CN113610841A

CN113610841A - Blood vessel abnormal image identification method and device, electronic equipment and storage medium

Info

Publication number: CN113610841A
Application number: CN202110991077.XA
Authority: CN
Inventors: 卢洁; 傅璠; 单艺
Original assignee: Xuanwu Hospital
Current assignee: Xuanwu Hospital
Priority date: 2021-08-26
Filing date: 2021-08-26
Publication date: 2021-11-05
Anticipated expiration: 2041-08-26
Also published as: CN113610841B

Abstract

The scheme discloses a blood vessel abnormal image identification method, a blood vessel abnormal image identification device, an electronic device and a storage medium, wherein the method comprises the following steps: marking voxel points in a blood vessel image to be identified to obtain voxel point marking information; determining a maximum communication domain in a blood vessel region based on the voxel point marking information; and identifying abnormal vessel images in the vessel images to be identified by using the maximum connected domain. According to the scheme, the complex blood vessel region in the blood vessel image is rapidly identified by using the maximum communication domain in the blood vessel region, and the blood vessel abnormal image is identified by screening the complex blood vessel region; by the method, the complex blood vessel region can be rapidly and accurately sorted from the normal blood vessel, and then the real attribute of the complex blood vessel is identified, so that the identification accuracy is improved, and the misdiagnosis rate and the missed diagnosis rate are reduced.

Description

Blood vessel abnormal image identification method and device, electronic equipment and storage medium

Technical Field

The invention relates to the technical field of medical image recognition. And more particularly, to an identification method, apparatus, electronic device, and computer-readable storage medium for identifying an abnormal blood vessel image in a physiological site having a complex blood vessel region.

Background

In the field of medical imaging, a variety of detection systems have made it possible to directly generate medical images for screening and evaluating medical conditions. Such as Computed Tomography (CT) imaging, Magnetic Resonance (MR) imaging, Positron Emission Tomography (PET), and so forth. These imaging methods allow visual identification of various lesions or abnormalities such as colon polyps, aneurysms, lung nodules, hardening of cardiac or arterial tissue, cancer microcalcifications or masses in breast tissue.

However, due to the complexity of human physiological tissues, the complexity of blood vessels at many positions is high, and the complicated blood vessel region and the abnormal blood vessel region are not easily distinguished and identified only by medical imaging, which often causes the problems of misdiagnosis and missed diagnosis.

Disclosure of Invention

The invention aims to provide an identification method, an identification device, an electronic device and a computer readable storage medium for identifying abnormal blood vessel images in a physiological part with a complex blood vessel region.

In order to achieve the purpose, the technical scheme is as follows:

in a first aspect, the present disclosure provides a blood vessel abnormality image recognition method, including:

marking voxel points in a blood vessel image to be identified to obtain voxel point marking information;

determining a maximum communication domain in a blood vessel region based on the voxel point marking information;

and identifying abnormal vessel images in the vessel images to be identified by using the maximum connected domain.

In a preferred embodiment, the step of labeling the voxel points in the blood vessel image to be identified, and the step of obtaining the voxel point labeling information includes:

and marking voxel points in different blood vessel regions in the blood vessel image to be identified by adopting different types of labels to obtain voxel point marking information.

In a preferred embodiment, the step of determining a maximum communication domain in the blood vessel region based on the voxel point labeling information comprises:

counting the number of voxel points in different blood vessel areas according to the voxel point marking information;

taking the blood vessel area with the number of voxel points not lower than a first preset threshold value as a maximum communication area; alternatively, the first and second electrodes may be,

and sorting the blood vessel regions according to the number of the voxel points, and taking the blood vessel regions sorted before the preset ranking as the maximum communication domain.

In a preferred embodiment, the step of identifying an abnormal blood vessel image in the blood vessel images to be identified by using the maximum connected domain includes:

and respectively removing the venous vessel region and the main vessel region in the maximum communication region to obtain a first complex vessel region.

In a preferred embodiment, the step of identifying an abnormal blood vessel image in the blood vessel image to be identified by using the maximum connected domain includes:

and taking the blood vessel region of the non-maximum communication region as a second complex blood vessel region.

