WO2021030995A1

WO2021030995A1 - Inferior vena cava image analysis method and product based on vrds ai

Info

Publication number: WO2021030995A1
Application number: PCT/CN2019/101165
Authority: WO
Inventors: 李戴维伟; 李斯图尔特平
Original assignee: 未艾医疗技术(深圳)有限公司
Priority date: 2019-08-16
Filing date: 2019-08-16
Publication date: 2021-02-25
Also published as: CN114365188A

Abstract

Disclosed are an inferior vena cava image analysis method and product based on VRDS AI, applied to a medical imaging apparatus. The method comprises: acquiring a scanned image including an inferior vena cava of a target user (S201); processing the scanned image to obtain a target image, wherein the target image comprises three-dimensional spatial image data of the inferior vena cava (S202); extracting a reference feature data set according to the target image, wherein the reference feature data set is used to reflect a physiological feature of the inferior vena cava of the target user (S203); determining an abnormality type of the inferior vena cava according to the reference feature data set (S204); and outputting the abnormality type of the inferior vena cava (S205). The method helps to improve the comprehensiveness, accuracy and detection efficiency of a medical imaging apparatus for the analysis of a human inferior vena cava.

Description

Analysis methods and products based on VRDS AI inferior vena cava image

Technical field

This application relates to the technical field of medical imaging devices, in particular to an analysis method and product based on VRDS AI inferior vena cava images.

Background technique

At present, doctors use Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Diffusion Tensor Imaging (DTI), and Positron Emission Computed Tomography (Computed Tomography). Tomography, PET) and other technologies obtain information such as the shape, location, and topology of the diseased tissue. Doctors still use continuous two-dimensional slice data to view and read to diagnose the condition. As current medical imaging equipment cannot visually present the three-dimensional image data of the inferior vena cava, it is currently impossible to realize health diagnosis based on veins.

Summary of the invention

The embodiments of the present application provide an analysis method and product based on VRDS AI inferior vena cava image, in order to improve the comprehensiveness, accuracy and detection efficiency of the medical imaging device's analysis of the human inferior vena cava.

In the first aspect, an embodiment of the present application provides an analysis method based on VRDS AI inferior vena cava image, which is applied to a medical imaging device; the method includes:

Acquiring a scanned image of the inferior vena cava containing the target user;

Processing the scanned image to obtain a target image, the target image including three-dimensional spatial image data of the inferior vena cava;

Extracting a reference feature data set according to the target image, the reference feature data set being used to reflect the physiological characteristics of the inferior vena cava of the target user;

Determine the abnormal category of the inferior vena cava according to the reference feature data set;

Output the abnormal category of the inferior vena cava.

In a second aspect, the embodiments of the present application provide a medical imaging device, which is applied to a medical imaging device; the medical imaging device includes a processing unit and a communication unit, wherein,

The processing unit is used to obtain a scanned image of the inferior vena cava including the target user through the communication unit; and used to process the scanned image to obtain a target image, the target image including the three-dimensional space image of the inferior vena cava Data; and used to extract a reference feature data set based on the target image, the reference feature data set used to reflect the physiological characteristics of the inferior vena cava of the target user; and used to determine the next feature data set based on the reference feature data set The abnormal category of the vena cava; and the abnormal category for outputting the inferior vena cava.

In a third aspect, an embodiment of the present application provides a medical imaging device, including a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured by the above Executed by a processor, the above-mentioned program includes instructions for executing steps in any method of the first aspect of the embodiments of the present application.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the foregoing computer-readable storage medium stores a computer program for electronic data exchange, wherein the foregoing computer program enables a computer to execute In one aspect, some or all of the steps described in any method.

In a fifth aspect, embodiments of the present application provide a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute For example, some or all of the steps described in any method of the first aspect. The computer program product may be a software installation package.

It can be seen that in the embodiment of the present application, the medical imaging device first obtains a scanned image of the inferior vena cava containing the target user, and secondly, processes the scanned image to obtain a target image of the inferior vena cava. The target image includes three-dimensional spatial image data of the inferior vena cava. , Again, extract the reference feature data set of the inferior vena cava according to the target image, the reference feature data set is used to reflect the physiological characteristics of the inferior vena cava of the target user, and then the abnormal category of the inferior vena cava is determined according to the reference feature data set, and finally , Output the abnormal category of the inferior vena cava. It can be seen that the medical imaging device of the present application processes the scanned image of the inferior vena cava of the target user to obtain the target image including the three-dimensional spatial image data of the inferior vena cava, thereby comprehensively analyzing the physiological characteristics of the inferior vena cava, accurately identifying and outputting the inferior vena cava The abnormal types of veins help to improve the comprehensiveness, accuracy and efficiency of the inferior vena cava analysis performed by the medical imaging device.

Description of the drawings

In order to more clearly describe the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

FIG. 1 is a schematic structural diagram of a medical image intelligent analysis and processing system based on VRDS Ai according to an embodiment of the present application;

Fig. 2a is a schematic flow chart of an analysis method based on VRDS AI inferior vena cava image provided by an embodiment of the present application;

Fig. 2b is a schematic diagram of a disease entry interface provided by an embodiment of the present application;

3 is a schematic structural diagram of a medical imaging device provided by an embodiment of the present application;

Fig. 4 is a block diagram of functional units of a medical imaging device provided by an embodiment of the present application.

detailed description

In order to enable those skilled in the art to better understand the solution of the application, the technical solutions in the embodiments of the application will be clearly and completely described below in conjunction with the drawings in the embodiments of the application. Obviously, the described embodiments are only It is a part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of this application.

