CN114340496A

CN114340496A - Analysis method and related device of heart coronary artery based on VRDS AI medical image

Info

Publication number: CN114340496A
Application number: CN201980099742.5A
Authority: CN
Inventors: 戴维伟·李; 斯图尔特平·李
Original assignee: Weiai Medical Technology Shenzhen Co ltd
Current assignee: Cao Sheng
Priority date: 2019-10-29
Filing date: 2019-10-29
Publication date: 2022-04-12
Also published as: WO2021081771A1

Abstract

A method and a related device for analyzing cardiac coronary artery based on VRDS AI medical images comprise the following steps: acquiring a scan image of a coronary artery part of a heart, wherein the scan image further comprises the coronary artery part of the heart and blood vessels (201) around the coronary artery of the heart; performing image processing according to the scanned image to obtain a target image data set (202); determining the lesion degree of the coronary artery of the heart of the target user according to the target image data set (203); the type of coronary stent is determined according to the lesion degree of the coronary artery of the heart (204). Can improve the accuracy and intelligence of selecting coronary artery stent in the coronary artery disease of heart.

Description

Analysis method and related device of heart coronary artery based on VRDS AI medical image

Technical Field

The present application relates to the field of medical imaging device technology, and in particular, to a method and related device for analyzing coronary artery based on VRDS AI medical images.

Background

Currently, doctors acquire information such as the morphology, position, and topology of coronary vessels of the heart by using techniques such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Diffusion Tensor Imaging (DTI), and Positron Emission Tomography (PET). Doctors still diagnose the disease condition by watching and reading the continuous two-dimensional slice scanning images. However, the two-dimensional slice scan image cannot present the spatial structure characteristics of the coronary artery of the heart, which affects the diagnosis of diseases by doctors. With the rapid development of medical imaging technology, people put new demands on medical imaging.

Disclosure of Invention

The embodiment of the application provides a method and a related device for analyzing coronary artery based on VRDS AI medical images, and the method and the device can be implemented to improve the accuracy and intelligence for selecting coronary artery stents in coronary artery diseases.

In a first aspect, an embodiment of the present application provides a method for analyzing a coronary artery of a heart based on VRDS AI medical images, including:

acquiring a scan image of a coronary artery region of a heart, wherein the scan image further comprises the coronary artery region of the heart and blood vessels around the coronary artery of the heart;

performing image processing according to the scanned image to obtain a target image data set;

determining the lesion degree of the coronary artery of the target user according to the target image data set;

and determining the type of the coronary stent according to the lesion degree of the coronary artery of the heart.

A second aspect of embodiments of the present application provides a medical imaging apparatus, including:

an acquisition unit for acquiring a scan image of a coronary artery region of a heart, wherein the scan image further comprises the coronary artery region of the heart and blood vessels around the coronary artery of the heart;

the processing unit is used for carrying out image processing according to the scanning image to obtain a target image data set;

the determining unit is used for determining the lesion degree of the coronary artery of the heart of the target user according to the target image data set;

the determining unit is further used for determining the type of the coronary stent according to the lesion degree of the coronary artery of the heart.

A third aspect of embodiments of the present application provides a medical imaging apparatus comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and are generated as instructions to be executed by the processor to perform the steps of the method of any of the first aspects of the claims above.

A fourth aspect of embodiments of the present application provides a computer-readable storage medium for storing a computer program, the stored computer program being executable by the processor to implement the method of any one of the above first aspects.

It can be seen that, in the above technical solution, a scanned image of a coronary artery region of a heart is obtained, where the scanned image further includes the coronary artery region of the heart and blood vessels around the coronary artery of the heart, then, image processing is performed according to the scanned image to obtain a target image data set, then, a lesion degree of the coronary artery of the target user is determined according to the target image data set, and finally, a type of a coronary stent is determined according to the lesion degree of the coronary artery of the heart. The coronary stent is quantitatively selected, so that the coronary stent can cover plaques after being expanded to support the vascular stent, the plaques cannot fall off, the surgical risk of the vascular stent implantation is reduced, and the accuracy and the intelligence for selecting the coronary stent in the coronary heart disease are improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic structural diagram of a system for analyzing coronary artery based on VRDS AI medical images according to an embodiment of the present disclosure;

fig. 2 is a schematic flowchart of a method for analyzing coronary artery based on VRDS AI medical images according to an embodiment of the present disclosure;

fig. 3 is a schematic flowchart of another method for analyzing coronary artery based on VRDS AI medical image according to an embodiment of the present disclosure;

fig. 4 is a schematic flowchart of another method for analyzing coronary artery based on VRDS AI medical image according to an embodiment of the present application;

FIG. 5 is a schematic view of a medical imaging apparatus provided by an embodiment of the present application;

fig. 6 is a schematic structural diagram of a medical imaging apparatus in a hardware operating environment according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The following are detailed below.

The terms "first," "second," and "third" in the description and claims of this application and in the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions.

The medical imaging apparatus according to the embodiments of the present application refers to various apparatuses that reproduce the internal structure of a human body as an image using various media as information carriers, and the image information corresponds to the actual structure of the human body in terms of spatial and temporal distribution. The "DICOM data" refers to original image file data which reflects internal structural features of a human body and is acquired by medical equipment, and may include information such as computed tomography CT, magnetic resonance MRI, diffusion tensor imaging DTI, positron emission tomography PET-CT, and the "map source" refers to Texture2D/3D image volume data generated by analyzing the original DICOM data. "VRDS" refers to a Virtual Reality medical system (VRDS).