In a preferred embodiment, the step of removing the venous vessel region and the main vessel region in the maximum communication region, respectively, to obtain the first complex vessel region comprises:

recognizing a vein blood vessel region in the maximum communication region by using a vein blood vessel model, and removing the vein blood vessel region;

and matching the main blood vessel template to the maximum communication domain based on the blood vessel key points, and removing the main blood vessel region on the maximum communication domain to obtain a first complex blood vessel region.

judging whether the first complex blood vessel region and/or the second complex blood vessel region are abnormal blood vessel images or not;

if the number of the voxel points in the complex blood vessel region is larger than a second preset threshold value, determining the complex blood vessel region as an abnormal blood vessel image;

and if the number of the voxel points in the complex blood vessel region is not larger than a second preset threshold, determining that the complex blood vessel region is a non-abnormal blood vessel image.

In a second aspect, the present solution provides an image recognition apparatus, including:

the acquisition module is used for determining a region with a surface vector field in the blood vessel image to be identified;

and the identification module is used for determining a raised image on the blood vessel according to the identification characteristics of the surface vector field.

In a third aspect, the present solution provides a computer storage medium having stored thereon a computer program which, when executed by a processor, implements the method as described above.

In a fourth aspect, the present solution provides a computing device comprising: a processor; and a memory for storing executable instructions of the processor;

wherein the processor is configured to perform the method as described above via execution of the executable instructions.

The invention has the following beneficial effects:

according to the scheme, the complex blood vessel region in the blood vessel image is rapidly identified by using the maximum communication domain in the blood vessel region, and the blood vessel abnormal image is identified by screening the complex blood vessel region; by the method, the complex blood vessel region can be rapidly and accurately sorted from the normal blood vessel, and then the real attribute of the complex blood vessel is identified, so that the identification accuracy is improved, and the misdiagnosis rate and the missed diagnosis rate are reduced.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 shows a schematic diagram of a blood vessel abnormality image identification method according to the scheme;

fig. 2 is a schematic diagram showing an example of the complex blood vessel region in the present scheme in the region of the main blood vessel region and the venous blood vessel region;

FIG. 3 is a schematic diagram showing an example of an isolated or abnormal complex blood vessel according to the present embodiment;

fig. 4 shows a schematic diagram of an image recognition apparatus according to the present solution;

fig. 5 shows a schematic diagram of an electronic device according to the present solution.

Detailed Description

In order to make the technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings. It is clear that the described embodiments are only a part of the embodiments of the present application, and not an exhaustive list of all embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.

In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be considered as limiting the present invention. Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.

In this application, the word "exemplary" is used to mean "serving as an example, instance, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the invention. In the following description, details are set forth for the purpose of explanation. It will be apparent to one of ordinary skill in the art that the present invention may be practiced without these specific details. In other instances, well-known structures and processes are not shown in detail to avoid obscuring the description of the invention with unnecessary detail. Thus, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

It should be noted that, since the method in the embodiment of the present application is executed in the computing device, the processing objects of each computing device exist in the form of data or information, for example, time, which is substantially time information, it can be understood that, in the subsequent embodiments, if the size, the number, the position, and the like are mentioned, corresponding data exist, so that the electronic device performs processing, and details are not described herein.

In a typical configuration of the present application, a terminal or a trusted party, etc. includes one or more processors, such as a Central Processing Unit (CPU), an input/output interface, a network interface, and a memory. The Memory may include forms of volatile Memory, Random Access Memory (RAM), and/or non-volatile Memory in a computer-readable medium, such as Read Only Memory (ROM) or Flash Memory. Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, Phase-Change Memory (PCM), Programmable Random Access Memory (PRAM), Static Random-Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), electrically Erasable Programmable Read-Only Memory (EEPROM), flash Memory or other Memory technology, Compact Disc Read-Only Memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device.

The device referred to in this application includes, but is not limited to, a user device, a network device, or a device formed by integrating a user device and a network device through a network. The user equipment includes, but is not limited to, any mobile electronic product, such as a smart phone, a tablet computer, etc., capable of performing human-computer interaction with a user (e.g., human-computer interaction through a touch panel), and the mobile electronic product may employ any operating system, such as an Android operating system, an iOS operating system, etc. The network Device includes an electronic Device capable of automatically performing numerical calculation and information processing according to a preset or stored instruction, and the hardware includes, but is not limited to, a microprocessor, an Application Specific Integrated Circuit (ASIC), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), an embedded Device, and the like. The network device includes but is not limited to a computer, a network host, a single network server, a plurality of network server sets or a cloud of a plurality of servers; here, the Cloud is composed of a large number of computers or web servers based on Cloud Computing (Cloud Computing), which is a kind of distributed Computing, one virtual supercomputer consisting of a collection of loosely coupled computers. Including, but not limited to, the internet, a wide area network, a metropolitan area network, a local area network, a VPN network, a wireless Ad Hoc network (Ad Hoc network), etc. Preferably, the device may also be a program running on the user device, the network device, or a device formed by integrating the user device and the network device, the touch terminal, or the network device and the touch terminal through a network.