The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific sequence. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes unlisted steps or units, or optionally also includes Other steps or units inherent to these processes, methods, products or equipment.

Reference to "embodiments" herein means that a specific feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art clearly and implicitly understand that the embodiments described herein can be combined with other embodiments.

The medical imaging devices involved in the embodiments of this application refer to various instruments that use various media as information carriers to reproduce the internal structure of the human body as images. The image information and the actual structure of the human body have spatial and temporal distributions. Correspondence. "DICOM data" refers to the original image file data collected by medical equipment that reflects the internal structural characteristics of the human body. It can include electronic computed tomography CT, magnetic resonance MRI, diffusion tensor imaging DTI, and positron emission computed tomography PET- For information such as CT, "image source" refers to the Texture2D/3D image volume data generated by analyzing the original DICOM data. "VRDS" refers to the Virtual Reality Doctor system (VRDS for short).

Please refer to FIG. 1. FIG. 1 is a schematic structural diagram of a VRDS Ai medical image intelligent analysis and processing system 100 based on an embodiment of the present application. The system 100 includes a medical imaging device 110 and a network database 120. The medical imaging device 110 may include The local medical imaging device 111 and/or the terminal medical imaging device 112, the local medical imaging device 111 or the terminal medical imaging device 112 are used for the analysis algorithm based on the VRDS AI inferior vena cava image presented in the embodiment of this application based on the original DICOM data Based on the recognition, positioning, four-dimensional volume rendering, and abnormal analysis of the inferior vena cava region of the human body, four-dimensional three-dimensional imaging effects are realized (the four-dimensional medical image specifically refers to the medical image including the internal spatial structure characteristics and external spatial structure of the displayed tissue The internal spatial structure characteristic means that the slice data inside the tissue is not lost, that is, the medical imaging device can present the internal structure of the target organ, blood vessel and other tissues. The external spatial structural characteristic refers to the environmental characteristics between the tissue and the tissue, including The spatial location characteristics between tissues (including crossing, spacing, fusion, etc., such as the edge structure characteristics of the crossing position between the kidney and the artery, etc.), the local medical imaging device 111 can also be used relative to the terminal medical imaging device 112 To edit the source data of the image to form the transfer function result of the four-dimensional human body image, the transfer function result can include the transfer function result of the surface of the internal organs of the human body and the tissue structure of the internal organs of the human body, and the transfer function result of the cube space, such as transfer The cube edit box and arc edit array quantity, coordinates, color, transparency and other information needed by the function. The network database 120 may be, for example, a cloud server. The network database 120 is used to store the image source generated by analyzing the original DICOM data and the transfer function result of the four-dimensional human body image edited by the local medical imaging device 111. The image source may be from multiple sources. A local medical imaging device 111 to realize interactive diagnosis of multiple doctors.

When the user performs specific image display through the above-mentioned medical imaging device 110, he can select a display or a head-mounted display of virtual reality VR (Head mounted Displays Set, HMDS) to display in combination with operating actions. The operating actions refer to the user’s actions through the medical imaging device. An external intake device, such as a mouse, keyboard, etc., controls the operation of the four-dimensional human body image to achieve human-computer interaction. The operation action includes at least one of the following: (1) Change the color and/or of a specific organ/tissue Transparency, (2) positioning zoom view, (3) rotating view, realizing multi-view 360-degree observation of four-dimensional human body image, (4) "entering" human organs to observe internal structure, real-time clipping effect rendering, (5) moving up and down view.

The following describes in detail the tumor recognition algorithm based on VRDS Ai medical imaging involved in the embodiments of the present application.

Please refer to Fig. 2a. Fig. 2a is a schematic flowchart of an analysis method based on VRDS AI inferior vena cava image provided by an embodiment of the present application, which is applied to the medical imaging device described in Fig. 1; as shown in the figure, this is based on VRDS AI inferior vena cava image analysis methods include:

S201: The medical imaging device acquires a scanned image of the inferior vena cava containing the target user;

Wherein, the scan image includes any one of the following: CT image, MRI image, DTI image, PET-CT image.

S202: The medical imaging device processes the scanned image to obtain a target image, where the target image includes three-dimensional spatial image data of the inferior vena cava;

Wherein, the target image may include the image information of the inferior vena cava and at least one of the following veins: superior vena cava, external jugular vein, lateral thoracic vein, intercostal vein, superior thoracic-abdominal vein, azygos vein, inferior abdominal or vein, Common hip vein, great saphenous vein, vertebral vein, vertebral venous plexus, internal breast vein, semi-odd vein, ascending lumbar vein, inferior vena cava of abdominal wall.

S203: The medical imaging device extracts a reference feature data set according to the target image, where the reference feature data set is used to reflect the physiological characteristics of the inferior vena cava of the target user;

Wherein, the physiological characteristics of the inferior vena cava refer to the feature data set based on prior experience that can reflect the abnormal category of the inferior vena cava. Because the formation of the inferior vena cava of the human body includes complex connection processes and various embryonic veins The degenerative process is the main conduit for the venous return of the lower limbs and abdominal organs to the right atrium. Therefore, researchers can use big data test analysis to find multiple types of abnormalities that can be used to locate abnormal categories from multiple types of physiological characteristics of the inferior vena cava. Characteristic data.