First, referring to fig. 1, fig. 1 is a schematic structural diagram of a system 100 for analyzing a coronary artery based on a VRDS AI medical image according to an embodiment of the present application, where the system 100 includes a medical imaging apparatus 110 and a network database 120, where the medical imaging apparatus 110 may include a local medical imaging apparatus 111 and/or a terminal medical imaging apparatus 112, and the local medical imaging apparatus 111 or the terminal medical imaging apparatus 112 is configured to perform identification, positioning, four-dimensional volume rendering, and anomaly analysis of a coronary artery image region of a human heart based on original DICOM data based on an analysis algorithm of a coronary artery based on a VRDS AI medical image presented in an embodiment of the present application, so as to achieve a four-dimensional stereoscopic imaging effect (the 4-dimensional medical image specifically refers to a medical image including an internal spatial structural feature and an external spatial structural feature of a displayed tissue, the internal spatial structural feature refers to that slice data inside the tissue is not lost, that is, the medical imaging apparatus may present the internal structure of the tissues such as the coronary artery and the blood vessel, and the external spatial structural characteristic refers to the environmental characteristics between the tissues, including the spatial position characteristics (including intersection, spacing, fusion) between the tissues, such as the edge structural characteristics of the intersection position between the left coronary artery and the right coronary artery, and the like), the local medical imaging apparatus 111 may also be used to edit the scanned image with respect to the terminal medical imaging apparatus 112, to form the transfer function result of the four-dimensional human body image, which may include the transfer function result of the surface of the coronary artery of the human body and the tissue structure in the coronary artery of the human body, and the transfer function result of the cubic space, such as the number of the arrays, coordinates, colors, transparencies of the cubic edit box and arc line edit required by the transfer function. The network database 120 may be, for example, a cloud medical imaging apparatus, and the like, and the network database 120 is used to store the image sources generated by parsing the raw DICOM data and the transfer function results of the four-dimensional human body images edited by the local medical imaging apparatus 111, and the scanned images may be from a plurality of local medical imaging apparatuses 111 to realize interactive diagnosis of a plurality of doctors.

When the user performs specific image display by using the medical imaging apparatus 110, the user may select a display or a Head Mounted Display (HMDS) of the virtual reality VR to display in combination with an operation action, where the operation action refers to operation control performed on a four-dimensional human body image by the user through an external shooting device of the medical imaging apparatus, such as a mouse, a keyboard, a tablet computer (Pad), an ipad (internet portable device), and the like, so as to implement human-computer interaction, and the operation action includes at least one of the following: (1) changing the color and/or transparency of a specific organ/tissue, (2) positioning a zoom view, (3) rotating the view to realize multi-view 360-degree observation of a four-dimensional human body image, (4) entering 'internal observation of a human body organ, real-time shearing effect rendering', and (5) moving the view up and down.

Referring to fig. 2, fig. 2 is a flowchart illustrating a method for analyzing a coronary artery of a heart based on VRDS AI medical images according to an embodiment of the present application. As shown in fig. 2, a method for analyzing coronary artery based on VRDS AI medical image according to an embodiment of the present application may include:

201. a medical imaging apparatus acquires a scan image of a coronary heart site, wherein the scan image further includes the coronary heart site and blood vessels surrounding the coronary heart.

Wherein the scanned image comprises any one of: CT images, MRI images, DTI images, PET-CT images.

Wherein blood enters the heart via two main coronary arteries and passes through a network of blood vessels on the surface of the heart muscle, and the coronary arteries of the heart are divided into 22 branches, each branch corresponding to a respective blood vessel.

202. And the medical imaging device performs image processing according to the scanning image to obtain a target image data set.

Wherein the image data comprises three-dimensional spatial image data of the coronary vessels of the heart.

Optionally, in a possible implementation, the performing image processing according to the scanned image to obtain a target image data set includes: carrying out image preprocessing according to the scanned image to obtain first image data; generating an original image data set of the coronary artery part of the heart according to the first image data; and carrying out boundary optimization processing on the original image data set to obtain a target image data set.

Among them, the coronary vessels of the heart include coronary arteries and coronary veins. Further, the coronary arteries may include, for example: left coronary artery, right coronary artery, etc. Coronary veins may include, for example, the great cardiac vein, the central vein, the small cardiac vein, the posterior left ventricular vein, the left oblique atrial vein, and the like.

Wherein the boundary optimization process comprises at least one of: 2D boundary optimization processing, 3D boundary optimization processing and data enhancement processing.

Wherein the 2D boundary optimization process comprises: and acquiring low-resolution information and high-resolution information through multiple sampling, wherein the low-resolution information can provide context semantic information of the segmented target in the whole image, namely characteristics reflecting the relation between the segmented target and the environment, the characteristics are used for judging the object type, and the high-resolution information is used for providing more fine characteristics such as gradient and the like for the segmented target.

Wherein the segmentation targets comprise coronary artery, coronary artery and coronary vein of heart.

Wherein the 3D boundary optimization process comprises: 3D convolution, 3D max pooling, and 3D up-convolution layers, the input data size is a1, a2, a3, the number of channels is c, the filter size is f, i.e., the filter dimension is f c, and the filter number is n, then the final output of the 3-dimensional convolution is:

(a1-f+1)*(a2-f+1)*(a3-f+1)*n

there are an analysis path and a synthesis path. In the analysis path, each layer contains two convolution kernels of 3 × 3, each followed by an activation function (Relu), and then a maximum pooling of 2 × 2 in each dimension merges two steps. In the synthesis path, each layer consists of 2 × 2 upward convolutions, with 2 steps in each dimension, followed by two 3 × 3 convolutions, and then Relu. The shortcut connections from the equal resolution layers in the analysis path then provide the basic high resolution features of the composite path. In the last layer, 1 × 1 convolution reduces the number of output channels.

Further, the 3D boundary optimization process includes the following operations: inputting the original image data into a 3D convolution layer for 3D convolution operation to obtain a characteristic diagram; inputting the feature map into a 3D pooling layer for pooling and nonlinear activation to obtain a first feature map; and carrying out cascade operation on the first feature map to obtain a prediction result.