Of course, those skilled in the art will appreciate that the foregoing is by way of example only, and that other existing or future devices, which may be suitable for use in the present application, are also encompassed within the scope of the present application and are hereby incorporated by reference.

The application provides an image recognition method which is mainly applied to computing equipment and used for detecting an image of a raised area in a blood vessel image. The scheme is used in the field of medical image processing, and is used for diagnosing the coronary artery, the head and the neck of a patient by means of electronic equipment such as a computer when a doctor diagnoses vascular aneurysms, providing an accurate vascular diagnosis report and helping the doctor to make a reasonable vascular diagnosis result.

The computing device may include a camera module for acquiring a CT (Computed Tomography) image of the head and neck region of a target user, such as a CT scanner, and performs a cross-sectional scan around a certain part of a human body together with a detector having a very high sensitivity one by one, mainly using a precisely collimated X-ray beam, gamma rays, ultrasonic waves, and the like.

The computing device comprises a communication module, establishes communication connection with network equipment or other equipment (such as external camera equipment) and the like, and sends and receives information, and the like, for example, the computing device uploads a scanning result to the network equipment so that other user equipment can inquire a diagnosis result of a CT image and the like through the network equipment, or downloads CT image information of a target user through the network equipment; for example, the computing device does not currently include a camera device, establishes a communication connection with an external camera device through a communication connection, and receives CT image information and the like about the head and neck region of the target user, which are sent by the external camera device. The external Imaging device includes, but is not limited to, the acquisition device 12 may include a Magnetic Resonance Imaging (MRI), a computed tomography device, and the like.

The computing device includes a data processing module for collecting, storing, retrieving, processing, transforming, and transmitting data, such as identifying corresponding lesion regions via CT images.

Computing devices include, but are not limited to, user devices and network devices, where user devices include, but are not limited to, any mobile electronic product that can interact with a user (e.g., via a touch pad), such as a smart phone, a tablet computer, or a medical device; network devices include, but are not limited to, computers, network hosts, a single network server, multiple sets of network servers, or a cloud of multiple servers.

Through analysis and research on the prior art, many parts of a human body, such as the cranium, the neck, the heart and the like, usually have dense blood vessels, and if abnormal blood vessel regions exist in the dense blood vessel regions, the dense blood vessel regions are difficult to distinguish from normal blood vessels. In the prior art, a neural network is usually adopted to extract blood vessels with complete complex physiological tissue structures such as intracranial structures, neck structures, hearts and the like, and convenience is provided for clinical lesion identification. However, when a certain area with dense and complex blood vessels is encountered and vascular lesions appear, the problem that the blood vessel images extracted by the neural network are not complete enough, and false identification or missing identification of abnormal areas of the blood vessels is likely to be caused may occur.

Therefore, the scheme aims to provide the identification method for identifying the abnormal blood vessel image in the physiological part with the complex blood vessel region, the method screens out the complex blood vessel regions with the types of free blood vessels, adhesion blood vessels, pathological changes and the like by utilizing the maximum communication region, then performs abnormal identification on the complex blood vessel regions, and determines the abnormal blood vessel image in the blood vessel image to be identified, so that the identification accuracy is improved, and the misdiagnosis rate and the missed diagnosis rate are reduced.

Hereinafter, a blood vessel abnormality image recognition method proposed by the present scheme will be described in detail with reference to the accompanying drawings.

As shown in fig. 1, the blood vessel abnormal image identification method according to the present scheme may be used for identifying abnormal blood vessel images in physiological tissue structures such as intracranial structures, neck structures, hearts and the like having dense blood vessel regions, and specifically includes:

step S1, marking voxel points in the blood vessel image to be identified to obtain voxel point marking information;

step S2, determining the maximum communication domain in the blood vessel region based on the voxel point marking information;

and step S3, identifying abnormal blood vessel images in the blood vessel images to be identified by using the maximum communication domain.