S204: The medical imaging device determines the abnormal category of the inferior vena cava according to the reference feature data set;

Among them, the abnormal categories include: absence of inferior vena cava (also known as drainage of inferior vena cava through azygos or semi-odd vena), duplication of inferior vena cava, ectopic left side of inferior vena cava, and continuation of inferior vena cava Thoracic vein, posterior inferior vena cava, ureter, tumor involving tumor thrombus, tumor involving thrombus,

Wherein, the category of the feature data includes at least one of the following feature data associated with the abnormal category: collateral distribution characteristic data, bilateral superior main vein status analysis data, analysis of the relationship between the superior renal inferior vena cava and azygos vein or semi-azygos vein Data, inferior vena cava development analysis data, tumor analysis data. In specific implementation, the medical imaging device can extract the characteristic data comprehensively and accurately by analyzing the three-dimensional spatial image data of the inferior vena cava.

In this possible example, the characteristic data of the inferior vena cava corresponding to the absence of the inferior vena cava in the abnormal database includes collateral distribution characteristic data, and the tissue structure defect presented by the collateral distribution characteristic data includes at least one of the following Species: venous insufficiency of the lower extremities, idiopathic deep vein thrombosis, collateral circulation of the lumbar vein.

Wherein, the collateral distribution characteristic data is obtained by analyzing the data of the area of the main vein on both sides of the target image.

In this possible example, the characteristic data of the inferior vena cava corresponding to the repeated malformation of the inferior vena cava in the abnormal database includes bilateral superior main vein state analysis data, and the bilateral superior main vein state analysis data presents The histological structural defects include the preservation of the bilateral superior main veins without degeneration.

Wherein, the bilateral superior main vein state analysis data is obtained by analyzing the data of the bilateral superior main vein area of the target image.

In this possible example, the characteristic data of the inferior vena cava corresponding to the thoracic vein in the abnormal database includes the analysis data of the relationship between the superior renal inferior vena cava and the odd vein or the semi-odd vein, and the renal The tissue structure defects presented in the analysis data of the relationship between the upper inferior vena cava and azygos or semi-odd veins include at least one of the following: abnormal continuation of the inferior vena cava and azygos or semi-odd vein, and the inferior vena cava of the upper kidney enters the atypical vein It flows back into the heart through the superior vena cava, or into the hemi-odd vein and then into the odd vein, the hemi-odd vein directly drains into the coronary sinus through the permanent left superior vena cava or drains into the left brachiocephalic vein through the accessory hemi-odd vein.

In this possible example, the characteristic data of the inferior vena cava corresponding to the posterior ureter of the inferior vena cava in the abnormal database includes the analysis data of the development of the inferior vena cava of the inferior kidney, and the analysis data of the development of the inferior vena cava of the inferior kidney The presented tissue structural defects include at least one of the following: the inferior renal vena cava develops from the right posterior main vein instead of the right upper main vein.

In this possible example, the characteristic data of the inferior vena cava corresponding to the tumor-involved tumor thrombus and the tumor-involved thrombosis in the abnormal database includes tumor analysis data.

S205: The medical imaging device outputs the abnormal category of the inferior vena cava.

In specific implementation, the medical imaging device can specifically output probability distributions of multiple abnormal categories.

In a possible example, the medical imaging device determining the abnormal category of the inferior vena cava according to the reference feature data set includes: the medical imaging device obtains an abnormal database, and the abnormal database includes the characteristics of the inferior vena cava Correspondence between the data and the abnormal category of the inferior vena cava; using the reference feature data set as a query identifier, query the abnormal database to obtain an abnormal category matching the reference feature data set.

It can be seen that, in this example, the medical imaging device can quickly obtain the abnormal category of the inferior venous vein of the user currently to be tested by means of a table lookup, which improves the efficiency of detection and analysis and ensures real-time performance.

In a possible example, the medical imaging device determining the abnormal category of the inferior vena cava according to the reference feature data set includes: the medical imaging device obtains a pre-trained abnormal category identification model; and comparing the reference feature The data set is imported into the abnormal category identification model to obtain an output result. The output result includes the probability distribution of a single abnormal category or multiple abnormal categories.

Among them, the abnormal category identification model can be a commonly used neural network model, etc., which is not uniquely limited here.

It can be seen that, in this example, the medical imaging device is based on the artificial intelligence AI processing mechanism, which has strong processing capabilities and a wider range of applications, which is of significant help in solving intractable diseases.

In a possible example, the medical imaging device extracting the reference feature data set of the inferior vena cava according to the target image includes: the medical imaging device obtains the information entered by the doctor for the target user through the condition entry interface. The preliminary diagnosis result data of the inferior vena cava, the preliminary diagnosis result data includes description information for the abnormal category of the inferior vena cava; at least one category of feature data to be extracted is determined according to the preliminary diagnosis result data; The target image extracts feature data of the at least one category.