Wherein the data enhancement processing includes any one of: data enhancement based on arbitrary angle rotation, data enhancement based on histogram equalization, data enhancement based on white balance, data enhancement based on mirroring operation, data enhancement based on random clipping, and data enhancement based on simulating different illumination variations.

Wherein the image data comprises a data set of the cardiac coronary artery, and a data set of the cardiac coronary vein.

It can be seen that, in the above technical solution, a target image data set with a clear boundary is obtained by processing the image of the scanned image, so as to assist a doctor to perform rapid diagnosis.

Optionally, in a possible implementation manner, after performing boundary optimization processing on the original image data set to obtain a target image data set, the method further includes: acquiring an image quality score corresponding to the image data; screening enhanced data with an image quality score larger than a preset image quality score from the image data according to the image quality score; setting the enhancement data to VRDS 4D imaging data; displaying the VRDS 4D imaging data on an output device.

It can be seen that, in the above technical scheme, enhancement data with an image quality score larger than a preset image quality score is screened out from image data according to the image quality score by acquiring the image quality score corresponding to the image data, the enhancement data is set as VRDS 4D imaging data, and finally, the VRDS 4D imaging data is displayed on an output device, thereby assisting a doctor in making a quick diagnosis.

203. And the medical imaging device determines the lesion degree of the coronary artery of the heart of the target user according to the target image data set.

Wherein, the disease condition expression information which may appear in the pathological change degree comprises: angina pectoris, myocardial infarction, heart failure, sudden cardiac death, etc.

Optionally, in a possible implementation, the determining a lesion degree of the coronary artery of the heart of the target user according to the target image data set includes: determining a lesion area of the coronary artery of the target user according to the target image data set; identifying a degree of stenosis of a blood vessel of the plaque and a degree of calcification of the plaque in the lesion region; determining a target size of the blood vessel around the coronary artery of the heart after expansion according to the stenosis degree and the calcification degree of the plaque; determining blood vessel parameters around the coronary artery of the heart according to the target size, wherein the blood vessel parameters comprise blood vessel type, diameter and length; and determining the lesion degree of the heart coronary artery of the target user according to the blood vessel parameters.

The target size can be obtained by inquiring a preset medical database to obtain blood vessel parameters around the coronary artery of the heart, inquiring the preset medical database to obtain the blood vessel parameters matched with the target size in the preset medical database, wherein the preset medical database comprises a mapping relation between the size and the blood vessel parameters.

The mapping relationship may be one-to-one, one-to-many, and many-to-many, and is not limited herein.

According to the technical scheme, the lesion characteristics in the lesion area are obtained by analyzing the lesion area, the target size of the expanded blood vessel is further obtained according to the lesion characteristics, and finally the lesion degree is determined according to the target size of the expanded blood vessel, so that the coronary artery disease information of the heart is accurately obtained.

Further, in a possible embodiment, the determining the lesion degree of the coronary artery of the heart of the target user according to the blood vessel parameter includes: acquiring a preset heart coronary artery part recognition model; inputting the blood vessel parameters into the preset heart coronary artery part identification model to obtain a lesion value of each blood vessel parameter in the blood vessel parameters; and determining the lesion degree corresponding to the blood vessel parameter according to a preset mapping relation between the lesion value and the lesion degree.

According to the technical scheme, the lesion degree is rapidly determined by presetting the identification model of the coronary artery part of the heart, more accurate acquisition of the information of the coronary artery vascular disease of the heart is realized, and the determination efficiency is also improved.

Further, in a possible embodiment, the identifying the stenosis degree of the plaque blood vessel and the calcification degree of the plaque in the lesion region includes: acquiring a lesion image of a lesion area in the target image data set; extracting a shadow part in the lesion image; detecting a shadow area size of the shadow portion; determining the calcification degree of the plaque according to the size of the shadow area; detecting a size of a blood vessel around the shadow portion; and determining the stenosis degree of the blood vessel according to the size of the blood vessel.

The plaque calcification degree determined according to the shadow area size can be obtained through a preset model or big data analysis, and the determination is not limited uniquely here.

According to the technical scheme, the calcification degree can be determined through the shadow area, the stenosis degree can be determined through the size of the blood vessel, and the accuracy and convenience in obtaining the heart coronary artery blood vessel lesion information are improved.

Optionally, in a possible implementation, the determining a lesion degree of the coronary artery of the heart of the target user according to the target image data set includes: respectively establishing a coordinate system in each target image data in the target image data set, wherein the origin of the coordinate system is the central position of a coronary artery central line in the cardiac coronary artery; performing region segmentation on the heart coronary artery according to the coordinate system to obtain region image data in each target image data; detecting the regional image data through the coordinate system to obtain spatial positions of a plurality of target pixel points, wherein the spatial positions of the plurality of target pixel points record the spatial positions corresponding to a first pixel point according to the fact that the gray value corresponding to the first pixel point belongs to the gray value corresponding to the blood vessel cell data of the heart coronary artery; dividing the spatial positions of the target pixel points according to regions to obtain a regional image data set of each region in the same region in the target image data set; obtaining a plurality of outermost vascular cell data sets corresponding to the cardiac coronary artery according to the regional image data set of each region, wherein each outermost vascular cell data set comprises a plurality of outermost vascular cell data; and searching the lesion degrees corresponding to the plurality of outmost vascular cell data sets as target lesion degrees.

Further, in a possible embodiment, the method further includes: for each of the outermost vascular cell datasets, performing the steps of: acquiring a characteristic curve of the outermost layer vascular cell data set projection in any plane; selecting any point of the characteristic curve as a starting point; starting from the starting point, marking pixel points continuously along the positive direction and the negative direction of the characteristic curve, stopping marking when a target pixel point is marked, wherein the positive direction of the characteristic curve is the transverse positive direction of the image data, the negative direction of the characteristic curve is the transverse negative direction of the image data, the target pixel point is a pixel point with the largest curvature change of a target coronary artery blood vessel section, the target blood vessel section is a blood vessel between the starting point and a target space position of a target blood vessel, the target blood vessel corresponds to an outermost layer blood vessel cell data set which is processed currently, and the target space position is a position corresponding to the target pixel point; obtaining the corresponding curvature of the target blood vessel section; and setting the curvature corresponding to the target blood vessel section as the corresponding curvature of the target blood vessel.