In the scheme, the initial medical image of the human physiological tissue containing the blood vessel image can be acquired by imaging systems such as Computed Tomography (CT), spiral CT, X-ray, Positron Emission Tomography (PET), 4D ultrasound, Magnetic Resonance (MR) and the like. And then, carrying out blood vessel segmentation processing on the initial medical image by utilizing deep learning neural network models such as a U-net network, a V-net network and the like to obtain a blood vessel image to be identified.

Because of the complexity of human body tissues, such as intracranial, cervical, cardiac, pulmonary, and hepatic tissues, the shapes of blood vessels associated with these human body tissues present a singular shape or arrangement. Therefore, the matched blood vessel segmentation model can be trained correspondingly aiming at different physiological tissues so as to more accurately separate the blood vessel images in the corresponding human physiological tissues and provide more accurate image basis for the subsequent blood vessel image processing.

In addition, in order to reduce the processing time of subsequent image recognition and reduce the dependence on manual work, the method can also utilize a binary classification segmentation method, a mask segmentation method and other modes to separate human tissues such as bones, lungs, hearts, livers and the like in the initial image from blood vessels to form a blood vessel image to be recognized, so as to avoid the interference of the human tissues except the blood vessels on the blood vessel image processing process.

In step S1, since there may be a plurality of types of blood vessels, such as free blood vessels, venous blood vessels, arterial blood vessels, and adhesion blood vessels, due to different types of blood vessels, the blood vessel regions in the blood vessel image to be recognized are not necessarily all connected. By the characteristic, the blood vessel images of different areas can be divided firstly. Specifically, different types of labels can be adopted to mark voxel points in different blood vessel regions in the blood vessel image to be identified, so as to obtain voxel point marking information, and the different blood vessel regions are divided by using the voxel point marking information.

In one example, the blood vessel image to be identified in this example may be an image that has been subjected to segmentation processing and contains only a blood vessel region. The voxel points in the vessel region in the vessel image to be identified are labeled numerically, so that each voxel point in the vessel region has labeling information. Wherein the voxel points in the continuous region are marked with a number. Different vessel regions are demarcated by regions represented by different numbers.

It is noted here that the label used to label the blood vessel can be in many different forms, for example, a number, letter, color, symbol, shade, etc. that can form a visually distinguishable label can be used as a label for the voxel point. In the same image to be identified, one or more labels can be adopted to mark voxel points of different blood vessel regions, so as to divide the different blood vessel regions.

In general, in the process of segmenting a blood vessel image, due to the problems of accuracy of a segmentation model, definition of image information and the like, some problems of discontinuous blood vessel regions, unclear boundaries, incomplete blood vessels and the like may occur, and particularly, complex blood vessel regions are easy to segment and are unclear. In this case, the main blood vessel region can be used as the maximum communication region, and some special types of blood vessels can be identified through the maximum communication region. In step S2, the maximum connected domain in the blood vessel region is determined using the voxel point labeling information. By the method, the blood vessel of the free type can be quickly identified, and the complexity of subsequent image processing is reduced.

In particular, the number of voxel points may be counted based on the voxel point labeling information. And if the number of voxel points in a certain blood vessel region is not lower than a first preset threshold value in the statistical result, determining the region as a maximum communication region.

In one example, the image is an intracranial blood vessel image, and the blood vessel image in this region has higher complexity and more blood vessels, so the first preset threshold value can be set to 5000. Counting the number of voxel points in different blood vessel regions, and if the number of voxel points in a certain region is not lower than 5000, determining the region as the maximum communication region. The type and number of vessels included may vary depending on the image acquisition region, and may include many arterial vessels. Therefore, the maximum communication areas can be multiple, and convenience is provided for judging whether vascular lesions exist in different positions at the same time subsequently through the multiple maximum communication areas.

In the scheme, a sequencing mode can be utilized to determine the plurality of maximum communication areas at one time. Specifically, the number of voxel points in the vessel region is counted and ordered. The blood vessel regions located in the first few positions are taken as the maximum communication region.

In one example, the image is a pulmonary vessel image, the number of voxel points in the vessel region is counted, and the vessel region is sorted according to the number of voxel points. According to the characteristics of the pulmonary vessel image, the rank of the preset rank can be set to be 7, namely, the vessel regions in the top 7 (including the 7 th) are all determined as the maximum communication region. Therefore, a plurality of maximum communication domains can be screened out at one time, and the identification speed is effectively improved.