Among them, as shown in Figure 2b is a schematic diagram of a disease entry interface, which includes a schematic diagram of the superior and inferior vena cava, a prompt box for the selected area, a prompt box for abnormal categories associated with the selected area, and a display box for preliminary diagnosis result data. After selecting the corresponding area, the possible abnormal category can be displayed in the abnormal category prompt box associated with the selected area, and the selected area is output in the selected area prompt box, and the user selects a certain abnormal category in the abnormal category prompt box associated with the selected area. After an abnormal category, the corresponding abnormal category is displayed in the display box of the preliminary diagnosis result data. The preliminary diagnosis result data may include one or more abnormalities selected by the doctor from the abnormality category display interface, and the multiple abnormality categories presented on the abnormality category display interface support statistical probability display, that is, the percentage shown in the figure, so as to be accurate Conveniently operate for doctors to improve operation convenience and accuracy.

It can be seen that, in this example, the medical imaging device can first input the preliminary diagnosis result based on the doctor's experience, and then perform exclusive disease analysis and data processing, thereby improving processing efficiency.

In a possible example, the medical imaging device extracting the reference feature data set of the inferior vena cava according to the target image includes: the medical imaging device obtains the inferior cavity entered by the target user through a disease entry interface The condition description data of the vein; the condition description data is imported into the pre-trained condition prediction model to obtain the condition prediction result, the condition prediction result includes the abnormal category of the inferior vena cava; the condition to be extracted is determined according to the condition prediction result At least one category of feature data; extracting the feature data of the at least one category according to the target image.

Among them, the condition entry interface can output as comprehensively and accurately as possible issues or topics that are strongly related to various diseases of the inferior vena cava based on the expert database, so as to facilitate accurate entry and improve accuracy.

It can be seen that in this example, the medical imaging device can predict potential abnormalities based on the description of the condition entered by the user, and then exclusively process the image data to improve the processing efficiency.

In a possible example, processing the scanned image by the medical imaging device to obtain the target image of the inferior vena cava includes: the medical imaging device generates a bitmap BMP data source according to the scanned image; and according to the BMP The data source generates first venous image data, the first venous image data includes an original data set of the inferior vena cava, and the original data set is an image of the surface of the inferior vena cava and the tissue structure inside the inferior vena cava A transfer function result in a cube space; generating second vein image data according to the first vein image data, the second vein image data including a segmentation data set of the inferior vena cava, the segmentation data set including a cross position relationship Mutually independent image data of the inferior vena cava; processing the second venous image data to obtain a target image of the inferior vena cava.

It can be seen that in this example, the medical imaging device processes the scanned image into image data that can reflect the spatial structure characteristics of the inferior vena cava through a series of data processing, and the venous image data at the crossing position are independent of each other, supporting accurate presentation in three-dimensional space, and improving Accuracy and comprehensiveness of data processing.

In addition, the specific implementation of generating a bitmap BMP data source according to the scanned image includes: using the scanned image as the medical digital imaging and communication DICOM data of the target user; parsing the DICOM data to generate the image source of the target user, the The image source includes texture 2D/3D image volume data; the BMP data source is obtained by performing first preset processing on the image source, and the first preset processing includes at least one of the following operations: VRDS restricted contrast adaptive histogram Image equalization, hybrid partial differential denoising, VRDS Ai elastic deformation processing.

Among them, the DICOM (Digital Imaging and Communications in Medicine) refers to medical digital imaging and communication, and is an international standard for medical images and related information. In a specific implementation, the medical imaging device first acquires multiple scanned images that reflect the internal structural characteristics of the target user's human body, and can screen out at least one suitable scanned image that contains the target organ through sharpness and accuracy. Further processing is performed on the scanned image to obtain a bitmap BMP data source. It can be seen that, in this example, the medical imaging device can obtain a bitmap BMP data source after filtering, parsing, and first preset processing based on the acquired scanned image, which improves the accuracy and clarity of medical image imaging.

Wherein, the VRDS limited contrast adaptive histogram equalization includes the following steps: regional noise ratio limiting, global contrast limiting; dividing the local histogram of the image source into multiple partitions, and for each partition, according to the neighbors of the partition The slope of the cumulative histogram of the domain determines the slope of the transformation function, and the degree of contrast amplification around the pixel value of the partition is determined according to the slope of the transformation function, and then the limit cropping process is performed according to the degree of contrast amplification to generate the effective histogram. At the same time, it also generates effective and usable neighborhood size values, and evenly distributes these cropped parts of the histogram to other areas of the histogram; the hybrid partial differential denoising includes the following steps: driving by VRDS Ai curvature and The VRDS Ai high-order hybrid denoising makes the curvature of the image edge less than the preset curvature, and realizes a hybrid partial differential denoising model that can protect the image edge and avoid the step effect in the smoothing process; the VRDS Ai elastic deformation processing includes The following steps: On the image dot matrix, superimpose the positive and negative random distances to form a difference position matrix, and then form a new dot matrix with the grayscale at each difference position, which can realize the distortion and deformation of the image, and then the image Perform rotation, twist, and translation operations.

