blood vessel age diagnosis apparatus 200 may infer a blood vessel age of a subject using a warping value corresponding to a preset blood vessel age. The blood vessel age diagnosis apparatus 200 inputs and learns a blood vessel age diagnosed by a diagnostic doctor or a biological blood vessel age of an actual subject along with a set of warpage values for each region, and the blood vessel age of the subject with respect to the set of warpage values of the subject You can also infer.

As shown in FIG. 5, the learning model device 300 may train a vascular age diagnosis model in conjunction with a vascular age diagnosis device.

In S210, the learning model device 300 receives a captured image of the object from the image capturing device or the vascular age diagnosis device.

In S220, the learning model device 300 extracts data for vascular age determination from the captured image. The learning model device 300 may extract data such as a 3D binary blood vessel image, a blood vessel split image divided into a blood vessel stem, and a warping value for blood vessel stems as data for determining blood vessel age.

In S230, the learning model device 300 selects data for vascular age determination from the extracted data. In S240, the learning model device 300 trains a blood vessel age diagnosis model using data for determining blood vessel age.

In S250, the learning model device 300 evaluates the vascular age diagnosis model by inputting data such as vascular age that is preset in the learned vascular age diagnosis model.

In S260, the learning model device 300 updates the vascular age diagnosis model that satisfies a predetermined evaluation criterion by transferring it to the storage unit of the vascular age diagnosis device.

Through this, the vascular age diagnosis apparatus 200 according to embodiments of the present invention may receive a vascular age diagnosis model through the learning model device 300 and infer the vascular age of the subject using the vascular age diagnosis model. .

7 to 9 are exemplary views of a user interface provided through a vascular age diagnosis device. The blood vessel age diagnosis apparatus 200 provides an image S1 of an object photographed through the input / output unit 240 and a 3D binary blood vessel image S2 extracted from the image.

The start point is set by I1 for the 3D binary blood vessel image S2, and the end point is set by I2. The vascular age diagnosis apparatus 200 may extract centerlines of blood vessels in consideration of I1 and I2.

In FIG. 8, a portion marked with ellipse indicates a left middle cerebral artery vascular region, and is an exemplary view showing overlaying the center points tracked through tracking on a blood vessel when a start point and an end point are given.

The vascular age diagnosis apparatus 200 may provide buttons (item1, item2, item3, item4) for calculating the warpage value of each vascular stem by user input.

10 is an exemplary diagram for inferring vascular age through a model trained from a warpage value of a blood vessel. Here, the distortion values of blood vessels are basilar artery, Posterior cerebral artery, Vertebral artery, and the like, but are not limited thereto and can be calculated by various algorithms. For example, after training models such as a linear regression model, a support vector machine (SVM) regression model, and a random forests regression model, the trained model can be used to predict vascular age for new blood vessels. In supervised learning, the input section is a set of blood vessel warpage and the output section is the blood vessel age. At this time, the input / output data for supervised learning can be obtained through the data of the normal group subject. When a missing value occurs in input data during training or testing, it can be replaced with another value using methods such as multiple imputation and inference using statistical information of samples without missing. .

The device described above may be implemented with hardware components, software components, and / or combinations of hardware components and software components. For example, the devices and components described in the embodiments include, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors (micro signal processors), microcomputers, field programmable gate arrays (FPGAs). , A programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions, may be implemented using one or more general purpose computers or special purpose computers. The processing device may perform an operating system (OS) and one or more software applications running on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to execution of the software. For convenience of understanding, a processing device may be described as one being used, but a person having ordinary skill in the art, the processing device may include a plurality of processing elements and / or a plurality of types of processing elements. It can be seen that may include. For example, the processing device may include a plurality of processors or a processor and a controller. In addition, other processing configurations, such as parallel processors, are possible.