Along with the different complexities of different physiological tissues of human bodies, the complexities of blood vessels also can be greatly different. Especially, when the arteriovenous box is close or the adjacent vessels have adhesion problems, the real boundary of the vessels is difficult to distinguish, and the area where the blood vessel abnormality exists can not be identified. Therefore, in step S3, the maximum communication domain and the recognition model of a special type or the template of the important blood vessel can be used to identify the complex blood vessel region, and then it is determined whether there is a real abnormal blood vessel image in the blood vessel image to be identified.

Specifically, the vein vessel region and the main vessel region in each maximum connected region are relatively obvious vessel regions, and the vein vessel region and the main vessel region in the maximum connected region can be removed by using the characteristics of the obvious vessel regions, so as to obtain the first complex vessel region.

In one embodiment, due to the complexity of the physiological tissues, the complex vessel region may be located in the vicinity of a main vessel region containing multiple vessel branches and a venous vessel region. As shown in fig. 1, the main vessel region includes the aorta, the primary vessel branch, and the secondary vessel branch. In the image to be identified, the venous vessel region may intersect with a significant vessel branch, a normal capillary vessel, or a vessel in which an abnormality exists. In order to better split the real complex blood vessel region, the vein blood vessel region and the main blood vessel region need to be removed. For example, the blood vessel image to be identified is an intracranial blood vessel image. At the moment, the vein vessel region in the maximum communication domain is identified by using a vein vessel identification model which is trained by a deep learning neural network in advance, and the region is removed from the maximum communication domain after the vein vessel region is determined. Then, according to the condition of the intracranial blood vessel image, determining a main blood vessel template corresponding to the blood vessel image to be identified (the main blood vessel template is a blood vessel pattern or model constructed through medical experience), and determining a blood vessel key point on the main blood vessel template. Matching the key points of the blood vessels into the maximum communication domain, matching the main blood vessel template onto the maximum communication domain, determining a main blood vessel region, and removing the main blood vessel region from the maximum communication domain, thereby obtaining a first complex blood vessel region. The first complex vascular region may be a blood vessel region that is not easily recognized, such as a normal adhesion blood vessel or a lesion blood vessel in which an abnormality exists.

For matching of key points of blood vessels in this example, two ways can be adopted:

the first mode is as follows: and obtaining a key point recognition model after secondary training by using the blood vessel key points and intracranial blood vessel images as input and utilizing a deep learning neural network. And identifying the positions of the key points of the blood vessels in the maximum connected domain by using the key point identification model. And then, taking the key points of the blood vessels as a reference, and enabling the matching value of the main blood vessel template to be on the maximum communication domain.

The second mode is as follows: and extracting a center line point set of the blood vessel in the image to be identified by utilizing a model-collecting algorithm or a skeleton algorithm, traversing the blood vessel center line point set according to the position of the blood vessel key point, and marking the blood vessel center line point as the blood vessel key point if the position error of a certain blood vessel center line point in the point set and the blood vessel key point is less than a preset threshold range. And then, taking the key points of the blood vessels as a reference, and enabling the matching value of the main blood vessel template to be on the maximum communication domain.

It should be noted here that, for the training processes of the blood vessel key point recognition model and the vein blood vessel recognition model, a neural network training method commonly used in the art is adopted, and details are not described here.

As can be seen from the above analysis, in the image segmentation process, a complicated blood vessel region is difficult to segment, and is prone to have problems such as unclear boundaries and incomplete segmentation, and therefore, a certain region independent of the maximum communication region may appear in the image, and the region may be similar to the region of the free blood vessel. To prevent the problem of missed detection, a blood vessel region other than the maximum communication region may be used as the second complex blood vessel region. As shown in fig. 2, the second complex blood vessel region may be an isolated blood vessel or a blood vessel region that is not easily recognized, such as a lesion blood vessel having an abnormality.

After the complex blood vessel region is determined by utilizing the maximum communication region, the complex blood vessel region can be identified by utilizing the anomaly identification model, so that whether a real blood vessel anomaly image exists in the blood vessel image to be identified or not is determined. Specifically, whether the first complex blood vessel region and/or the second complex blood vessel region is an abnormal blood vessel image is judged; if the number of the voxel points in the complex blood vessel region is larger than a second preset threshold value, determining the complex blood vessel region as an abnormal blood vessel image; and if the number of the voxel points in the complex blood vessel region is not larger than a second preset threshold, determining that the complex blood vessel region is a non-abnormal blood vessel image.