According to the technical scheme, more accurate positioning of the disease is achieved according to at least one abnormal heart coronary vessel positioning disease in the image data, and the identification accuracy of the heart coronary disease is improved.

204. The medical imaging device determines the type of coronary stent according to the lesion degree of the coronary artery of the heart.

Optionally, in a possible implementation, the determining the type of the coronary stent according to the lesion degree of the coronary artery of the heart includes querying a first database, and obtaining a target type of the coronary stent matched with the lesion degree in the first database, where the first database includes a mapping relationship between the lesion degree and the type of the coronary stent.

Optionally, in a possible embodiment, the determining the type of coronary stent according to the lesion degree of the coronary artery of the heart includes: obtaining the survival time of a target user corresponding to the lesion degree; bringing the lesion degree into a preset lesion model, and simulating to obtain a lesion process in the survival time; analyzing the lesion process to obtain a superposition area of the plaque and the blood vessel in the lesion process; obtaining the parameter range of the coronary stent according to the overlapping area; and determining the type of the coronary stent according to the parameter range of the coronary stent.

According to the technical scheme, the proper coronary stent is further selected according to the query database, so that quantitative model selection of the coronary stent is realized, plaques can be covered after the coronary stent is expanded to support the blood vessel stent, the plaques cannot fall off, and the operation risk of the blood vessel stent implantation is reduced.

Optionally, in a possible implementation, the medical imaging apparatus performs image processing on the scan image to obtain a target image data set, including: performing first preset processing on the scanned image to obtain a bitmap BMP data source; importing the BMP data source into a preset VRDS medical network model to obtain first medical image data, wherein the first medical image data comprise a data set of the coronary artery and a data set of the coronary artery, the data set of the coronary artery comprises fusion data of the intersection position of a left coronary artery and a right coronary artery, the data set of the coronary artery is a transfer function result of a cubic space of a tissue structure on the surface of the coronary artery and in the coronary artery, and the data set of the coronary artery is a transfer function result of a cubic space of a tissue structure on the surface of the coronary artery and in the coronary artery; importing the first medical image data into a preset cross blood vessel network model to obtain second medical image data, wherein the second medical image data comprises a data set of the cardiac coronary artery, a data set of the left coronary artery and a data set of the right coronary artery, first data in the data set of the left coronary artery and second data in the data set of the right coronary artery are independent of each other, the first data is data associated with the cross position, and the second data is data associated with the cross position; and executing second preset processing on the second medical image data to obtain the image data.

Wherein the first preset processing comprises at least one of the following operations: VRDS restriction contrast self-adaptive histogram equalization, mixed partial differential denoising and VRDS Ai elastic deformation processing.

Wherein the VRDS limited contrast adaptive histogram equalization comprises the steps of: performing regional noise contrast clipping and global contrast clipping on the map source; dividing the local histogram of the graph source into a plurality of partitions; determining a plurality of slopes of a plurality of transform functions from slopes of the cumulative histogram of the neighborhood of each of the plurality of partitions; determining a degree of contrast magnification around a pixel value of each of the plurality of partitions from the plurality of slopes; performing limit clipping processing on the plurality of partitions according to the contrast amplification degree of the periphery of the pixel value of each partition in the plurality of partitions to obtain the distribution of an effective histogram and the value of the effective available neighborhood size; and uniformly distributing the histogram with the cut-off limit to other areas of the local histogram of the graph source.

The hybrid partial differential denoising comprises the following steps: processing the image source through VRDS Ai curvature drive and VRDS Ai high-order mixed denoising, so that the curvature of the image edge is smaller than the preset curvature, and a mixed partial differential denoising model which can protect the image edge and can avoid the step effect in the smoothing process is realized;

the VRDS Ai elastic deformation processing comprises the following steps: and acquiring an image lattice of the image source, superposing positive and negative random distances on the image lattice to form a difference position matrix, performing gray level processing on each difference position in the difference position matrix to obtain a new difference position matrix, thereby realizing distortion deformation in the image, and then performing rotation, distortion and translation operations on the image.

Wherein the hybrid partial differential denoising is performed by the medical imaging device on the image source by using CDD and a higher-order denoising model.

The CDD model (Curvature drive diffusion) is formed by introducing Curvature drive on the basis of a TV (Total variation) model, and the problem that the TV model cannot repair image visual connectivity is solved.

The high-order denoising refers to denoising the image based on a Partial Differential Equation (PDE) method. In specific implementation, the graph source is subjected to noise filtering according to the change of a specified differential equation function, so as to obtain the BMP data source. The partial differential equation is the BMP data source obtained after high-order denoising, and the image denoising method based on the PDE has the characteristic of anisotropic diffusion, so that the diffusion effect of different degrees can be performed in different areas of the image source, and the effect of inhibiting noise and protecting image edge texture information is achieved.

It can be seen that in the present example, the medical imaging apparatus operates by at least one of the following image processing operations: the VRDS limits contrast self-adaptive histogram equalization, mixed partial differential denoising and VRDS Ai elastic deformation processing, improves the execution efficiency of image processing, also improves the image quality and protects the edge texture of the image.

In a possible implementation manner, the performing a first preset process on the scan image to obtain a bitmap BMP data source includes: setting the scan image as digital imaging and communications in medicine (DICOM) data of the user; parsing the DICOM data to generate a map source of the user, the map source comprising Texture2D/3D image volume data; and executing the first preset treatment on the graph source to obtain the BMP data source.