Among them, the hybrid partial differential denoising can use CDD and high-order denoising models to process the image source; the CDD model (Curvature Driven Diffusions) model is based on the TV (Total Variation) model with the introduction of curvature drive and It solves the problem that the TV model cannot repair the visual connectivity of the image. Among them, high-order denoising refers to denoising the image based on the partial differential equation (PDE) method. In the specific implementation, let the image source perform noise filtering according to the specified differential equation function change, thereby filtering out the noise in the image source, and the solution of the partial differential equation is the BMP data source obtained after denoising The PDE-based image denoising method has the characteristics of anisotropic diffusion, so it can perform different degrees of diffusion in different regions of the image source, thereby achieving the effect of suppressing noise while protecting the edge texture information of the image.

It can be seen that, in this example, the medical imaging device uses at least one of the following image processing operations: VRDS limited contrast adaptive histogram equalization, hybrid partial differential denoising, VRDS Ai elastic deformation processing, which improves the execution efficiency of image processing, and Improve the image quality and protect the edge texture of the image.

In this possible example, the generating the first vein image data according to the BMP data source includes: importing the BMP data source into a preset VRDS medical network model, and invoking the pre-stored delivery through the VRDS medical network model For each transfer function in the function set, the BMP data source is processed by multiple transfer functions in the transfer function set to obtain the first venous image data, and the transfer function set includes all presets set by a reverse editor. Describe the transfer function of the inferior vena cava.

Among them, BMP (full name Bitmap) is a standard image file format in the Windows operating system, which can be divided into two categories: device-dependent bitmap (DDB) and device-independent bitmap (DIB). The scan image includes any one of the following: CT image, MRI image, DTI image, PET-CT image.

The VRDS medical network model is a preset network model, and its training method includes the following three steps: image sampling and scale scaling; 3D convolutional neural network feature extraction and scoring; medical imaging device evaluation and network training. In the implementation process, first sampling will be required to obtain N BMP data sources, and then M BMP data sources will be extracted from N BMP data sources at a preset interval. It needs to be explained that the preset interval can be flexibly set according to the usage scenario. Sample M from N, then scale the sampled M BMP data sources to a fixed size (for example, the length is S pixels and the width is S pixels), and the resulting processing result is used as the input of the 3D convolutional neural network . In this way, M BMP data sources are used as the input of the 3D convolutional neural network. Specifically, a 3D convolutional neural network is used to perform 3D convolution processing on the BMP data source to obtain a feature map.

In this possible example, the generating second vein image data according to the first vein image data includes: importing the first vein image data into a preset cross blood vessel network model, and pass the cross blood vessel network model Perform spatial segmentation processing on the original data at the intersection to obtain mutually independent image data of multiple inferior vena cava at the intersection; update the original data set through the independent image data to obtain a second vein Image data.

In this possible example, the processing the second vein image data to obtain the target image of the inferior vena cava includes: performing at least one of the following processing operations on the second vein image data to obtain the inferior vena cava The target image: 2D boundary optimization processing, 3D boundary optimization processing, data enhancement processing.

Wherein, the 2D boundary optimization processing includes the following operations: multiple sampling to obtain low-resolution information and high-resolution information, where the low-resolution information can provide contextual semantic information of the segmented target in the entire image, that is, reflect the relationship between the target and the environment. Characteristics of the inter-relationship, the segmentation target includes the target vein. The 3D boundary optimization processing includes the following operations: putting the second medical image data into a 3D convolution layer to perform a 3D convolution operation to obtain a feature map; the 3D pooling layer compresses the feature map and performs Non-linear activation; cascade operation is performed on the compressed feature maps to obtain the prediction result image output by the model. The data enhancement processing includes at least one of the following: data enhancement based on arbitrary angle rotation, data enhancement based on histogram equalization, data enhancement based on white balance, data enhancement based on mirroring operations, data enhancement based on random cut, and data enhancement based on Data enhancement to simulate different lighting changes.

In a possible example, before acquiring the scanned image of the target part of the target user including the inferior vena cava, the method further includes: entering the original feature data set.

In addition, the method further includes: the medical imaging device displays the probability distribution of the abnormal category of the abnormal part of the inferior vena cava, assisting the doctor to make a quick diagnosis.

Consistent with the embodiment shown in FIG. 2a, please refer to FIG. 3. FIG. 3 is a schematic structural diagram of a medical imaging device 300 according to an embodiment of the present application. As shown in the figure, the medical imaging device 300 includes a processor 310, a memory 320, a communication interface 330, and one or more programs 321, wherein the one or more programs 321 are stored in the above-mentioned memory 320 and are configured to be executed by the above-mentioned processor 310, and the one or more The program 321 includes instructions for performing the following steps;

Obtaining a scanned image of the inferior vena cava containing the target user; and processing the scanned image to obtain a target image, the target image including three-dimensional spatial image data of the inferior vena cava; and for extracting reference features based on the target image Data set, the reference feature data set is used to reflect the physiological characteristics of the inferior vena cava of the target user; and used to determine the abnormal category of the inferior vena cava according to the reference feature data set; and used to output the The abnormal category of the inferior vena cava.

In a possible example, in terms of determining the abnormal category of the inferior vena cava according to the reference characteristic data set, the instructions in the program are specifically used to perform the following operations: obtaining an abnormal database, the abnormal database including Correspondence between the characteristic data of the inferior vena cava and the abnormal category of the inferior vena cava; and used to query the abnormal database with the reference characteristic data set as a query identifier, and obtain the abnormalities matching the reference characteristic data set category.