The software may include a computer program, code, instruction, or a combination of one or more of these, and configure the processing device to operate as desired, or process independently or collectively You can command the device. Software and / or data may be interpreted by a processing device, or to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. , Or may be permanently or temporarily embodied in the transmitted signal wave. The software may be distributed over networked computer systems, and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded in the medium may be specially designed and configured for the embodiments or may be known and usable by those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, and magnetic media such as floptical disks. -Hardware devices specifically configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, etc., as well as machine language codes produced by a compiler. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described by a limited embodiment and drawings, those skilled in the art can make various modifications and variations from the above description. For example, the described techniques are performed in a different order than the described method, and / or the components of the described system, structure, device, circuit, etc. are combined or combined in a different form from the described method, or other components Alternatively, even if replaced or substituted by equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

10: Vascular age diagnosis system
100: video recording device
200: vascular age diagnostic device
300: learning model device

Claims (13)

Receiving an image of an object captured by the vascular age diagnosis apparatus from the imaging apparatus;
Extracting a 3D binary blood vessel image from the image by the vessel age diagnosis apparatus;
Extracting vascular segmentation regions for each anatomically separated region from the 3D binary vessel image by the vascular age diagnosis apparatus;
Extracting a center line connecting the center points of the first blood vessel section by using the blood vessel segmentation regions;
Calculating, by the vascular age diagnosis apparatus, a warping value of a blood vessel based on the center line; And
The blood vessel age diagnostic apparatus inferring the blood vessel age of the subject based on the distortion value. According to claim 1,
The step of extracting the 3D binary blood vessel image
By interpolating the x, y, and z components of each voxel of the image, the z component of each voxel is corrected, and the x, y, and z components of each voxel of the image are corrected using the corrected result, A method of diagnosing vascular age, characterized by extracting a 3D binary vascular image by changing the position of each voxel with the corrected x, y, and z components.
According to claim 1,
Inferring the age of the blood vessels
A method of diagnosing vascular age, characterized in that the vascular age of the subject is inferred using a vascular age diagnostic model trained as a distortion value of vascular segmentation regions for each region included in the image. According to claim 1,
The step of extracting the centerline
A method of diagnosing vascular age, characterized in that the distortion value of a blood vessel is quantified by tracking center points connecting a start point and an end point set by user inputs on the 3D binary blood vessel image. According to claim 1,
After the reasoning,
And comparing the vascular age of the subject with the actual subject's age to evaluate the inference result of the vascular age. According to claim 1,
And overlapping the image and the center line extracted on the 3D binary blood vessel image and outputting it through an input / output unit. An image receiving unit configured to receive an image of an object from an image capturing apparatus;
A blood vessel extraction unit extracting a 3D binary blood vessel image from the image and extracting blood vessel segmentation regions for each anatomically separated region from the 3D binary blood vessel image;
A center line acquiring unit extracting a center line connecting the center points of the first blood vessel section using the blood vessel segmentation regions;
A distortion calculation unit calculating a distortion value of a blood vessel based on the center line; And
A vascular age diagnosis apparatus including a vascular age inference unit that infers a blood vessel age of the object based on the warpage value. The method of claim 7,
The blood vessel extraction unit
By interpolating the x, y, and z components of each voxel of the image, the z component of each voxel is corrected, and the x, y, and z components of each voxel of the image are corrected using the corrected result, A vascular age diagnosis apparatus characterized by extracting a 3D binary blood vessel image by changing the position of each voxel with the corrected x, y, and z components.
The method of claim 7,
The vascular age deduction unit
A blood vessel age diagnosis apparatus characterized by inferring a blood vessel age of the object using a blood vessel age diagnostic model that is trained as a warp value of blood vessel segmentation regions for each region included in the image. The method of claim 7,
The center line acquisition unit
And tracking a center point connecting a start point and an end point set by user inputs on the 3D binary blood vessel image, thereby extracting a center line that is a target for calculating a warpage value. The method of claim 7,
And further comprising an evaluation unit for evaluating the inference result of the vascular age by comparing the vascular age of the subject and the actual subject's age, vascular age diagnosis apparatus. The method of claim 7,
And an input / output unit that overlaps and outputs the center line extracted on the image and the 3D binary blood vessel image.
A computer program stored in a computer readable storage medium for executing the method of claim 1 using a computer.

Images