The dicom (digital Imaging and Communications in medicine), i.e., digital Imaging and Communications in medicine, is an international standard for medical images and related information. In specific implementation, the medical imaging device acquires a plurality of acquired scanning images reflecting the structural features of the coronary artery of the user, screens out at least one appropriate scanning image containing the coronary artery through definition, accuracy and the like, and then performs further processing on the scanning image to obtain a bitmap BMP data source.

In this example, the medical imaging apparatus may perform screening, analysis and first preset processing based on the acquired scan image to obtain a bitmap BMP data source, so as to improve the accuracy and definition of medical image imaging.

It can be seen that, in this example, the medical imaging apparatus processes the scan image into image data capable of reflecting the spatial structural characteristics of the coronary artery of the heart through a plurality of lines of data processing, and the left coronary artery image data and the right coronary artery image data at the intersection position are independent from each other, so that accurate presentation in a three-dimensional space is supported, and the data processing accuracy and comprehensiveness are improved.

In a possible example of the present application, the importing the BMP data source into a preset VRDS medical network model to obtain first medical image data includes: importing the BMP data source into a preset VRDS medical network model, calling each transfer function in a pre-stored transfer function set through the VRDS medical network model, and processing the BMP data source through a plurality of transfer functions in the transfer function set to obtain first medical image data, wherein the transfer function set comprises the transfer functions of the coronary artery and the coronary artery, which are preset through a reverse editor.

The BMP (full Bitmap) is a standard image file format in the Windows operating system, and can be divided into two types: a Device Dependent Bitmap (DDB) and a Device Independent Bitmap (DIB). The scan image includes any one of: CT images, MRI images, DTI images, PET-CT images.

The VRDS medical network model is a preset network model, and the training method comprises the following three steps: sampling an image and scaling; extracting and scoring the 3D convolutional neural network characteristics; and evaluating the medical imaging device and training the network. In the implementation process, sampling is firstly carried out to obtain N BMP data sources, and then M BMP data sources are extracted from the N BMP data sources according to a preset interval. It should be noted that the preset interval can be flexibly set according to the usage scenario. M BMP data sources are sampled from the N BMP, and then the sampled M BMP data sources are scaled to a fixed size (for example, S pixels long and S pixels wide), and the resulting processing result is used as an input to the 3D convolutional neural network. This takes the M BMP data sources as inputs to the 3D convolutional neural network. Specifically, 3D convolution processing is performed on the BMP data source by using a 3D convolution neural network, and a feature map is obtained.

Referring to fig. 3, fig. 3 is a flowchart illustrating a method for analyzing coronary artery of heart based on VRDS AI medical image according to an embodiment of the present application. The method comprises the following steps:

301. a medical imaging device acquires a scanned image of a coronary region of the heart.

302. The medical imaging device carries out image preprocessing according to the scanning image to obtain first image data.

303. A medical imaging device generates a raw image data set of the coronary region of the heart from the first image data.

304. And the medical imaging device performs boundary optimization processing on the original image data set to obtain a target image data set.

305. The medical imaging device determines a lesion region of the target user's coronary artery from the target set of image data.

306. A medical imaging device identifies a degree of stenosis of a blood vessel of the plaque and a degree of calcification of the plaque in the lesion region.

307. The medical imaging device determines a target size of the blood vessel around the coronary artery of the heart after expansion according to the stenosis degree and the calcification degree of the plaque.

308. The medical imaging device determines blood vessel parameters around the coronary artery of the heart according to the target size, wherein the blood vessel parameters comprise blood vessel type, diameter and length.

309. And the medical imaging device determines the lesion degree of the heart coronary artery of the target user according to the blood vessel parameters.

310. The medical imaging device determines the type of coronary stent according to the lesion degree of the coronary artery of the heart.

In addition, the lesion features in the lesion region are acquired by analyzing the lesion region, the target size of the expanded blood vessel is further acquired according to the lesion features, and finally, the lesion degree is determined according to the target size of the expanded blood vessel, so that the coronary artery disease information of the heart is acquired more accurately.

Referring to fig. 4, fig. 4 is a flowchart illustrating a method for analyzing coronary artery of heart based on VRDS AI medical image according to an embodiment of the present application.

401. A medical imaging device acquires a scanned image of a coronary region of the heart.

402. And the medical imaging device performs image processing according to the scanning image to obtain a target image data set.

403. And the medical imaging device determines the lesion degree of the coronary artery of the heart of the target user according to the target image data set.

404. The medical imaging device establishes a coordinate system in each target image data in the set of target image data, respectively.

405. And the medical imaging device performs region segmentation on the coronary artery of the heart according to the coordinate system to obtain region image data in each target image data.

406. The medical imaging device detects the regional image data through the coordinate system to obtain the spatial positions of a plurality of target pixel points, and the spatial positions of the plurality of target pixel points record the spatial positions corresponding to the first pixel point according to the fact that the gray value corresponding to the first pixel point belongs to the gray value corresponding to the blood vessel cell data of the coronary artery of the heart.

407. The medical imaging device divides the spatial positions of the target pixels according to regions to obtain a regional image data set of each region in the same region in the target image data set.

408. And the medical imaging device obtains a plurality of outermost vascular cell data sets corresponding to the coronary artery of the heart according to the regional image data set of each region, wherein each outermost vascular cell data set comprises a plurality of outermost vascular cell data.

409. The medical imaging device searches for a lesion extent corresponding to the plurality of outermost vascular cell data sets as a target lesion extent.

In addition, according to at least one heart coronary artery blood vessel positioning disease with abnormal image data, more accurate positioning disease is realized, and the identification accuracy of heart coronary artery diseases is improved.

Referring to fig. 5, an embodiment of the present application provides a schematic diagram of a medical imaging apparatus 500, and the medical imaging apparatus 500 may include:

an obtaining unit 501, configured to obtain a scan image of a coronary artery region of a heart, where the scan image further includes the coronary artery region of the heart and blood vessels around the coronary artery of the heart;

a processing unit 502, configured to perform image processing according to the scanned image to obtain a target image data set;

a determining unit 503, configured to determine a lesion degree of a coronary artery of the target user according to the target image data set;

the determining unit 503 is further configured to determine the type of the coronary stent according to the lesion degree of the coronary artery of the heart.