In this possible example, the characteristic data of the inferior vena cava corresponding to the repeated malformation of the inferior vena cava in the abnormal database includes bilateral superior main vein state analysis data, and the bilateral superior main vein state analysis data presents The histological defects include the preservation of bilateral superior main veins.

In a possible example, in terms of determining the abnormal category of the inferior vena cava according to the reference feature data set, the instructions in the program are specifically used to perform the following operations: obtaining a pre-trained abnormal category identification model; And for importing the reference feature data set into the abnormal category identification model to obtain an output result, the output result including the probability distribution of a single abnormal category or multiple abnormal categories.

In a possible example, in terms of extracting the reference feature data set based on the target image, the instructions in the program are specifically used to perform the following operations: obtain the information entered by the doctor for the target user through the disease entry interface Preliminary diagnosis result data of the inferior vena cava, the preliminary diagnosis result data including description information for the abnormal category of the inferior vena cava; and at least one category for determining the feature data to be extracted according to the preliminary diagnosis result data; And for extracting the feature data of the at least one category according to the target image.

In a possible example, in terms of extracting the reference feature data set according to the target image, the instructions in the program are specifically used to perform the following operations: obtain the inferior vena cava entered by the target user through the disease entry interface Condition description data; and used to import the condition description data into a pre-trained condition prediction model to obtain a condition prediction result, the condition prediction result including the abnormal category of the inferior vena cava; and used to predict according to the condition As a result, at least one category of feature data to be extracted is determined; and feature data for extracting the at least one category according to the target image.

In a possible example, in terms of processing the scanned image to obtain a target image, the instructions in the program are specifically used to perform the following operations: generating a bitmap BMP data source according to the scanned image; The BMP data source generates first venous image data, the first venous image data includes a raw data set of the inferior vena cava, the raw data set is the surface of the inferior vena cava and the tissue inside the inferior vena cava The transfer function result of the cubic space of the structure; and for generating second vein image data according to the first vein image data, the second vein image data including a segmentation data set of the inferior vena cava, the segmentation data set It includes mutually independent image data of the inferior vena cava having a cross position relationship; and is used to process the second venous image data to obtain a target image of the inferior vena cava.

The foregoing mainly introduces the solution of the embodiment of the present application from the perspective of the execution process on the method side. It can be understood that, in order to realize the above-mentioned functions, the medical imaging apparatus includes hardware structures and/or software modules corresponding to each function. Those skilled in the art should easily realize that in combination with the units and algorithm steps of the examples described in the embodiments provided herein, this application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software-driven hardware depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.

The embodiment of the present application may divide the medical imaging device into functional units according to the foregoing method examples. For example, each functional unit may be divided corresponding to each function, or two or more functions may be integrated into one processing unit. The above-mentioned integrated unit can be implemented in the form of hardware or software functional unit. It should be noted that the division of units in the embodiments of the present application is illustrative, and is only a logical function division, and there may be other division methods in actual implementation.

FIG. 4 is a block diagram of the functional unit composition of the medical imaging device 400 involved in an embodiment of the present application. The medical imaging device 400 is applied to a medical imaging device. The medical imaging device 400 includes a processing unit 401 and a communication unit 402, wherein,

The processing unit 401 is used to obtain a scanned image of the inferior vena cava of the target user through the communication unit 402; and used to process the scanned image to obtain a target image, the target image including the three-dimensional image of the inferior vena cava Spatial image data; and used to extract a reference feature data set based on the target image, the reference feature data set used to reflect the physiological characteristics of the inferior vena cava of the target user; and used to determine the reference feature data set based on the reference feature data set The abnormal category of the inferior vena cava; and the abnormal category used to output the inferior vena cava.

Wherein, the apparatus 400 may further include a storage unit 403, which is used to store program codes and data of the electronic device. The processing unit 401 may be a processor, the communication unit 402 may be a touch screen or a transceiver, and the storage unit 403 may be a memory.

In a possible example, in the aspect of determining the abnormal category of the inferior vena cava according to the reference feature data set, the processing unit 401 is specifically configured to: obtain an abnormal database, the abnormal database including the inferior vena cava Correspondence between the characteristic data and the abnormal category of the inferior vena cava; and used to query the abnormal database with the reference characteristic data set as a query identifier to obtain an abnormal category matching the reference characteristic data set.

In a possible example, in determining the abnormal category of the inferior vena cava according to the reference feature data set, the processing unit 401 is specifically configured to: obtain a pre-trained abnormal category identification model; and The reference feature data set is imported into the abnormal category identification model to obtain an output result, and the output result includes the probability distribution of a single abnormal category or multiple abnormal categories.

In a possible example, in terms of extracting the reference feature data set according to the target image, the processing unit 401 is specifically configured to: obtain the information entered by the doctor for the inferior vena cava of the target user through the condition entry interface Preliminary diagnosis result data, the preliminary diagnosis result data including description information for the abnormal category of the inferior vena cava; and at least one category for determining the feature data to be extracted according to the preliminary diagnosis result data; and The target image extracts feature data of the at least one category.