In one example, the first complex blood vessel region has a greater influence on the patient, and therefore, the determination of whether there is a real abnormal blood vessel in the first complex blood vessel region can be preferentially performed. For example, if the blood vessel image to be identified is a blood vessel image of a liver region, the second preset threshold may be set to 800. Counting the number of voxel points in each first complex blood vessel region; if the number of voxel points in a first complex blood vessel region is more than 800; the first complex blood vessel region is determined to be an abnormal blood vessel image. If not, determining that the first complex blood vessel region is a normal blood vessel image. Such as disordered capillaries, etc.

In one example, any kind of complex vascular region may be harmful to the patient for some important physiological tissues, for example, the vascular image to be identified is an intracranial vascular image. In this case, in order to prevent the risk of delaying the disease state due to missed detection, it is necessary to determine whether or not there is an abnormal blood vessel image in both the first complex blood vessel region and the second complex blood vessel region. At this time, the second preset threshold may be set to 500. And counting the number of voxel points in each first complex blood vessel region and each second complex blood vessel region. If the number of voxel points in a certain complex blood vessel area is more than 500; the complex blood vessel region is determined to be an abnormal blood vessel image. If not, determining that the complex blood vessel region is a normal blood vessel image. Such as disordered capillaries, etc.

The method of identifying the abnormal blood vessel image may be determined by using a method such as an area of the abnormal blood vessel region. The above method for identifying abnormal blood vessel images by using the number of voxel points is only an example for clearly illustrating the present invention, and is not a limitation to the embodiments of the present invention, and it will be obvious to those skilled in the art that other variations or modifications may be made on the basis of the above description, and all embodiments cannot be exhaustive.

In conclusion, the scheme utilizes the maximum communication domain in the blood vessel region to quickly identify the complex blood vessel region in the blood vessel image, and identifies the blood vessel abnormal image by screening the complex blood vessel region; by the method, the complex blood vessel region can be rapidly and accurately sorted from the normal blood vessel, and then the real attribute of the complex blood vessel is identified, so that the identification accuracy is improved, and the misdiagnosis rate and the missed diagnosis rate are reduced.

As shown in fig. 4, the present embodiment further provides an image recognition apparatus 101 implemented in cooperation with the above blood vessel abnormality image recognition method, the apparatus including: a marking module 102, a determination module 103 and a recognition module 104.

In order to shorten the processing time of the identification device 101, the acquired initial image may be segmented in advance to obtain the blood vessel image to be identified. When the device works, firstly, a to-be-identified blood vessel image is acquired, voxel points in the to-be-identified blood vessel image are marked by using an acquisition module 102, and voxel point marking information is acquired; then, determining the maximum communication domain in the blood vessel region by using the identification module 104 based on the voxel point marking information; finally, the identification module 104 is adopted to identify the abnormal blood vessel image in the blood vessel image to be identified by utilizing the maximum communication domain.

The device can be also provided with a calling module, and when vein region identification and blood vessel key point identification are required, the identification model is called from the model library, so that the time for the device to construct the model by itself is saved, and the identification speed of the image is accelerated.

In the scheme, when the marking model 102 is used for marking the voxel points, different types of labels are needed to be adopted to mark the voxel points in different blood vessel regions in the blood vessel image to be identified, so as to obtain voxel point marking information. In the scheme, the blood vessel image to be identified is an image subjected to blood vessel region segmentation, so that different blood vessel regions can be preliminarily identified according to the continuity of the voxel points, and meanwhile, the voxel points with continuity can be marked by the same type of label according to the continuity of the voxel points.

The specific process of determining the maximum communication domain in the blood vessel region by the obtaining module 103 is as follows: counting the number of voxel points in different blood vessel areas according to the voxel point marking information; taking the blood vessel area with the number of voxel points not lower than a first preset threshold value as a maximum communication area; or sorting the blood vessel regions according to the number of the voxel points, and taking the blood vessel regions sorted before the preset ranking as the maximum communication domain.