Optionally, the processing module 502 is specifically configured to perform image preprocessing according to the scanned image to obtain first image data; generating an original image data set of the coronary artery part of the heart according to the first image data; and carrying out boundary optimization processing on the original image data set to obtain a target image data set.

Optionally, the determining module 503 is specifically configured to determine a lesion region of the coronary artery of the target user according to the target image data set; identifying a degree of stenosis of a blood vessel of the plaque and a degree of calcification of the plaque in the lesion region; determining a target size of the blood vessel around the coronary artery of the heart after expansion according to the stenosis degree and the calcification degree of the plaque; determining blood vessel parameters around the coronary artery of the heart according to the target size, wherein the blood vessel parameters comprise blood vessel type, diameter and length; and determining the lesion degree of the heart coronary artery of the target user according to the blood vessel parameters.

Optionally, the determining module 503 is specifically configured to obtain a preset identification model of a coronary artery region of the heart; inputting the blood vessel parameters into the preset heart coronary artery part identification model to obtain a lesion value of each blood vessel parameter in the blood vessel parameters; and determining the lesion degree corresponding to the blood vessel parameter according to a preset mapping relation between the lesion value and the lesion degree.

Optionally, the determining module 503 is specifically configured to acquire a lesion image of a lesion area in the target image data set; extracting a shadow part in the lesion image; detecting a shadow area size of the shadow portion; determining the calcification degree of the plaque according to the size of the shadow area; detecting a size of a blood vessel around the shadow portion; and determining the stenosis degree of the blood vessel according to the size of the blood vessel.

Optionally, the determining module 503 is specifically configured to query a first database, and obtain a target coronary stent type matched with the lesion degree in the first database, where the first database includes a mapping relationship between the lesion degree and the coronary stent type.

Optionally, the determining module 503 is specifically configured to obtain the survival time of the target user corresponding to the lesion degree; bringing the lesion degree into a preset lesion model, and simulating to obtain a lesion process in the survival time; analyzing the lesion process to obtain a superposition area of the plaque and the blood vessel in the lesion process; obtaining the parameter range of the coronary stent according to the overlapping area; and determining the type of the coronary stent according to the parameter range of the coronary stent.

Optionally, the determining module 503 is specifically configured to respectively establish a coordinate system in each target image data in the target image data set, where an origin of the coordinate system is a central position of a coronary artery centerline in the coronary artery of the heart; performing region segmentation on the heart coronary artery according to the coordinate system to obtain region image data in each target image data; detecting the regional image data through the coordinate system to obtain spatial positions of a plurality of target pixel points, wherein the spatial positions of the plurality of target pixel points record the spatial positions corresponding to a first pixel point according to the fact that the gray value corresponding to the first pixel point belongs to the gray value corresponding to the blood vessel cell data of the heart coronary artery; dividing the spatial positions of the target pixel points according to regions to obtain a regional image data set of each region in the same region in the target image data set; obtaining a plurality of outermost vascular cell data sets corresponding to the cardiac coronary artery according to the regional image data set of each region, wherein each outermost vascular cell data set comprises a plurality of outermost vascular cell data; and searching the lesion degrees corresponding to the plurality of outmost vascular cell data sets as target lesion degrees.

Optionally, for each of the outermost vascular cell data sets, performing the following steps: acquiring a characteristic curve of the outermost layer vascular cell data set projection of the current treatment on any plane; selecting any point of the characteristic curve as a starting point; starting from the starting point, marking pixel points continuously along the positive direction and the negative direction of the characteristic curve, stopping marking when a target pixel point is marked, wherein the positive direction of the characteristic curve is the transverse positive direction of the image data, the negative direction of the characteristic curve is the transverse negative direction of the image data, the target pixel point is a pixel point with the largest curvature change of a target coronary artery blood vessel section, the target blood vessel section is a blood vessel between the starting point and a target space position of a target blood vessel, the target blood vessel corresponds to an outermost layer blood vessel cell data set which is processed currently, and the target space position is a position corresponding to the target pixel point; obtaining the corresponding curvature of the target blood vessel section; and setting the curvature corresponding to the target blood vessel section as the corresponding curvature of the target blood vessel.

Referring to fig. 6, fig. 6 is a schematic structural diagram of a medical imaging apparatus in a hardware operating environment according to an embodiment of the present application. As shown in fig. 6, a medical imaging apparatus in a hardware operating environment according to an embodiment of the present application may include:

a processor 601, such as a CPU.

The memory 602 may alternatively be a high speed RAM memory or a stable memory such as a disk memory.

A communication interface 603 for implementing connection communication between the processor 601 and the memory 602.

Those skilled in the art will appreciate that the configuration of the medical imaging device shown in FIG. 6 is not intended to be limiting and may include more or fewer components than shown, or some components in combination, or a different arrangement of components.

As shown in fig. 6, the memory 602 may include therein an operating system, a network communication module, and a program for information processing. An operating system is a program that manages and controls the hardware and software resources of the medical imaging device, a program that supports personnel management, and the execution of other software or programs. The network communication module is used for communication among the components in the memory 602 and with other hardware and software in the medical imaging apparatus.

In the medical imaging apparatus shown in fig. 6, a processor 601 is configured to execute a program for information migration stored in a memory 602, and implement the following steps: acquiring a scan image of a coronary artery region of a heart, wherein the scan image further comprises the coronary artery region of the heart and blood vessels around the coronary artery of the heart; performing image processing according to the scanned image to obtain a target image data set; determining the lesion degree of the coronary artery of the target user according to the target image data set; and determining the type of the coronary stent according to the lesion degree of the coronary artery of the heart.