In a possible example, in terms of extracting the reference feature data set according to the target image, the processing unit 401 is specifically configured to: obtain the condition description data of the inferior vena cava entered by the target user through the condition entry interface And used to import the condition description data into a pre-trained condition prediction model to obtain a condition prediction result, the condition prediction result including the abnormal category of the inferior vena cava; and used to determine the condition to be extracted according to the condition prediction result At least one category of feature data; and feature data for extracting the at least one category according to the target image.

In a possible example, in terms of processing the scanned image to obtain the target image, the processing unit 401 is specifically configured to: generate a bitmap BMP data source according to the scanned image; and to generate a bitmap BMP data source according to the BMP data source Generate first venous image data, the first venous image data including a raw data set of the inferior vena cava, the raw data set being a cubic space of the surface of the inferior vena cava and the tissue structure inside the inferior vena cava And is used to generate second vein image data according to the first vein image data, the second vein image data includes a segmented data set of the inferior vena cava, the segmented data set includes a cross position Mutually independent image data of the related inferior vena cava; and used to process the second venous image data to obtain a target image of the inferior vena cava.

An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any method as recorded in the above method embodiment , The aforementioned computer includes a medical imaging device.

The embodiments of the present application also provide a computer program product. The above-mentioned computer program product includes a non-transitory computer-readable storage medium storing a computer program. The above-mentioned computer program is operable to cause a computer to execute any of the methods described in the above-mentioned method embodiments. Part or all of the steps of the method. The computer program product may be a software installation package, and the computer includes a medical imaging device.

It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that this application is not limited by the described sequence of actions. Because according to this application, some steps can be performed in other order or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

In the above-mentioned embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in an embodiment, reference may be made to related descriptions of other embodiments.

In the several embodiments provided in this application, it should be understood that the disclosed device may be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the above-mentioned units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or integrated. To another system, or some features can be ignored, or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical or other forms.

The units described above as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.

In addition, the functional units in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.

If the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable memory. Based on this understanding, the technical solution of the present application essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, A number of instructions are included to enable a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the foregoing methods of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other various media that can store program codes.

Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above-mentioned embodiments can be completed by a program instructing relevant hardware. The program can be stored in a computer-readable memory, and the memory can include: a flash disk , Read-only memory (English: Read-Only Memory, abbreviation: ROM), random access device (English: Random Access Memory, abbreviation: RAM), magnetic disk or optical disc, etc.

The embodiments of the application are described in detail above, and specific examples are used in this article to illustrate the principles and implementation of the application. The descriptions of the above examples are only used to help understand the methods and core ideas of the application; A person of ordinary skill in the art, based on the idea of the present application, will have changes in the specific implementation and the scope of application. In summary, the content of this specification should not be construed as a limitation of the present application.

Claims

An analysis method based on the virtual reality doctor system VRDS AI inferior vena cava image, characterized in that it is applied to a medical imaging device; the method includes:

Acquiring a scanned image of the inferior vena cava containing the target user;

Processing the scanned image to obtain a target image, the target image including three-dimensional spatial image data of the inferior vena cava;

Extracting a reference feature data set according to the target image, the reference feature data set being used to reflect the physiological characteristics of the inferior vena cava of the target user;

Determine the abnormal category of the inferior vena cava according to the reference feature data set;

Output the abnormal category of the inferior vena cava.
The method according to claim 1, wherein the determining the abnormal category of the inferior vena cava according to the reference characteristic data set comprises:

Acquiring an abnormality database, the abnormality database including the correspondence between the characteristic data of the inferior vena cava and the abnormal category of the inferior vena cava;

Using the reference feature data set as a query identifier, query the anomaly database to obtain an anomaly category matching the reference feature data set.
The method according to claim 2, wherein the abnormal category of the inferior vena cava includes any one of the following: absence of inferior vena cava, duplication of inferior vena cava, ectopic left side of inferior vena cava, inferior vena cava Continue to thoracic vein, inferior vena cava and posterior ureter, tumor involving tumor thrombus, tumor involving thrombus.
The method according to claim 3, wherein the characteristic data of the inferior vena cava corresponding to the absence of the inferior vena cava in the abnormal database comprises collateral distribution characteristic data, and the tissue presented by the collateral distribution characteristic data Structural defects include at least one of the following: venous insufficiency of the lower extremities, idiopathic deep vein thrombosis, and collateral circulation of the lumbar vein.
The method according to claim 3, wherein the characteristic data of the inferior vena cava corresponding to the duplication of the inferior vena cava malformation in the abnormal database includes bilateral superior main venous state analysis data, and the bilateral superior main vein The tissue structure defects presented by the venous status analysis data include the preservation of the bilateral superior main veins.
The method according to claim 3, wherein the characteristic data of the inferior vena cava corresponding to the thoracic vein in the abnormal database includes an analysis of the relationship between the superior renal inferior vena cava and the azygos vein or the hemi-azygos vein Data, and the tissue structure defects presented in the analysis data of the relationship between the superior renal inferior vena cava and azygos or semi-odd veins include at least one of the following: abnormal continuation of the superior renal inferior vena cava and azygos or semi-odd veins, The vena cava enters the atypical vein and flows back into the heart through the superior vena cava, or into the hemi-odd vein and then into the azygos vein, the hemi-odd vein directly drains into the coronary sinus through the permanent left superior vena cava or drains into the left through the accessory hemi-odd vein Lateral brachiocephalic vein.
The method according to claim 3, wherein the characteristic data of the inferior vena cava corresponding to the posterior inferior vena cava ureter in the abnormal database includes analysis data of the development of the inferior vena cava of the inferior kidney, and The tissue structure defects presented by the analysis data of the development of the vena cava include at least one of the following: the inferior vena cava of the inferior renal segment develops from the right posterior main vein instead of the right upper main vein.
The method according to claim 3, wherein the characteristic data of the inferior vena cava corresponding to tumor involvement and thrombus formation in the abnormal database comprises tumor analysis data.
The method according to any one of claims 1-8, wherein the determining the abnormal category of the inferior vena cava according to the reference characteristic data set comprises:

Obtain a pre-trained abnormal category identification model;

The reference feature data set is imported into the abnormal category identification model to obtain an output result, and the output result includes the probability distribution of a single abnormal category or multiple abnormal categories.
The method according to claim 9, wherein said extracting a reference feature data set according to the target image comprises:

Obtain the preliminary diagnosis result data of the inferior vena cava of the target user entered by the doctor through the condition entry interface, the preliminary diagnosis result data including the description information of the abnormal category of the inferior vena cava;

Determine at least one category of feature data to be extracted according to the preliminary diagnosis result data;

Extracting the feature data of the at least one category according to the target image.
The method according to claim 9, wherein said extracting a reference feature data set according to the target image comprises:

Acquiring the condition description data of the inferior vena cava entered by the target user through the condition entry interface;

Importing the condition description data into a pre-trained condition prediction model to obtain a condition prediction result, the condition prediction result including the abnormal category of the inferior vena cava;

Determine at least one category of feature data to be extracted according to the condition prediction result;

Extracting the feature data of the at least one category according to the target image.
The method according to claim 1, wherein said processing said scanned image to obtain a target image comprises:

Generating a bitmap BMP data source according to the scanned image;

First venous image data is generated according to the BMP data source, the first venous image data includes the original data set of the inferior vena cava, the original data set is the surface of the inferior vena cava and the interior of the inferior vena cava The result of the transfer function of the cube space of the organizational structure;

Generate second vein image data according to the first vein image data, the second vein image data includes a segmented data set of the inferior vena cava, and the segmented data set includes mutually independent inferior vena cava having a cross position relationship Image data;

The second venous image data is processed to obtain the target image of the inferior vena cava.
A medical imaging device is characterized in that it comprises a processing unit and a communication unit, wherein:

The processing unit is used to obtain a scanned image of the inferior vena cava including the target user through the communication unit; and used to process the scanned image to obtain a target image, the target image including the three-dimensional space image of the inferior vena cava Data; and used to extract a reference feature data set based on the target image, the reference feature data set used to reflect the physiological characteristics of the inferior vena cava of the target user; and used to determine the next feature data set based on the reference feature data set The abnormal category of the vena cava; and the abnormal category for outputting the inferior vena cava.
The device according to claim 13, wherein, in the aspect of determining the abnormal category of the inferior vena cava according to the reference characteristic data set, the processing unit is specifically configured to: obtain an abnormal database, the abnormal database Including the correspondence between the characteristic data of the inferior vena cava and the abnormal category of the inferior vena cava; and using the reference characteristic data set as a query identifier to query the abnormal database to obtain the abnormal category matching the reference characteristic data set .
The device of claim 14, wherein the abnormal category of the inferior vena cava includes any one of the following: absence of inferior vena cava, duplication of inferior vena cava, ectopic left side of inferior vena cava, inferior vena cava Continue to thoracic vein, inferior vena cava and posterior ureter, tumor involving tumor thrombus, tumor involving thrombus.
The device according to claim 15, wherein the characteristic data of the inferior vena cava corresponding to the absence of the inferior vena cava in the abnormal database comprises collateral distribution characteristic data, and the tissue presented by the collateral distribution characteristic data Structural defects include at least one of the following: venous insufficiency of the lower extremities, idiopathic deep vein thrombosis, and collateral circulation of the lumbar vein.
The device according to claim 15, wherein the characteristic data of the inferior vena cava corresponding to the inferior vena cava duplication malformation in the abnormal database includes bilateral superior main venous state analysis data, and the bilateral superior main vein The tissue structure defects presented by the venous status analysis data include the preservation of the bilateral superior main veins.
The device according to claim 15, wherein the characteristic data of the inferior vena cava corresponding to the thoracic vein in the abnormal database includes an analysis of the relationship between the superior renal inferior vena cava and azygos vein or semi-azygos vein Data, and the tissue structure defects presented in the analysis data of the relationship between the superior renal inferior vena cava and azygos or semi-odd veins include at least one of the following: abnormal continuation of the superior renal inferior vena cava and azygos or semi-odd veins, The vena cava enters the atypical vein and flows back into the heart through the superior vena cava, or into the hemi-odd vein and then into the azygos vein, the hemi-odd vein directly drains into the coronary sinus through the permanent left superior vena cava or drains into the left through the accessory hemi-odd vein Lateral brachiocephalic vein.
A medical imaging device is characterized by comprising a processor, a memory, a communication interface, and one or more programs, the one or more programs are stored in the memory and configured to be executed by the processor The program includes instructions for executing the steps in the method according to any one of claims 1-12.
A computer-readable storage medium, characterized by storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the method according to any one of claims 1-12.