The identification module 104 in the device identifies the complex blood vessel region based on the maximum communication domain by combining the vein blood vessel identification model and the main blood vessel template. Specifically, a vein blood vessel model can be used for identifying a vein blood vessel region in the maximum communication domain and rejecting the vein blood vessel region; and matching the main blood vessel template to the maximum communication domain based on the blood vessel key points, and removing the main blood vessel region on the maximum communication domain to obtain a first complex blood vessel region. And then the blood vessel region of the non-maximum communication region is taken as a second complex blood vessel region. In this way, all the complex blood vessel regions with different types are identified, and then, whether a real abnormal blood vessel image exists in the complex blood vessel image is further judged by using the abnormal recognition model. The identification process of the abnormal identification model comprises the following steps: judging whether the first complex blood vessel region and/or the second complex blood vessel region are abnormal blood vessel images or not; if the number of the voxel points in the complex blood vessel region is larger than a second preset threshold value, determining the complex blood vessel region as an abnormal blood vessel image; and if the number of the voxel points in the complex blood vessel region is not larger than a second preset threshold, determining that the complex blood vessel region is a non-abnormal blood vessel image.

It should be understood that the various modules or units in the present solution may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, a discrete logic circuit having a logic Gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic Gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like is used.

On the basis of the above-mentioned blood vessel abnormality image recognition method, the present disclosure further provides a computer-readable storage medium. The computer-readable storage medium is a program product for implementing the above-described data acquisition method, which may employ a portable compact disc read only memory (CD-ROM) and include program code, and may be run on a device, such as a personal computer. However, the program product of the present solution is not limited thereto, and in this document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

A computer readable storage medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user computing device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).

On the basis of the above abnormal blood vessel image identification method, the present solution further provides an electronic device. The electronic device shown in fig. 5 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.

As shown in fig. 5, the electronic device 201 is in the form of a general purpose computing device. The components of the electronic device 201 may include, but are not limited to: at least one memory unit 202, at least one processing unit 203, a display unit 204 and a bus 205 for connecting different system components.

Wherein the storage unit 202 stores program codes executable by the processing unit 203, so that the processing unit 203 executes the steps of various exemplary embodiments described in the above-mentioned apparatus symptom information acquisition method. For example, the processing unit 203 may perform the steps as shown in fig. 1.

The memory unit 202 may include volatile memory units such as a random access memory unit (RAM) and/or a cache memory unit, and may further include a read only memory unit (ROM).

The storage unit 202 may also include programs/utilities with program modules including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.

The bus 205 may include a data bus, an address bus, and a control bus.

The electronic device 201 may also communicate with one or more external devices 207 (e.g., keyboard, pointing device, bluetooth device, etc.), which may be through an input/output (I/O) interface 206. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the electronic device 201, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

It should be understood that the above-mentioned embodiments of the present invention are only examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention, and it will be obvious to those skilled in the art that other variations or modifications may be made on the basis of the above description, and all embodiments may not be exhaustive, and all obvious variations or modifications may be included within the scope of the present invention.

Claims

1. A blood vessel abnormal image identification method is characterized by comprising the following steps:

2. The blood vessel abnormality image recognition method according to claim 1, wherein the step of labeling voxel points in the blood vessel image to be recognized and obtaining voxel point labeling information includes:

3. The blood vessel abnormality image recognition method according to claim 1, wherein the step of determining a maximum connected domain in a blood vessel region based on the voxel point labeling information includes:

4. The blood vessel abnormality image recognition method according to claim 1, wherein the step of recognizing an abnormal blood vessel image in the blood vessel image to be recognized by using the maximum connected domain includes:

5. The blood vessel abnormality image recognition method according to claim 4, wherein the step of recognizing the abnormal blood vessel image in the blood vessel image to be recognized by using the maximum connected domain includes:

6. The blood vessel abnormality image recognition method according to claim 4, wherein the step of removing the venous blood vessel region and the main blood vessel region in the maximum connected region respectively to obtain a first complex blood vessel region comprises:

7. The blood vessel abnormal image identification method according to claim 4 or 5, wherein the step of identifying an abnormal blood vessel image in the blood vessel image to be identified by using the maximum connected domain comprises:

8. An image recognition apparatus for blood vessel abnormality, comprising:

the marking module is used for marking the voxel points in the blood vessel image to be identified to obtain voxel point marking information;

a determining module for determining a maximum communication domain in the blood vessel region based on the voxel point marking information;

and the identification module is used for identifying the abnormal blood vessel image in the blood vessel image to be identified by utilizing the maximum communication domain.

9. A computer storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 7.

10. A computing device, comprising: a processor; and a memory for storing executable instructions of the processor;

wherein the processor is configured to perform the method of any of claims 1 to 7 via execution of the executable instructions.