For specific implementation of the medical imaging apparatus according to the present application, reference may be made to the above embodiments of the method for analyzing a coronary artery of a heart based on VRDS AI medical images, which are not described herein again.

The present application further provides a computer readable storage medium for storing a computer program, the stored computer program being executable by the processor to perform the steps of: acquiring a scan image of a coronary artery region of a heart, wherein the scan image further comprises the coronary artery region of the heart and blood vessels around the coronary artery of the heart; performing image processing according to the scanned image to obtain a target image data set; determining the lesion degree of the coronary artery of the target user according to the target image data set; and determining the type of the coronary stent according to the lesion degree of the coronary artery of the heart.

For specific implementation of the computer-readable storage medium according to the present application, reference may be made to the above-mentioned embodiments of the method for analyzing cardiac coronary artery based on VRDS AI medical images, which are not described herein again.

It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art should understand that the present application is not limited by the order of acts described, as some steps may be performed in other orders or simultaneously according to the present application. Further, those skilled in the art should also appreciate that the embodiments described in this specification are preferred embodiments and that the acts and modules involved are not necessarily required for this application.

In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical division, and in actual implementation, there may be other divisions, for example, multiple modules or components may be combined or integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or modules through some interfaces, and may be in an electrical or other form.

The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.

In addition, functional modules in the embodiments of the present application may be integrated into one processing module, or each of the modules may exist alone physically, or two or more modules are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode.

The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a medical imaging apparatus, or a network device) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.

The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.

Claims

The analysis method of the heart coronary artery based on the VRDS AI medical image is characterized by comprising the following steps:

acquiring a scan image of a coronary artery region of a heart, wherein the scan image further comprises the coronary artery region of the heart and blood vessels around the coronary artery of the heart;

performing image processing according to the scanned image to obtain a target image data set;

determining the lesion degree of the coronary artery of the target user according to the target image data set;

and determining the type of the coronary stent according to the lesion degree of the coronary artery of the heart.
The method of claim 1, wherein said performing image processing based on said scanned image to obtain a target image data set comprises:

carrying out image preprocessing according to the scanned image to obtain first image data;

generating an original image data set of the coronary artery part of the heart according to the first image data;

and carrying out boundary optimization processing on the original image data set to obtain a target image data set.
The method of claim 1 or 2, wherein said determining a lesion size of a coronary artery of a heart of the target user from the target set of image data comprises:

determining a lesion area of the coronary artery of the target user according to the target image data set;

identifying a degree of stenosis of a blood vessel of the plaque and a degree of calcification of the plaque in the lesion region;

determining a target size of the blood vessel around the coronary artery of the heart after expansion according to the stenosis degree and the calcification degree of the plaque;

determining blood vessel parameters around the coronary artery of the heart according to the target size, wherein the blood vessel parameters comprise blood vessel type, diameter and length;

and determining the lesion degree of the heart coronary artery of the target user according to the blood vessel parameters.
The method of claim 3, wherein said determining a lesion level of the coronary artery of the heart of the target user from the blood vessel parameter comprises:

acquiring a preset heart coronary artery part recognition model;

inputting the blood vessel parameters into the preset heart coronary artery part identification model to obtain a lesion value of each blood vessel parameter in the blood vessel parameters;

and determining the lesion degree corresponding to the blood vessel parameter according to a preset mapping relation between the lesion value and the lesion degree.
The method of claim 3, wherein the identifying of the degree of stenosis of the blood vessels of the plaque and the degree of calcification of the plaque in the diseased region comprises:

acquiring a lesion image of a lesion area in the target image data set;

extracting a shadow part in the lesion image;

detecting a shadow area size of the shadow portion;

determining the calcification degree of the plaque according to the size of the shadow area;

detecting a size of a blood vessel around the shadow portion;

and determining the stenosis degree of the blood vessel according to the size of the blood vessel.
The method of claim 1, wherein determining the type of coronary stent based on the extent of lesion in the coronary artery of the heart comprises:

and inquiring a first database to obtain the target coronary stent type matched with the lesion degree in the first database, wherein the first database comprises the mapping relation between the lesion degree and the coronary stent type.
The method of claim 1, wherein determining the type of coronary stent based on the extent of lesion in the coronary artery of the heart comprises:

obtaining the survival time of a target user corresponding to the lesion degree;

bringing the lesion degree into a preset lesion model, and simulating to obtain a lesion process in the survival time;

analyzing the lesion process to obtain a superposition area of the plaque and the blood vessel in the lesion process;

obtaining the parameter range of the coronary stent according to the overlapping area;

and determining the type of the coronary stent according to the parameter range of the coronary stent.
The method of claim 1 or 2, wherein said determining a lesion size of a coronary artery of a heart of the target user from the target set of image data comprises:

respectively establishing a coordinate system in each target image data in the target image data set, wherein the origin of the coordinate system is the central position of a coronary artery central line in the cardiac coronary artery;

performing region segmentation on the heart coronary artery according to the coordinate system to obtain region image data in each target image data;

detecting the regional image data through the coordinate system to obtain spatial positions of a plurality of target pixel points, wherein the spatial positions of the plurality of target pixel points record the spatial positions corresponding to a first pixel point according to the fact that the gray value corresponding to the first pixel point belongs to the gray value corresponding to the blood vessel cell data of the heart coronary artery;

dividing the spatial positions of the target pixel points according to regions to obtain a regional image data set of each region in the same region in the target image data set;

obtaining a plurality of outermost vascular cell data sets corresponding to the cardiac coronary artery according to the regional image data set of each region, wherein each outermost vascular cell data set comprises a plurality of outermost vascular cell data;

and searching the lesion degrees corresponding to the plurality of outmost vascular cell data sets as target lesion degrees.
The method of claim 8, further comprising:

for each of the outermost vascular cell datasets, performing the steps of:

acquiring a characteristic curve of the outermost layer vascular cell data set projection in any plane;

selecting any point of the characteristic curve as a starting point;

starting from the starting point, marking pixel points continuously along the positive direction and the negative direction of the characteristic curve, stopping marking when a target pixel point is marked, wherein the positive direction of the characteristic curve is the transverse positive direction of the image data, the negative direction of the characteristic curve is the transverse negative direction of the image data, the target pixel point is a pixel point with the largest curvature change of a target coronary artery blood vessel section, the target blood vessel section is a blood vessel between the starting point and a target space position of a target blood vessel, the target blood vessel corresponds to an outermost layer blood vessel cell data set which is processed currently, and the target space position is a position corresponding to the target pixel point; obtaining the corresponding curvature of the target blood vessel section;

and setting the curvature corresponding to the target blood vessel section as the corresponding curvature of the target blood vessel.
A medical imaging device, comprising:

an acquisition unit for acquiring a scan image of a coronary artery region of a heart, wherein the scan image further comprises the coronary artery region of the heart and blood vessels around the coronary artery of the heart;

the processing unit is used for carrying out image processing according to the scanning image to obtain a target image data set;

the determining unit is used for determining the lesion degree of the coronary artery of the heart of the target user according to the target image data set;

the determining unit is further used for determining the type of the coronary stent according to the lesion degree of the coronary artery of the heart.
The apparatus according to claim 10, wherein the processing module is specifically configured to perform image preprocessing according to the scanned image to obtain first image data; generating an original image data set of the coronary artery part of the heart according to the first image data; and carrying out boundary optimization processing on the original image data set to obtain a target image data set.
The apparatus according to claim 10 or 11, wherein the determining means is specifically configured to determine a lesion region of the coronary artery of the heart of the target user from the target set of image data; identifying a degree of stenosis of a blood vessel of the plaque and a degree of calcification of the plaque in the lesion region; determining a target size of the blood vessel around the coronary artery of the heart after expansion according to the stenosis degree and the calcification degree of the plaque; determining blood vessel parameters around the coronary artery of the heart according to the target size, wherein the blood vessel parameters comprise blood vessel type, diameter and length; and determining the lesion degree of the heart coronary artery of the target user according to the blood vessel parameters.
The apparatus according to claim 12, wherein the determining means is specifically configured to obtain a preset cardiac coronary location identification model; inputting the blood vessel parameters into the preset heart coronary artery part identification model to obtain a lesion value of each blood vessel parameter in the blood vessel parameters; and determining the lesion degree corresponding to the blood vessel parameter according to a preset mapping relation between the lesion value and the lesion degree.
The apparatus according to claim 12, wherein the determining module is specifically configured to acquire a lesion image of a lesion region in the target image data set; extracting a shadow part in the lesion image; detecting a shadow area size of the shadow portion; determining the calcification degree of the plaque according to the size of the shadow area; detecting a size of a blood vessel around the shadow portion; and determining the stenosis degree of the blood vessel according to the size of the blood vessel.
The apparatus according to claim 10, wherein the determining module is specifically configured to query a first database, to obtain a target coronary stent type matching the lesion degree in the first database, and the first database includes a mapping relationship between the lesion degree and the coronary stent type.
The device according to claim 10, wherein the determining module is specifically configured to obtain a survival time of the target user corresponding to the lesion degree; bringing the lesion degree into a preset lesion model, and simulating to obtain a lesion process in the survival time; analyzing the lesion process to obtain a superposition area of the plaque and the blood vessel in the lesion process; obtaining the parameter range of the coronary stent according to the overlapping area; and determining the type of the coronary stent according to the parameter range of the coronary stent.
The apparatus according to claim 10 or 11, wherein the determining module is specifically configured to establish a coordinate system in each target image data in the target image data set, and an origin of the coordinate system is a central position of a coronary centerline in the coronary artery of the heart; performing region segmentation on the heart coronary artery according to the coordinate system to obtain region image data in each target image data; detecting the regional image data through the coordinate system to obtain spatial positions of a plurality of target pixel points, wherein the spatial positions of the plurality of target pixel points record the spatial positions corresponding to a first pixel point according to the fact that the gray value corresponding to the first pixel point belongs to the gray value corresponding to the blood vessel cell data of the heart coronary artery; dividing the spatial positions of the target pixel points according to regions to obtain a regional image data set of each region in the same region in the target image data set; obtaining a plurality of outermost vascular cell data sets corresponding to the cardiac coronary artery according to the regional image data set of each region, wherein each outermost vascular cell data set comprises a plurality of outermost vascular cell data; and searching the lesion degrees corresponding to the plurality of outmost vascular cell data sets as target lesion degrees.
The apparatus according to claim 17, wherein for each outermost vascular cell data set, the following steps are performed: acquiring a characteristic curve of the outermost layer vascular cell data set projection in any plane; selecting any point of the characteristic curve as a starting point; starting from the starting point, marking pixel points continuously along the positive direction and the negative direction of the characteristic curve, stopping marking when a target pixel point is marked, wherein the positive direction of the characteristic curve is the transverse positive direction of the image data, the negative direction of the characteristic curve is the transverse negative direction of the image data, the target pixel point is a pixel point with the largest curvature change of a target coronary artery blood vessel section, the target blood vessel section is a blood vessel between the starting point and a target space position of a target blood vessel, the target blood vessel corresponds to an outermost layer blood vessel cell data set which is processed currently, and the target space position is a position corresponding to the target pixel point; obtaining the corresponding curvature of the target blood vessel section; and setting the curvature corresponding to the target blood vessel section as the corresponding curvature of the target blood vessel.
A medical imaging apparatus comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and generated as instructions for execution by the processor to perform the steps of the method of any of claims 1-9.
A computer-readable storage medium, characterized in that the computer-readable storage medium is used to store a computer program, which is executed by the processor, to implement the method of any of claims 1-9.