CN113538463A

CN113538463A - Aneurysm segmentation method, device and equipment

Info

Publication number: CN113538463A
Application number: CN202110832295.9A
Authority: CN
Inventors: 金海岚; 王文智; 印胤; 杨光明
Original assignee: Union Strong Beijing Technology Co ltd
Current assignee: Union Strong Beijing Technology Co ltd
Priority date: 2021-07-22
Filing date: 2021-07-22
Publication date: 2021-10-22

Abstract

The embodiment of the specification discloses an aneurysm segmentation method, device and equipment, and belongs to the technical field of medical images and computers. The method comprises the following steps: acquiring image data to be processed; inputting the image data to be processed into an aneurysm segmentation model, segmenting blood vessels and aneurysms in the image data to be processed to obtain blood vessel segmentation results and aneurysm segmentation results, wherein the aneurysm segmentation model takes points on the aneurysms as centers, selects image blocks with different sizes, and trains a model obtained in advance through a convolutional neural network. By adopting the method provided by the embodiment of the specification, the problems of hollowing, blood vessel defect, aneurysm defect and the like can be reduced, the aneurysm segmentation precision and the aneurysm segmentation speed are improved, and a better aneurysm segmentation effect is obtained.

Description

Aneurysm segmentation method, device and equipment

Technical Field

The present disclosure relates to the field of medical imaging and computer technologies, and in particular, to a method, an apparatus, and a device for segmenting an aneurysm.

Background

Cerebrovascular angiography has become an important clinical diagnostic method, and plays an irreplaceable role particularly in interventional therapy. Cerebrovascular angiography plays an important role in diagnosis and treatment of cerebrovascular diseases, and three commonly used methods of cerebrovascular angiography include: cerebrovascular Angiography performed by nuclear Magnetic Resonance (MRA), cerebrovascular Angiography performed by CT (CTA) and cerebrovascular Angiography performed by Digital Subtraction Angiography (DSA). The segmented cerebral artery blood vessel can be used for auxiliary diagnosis of cerebral artery blood vessel pathological changes and mining of more cerebral artery blood vessel geometric morphology information, is beneficial to quantitative description of the cerebral artery blood vessel pathological changes, and provides more possibility for the application of imaging in clinic. Therefore, the method for segmenting the aneurysm based on the cerebrovascular image data has important significance.

In the prior art, the problems of hollow-out, blood vessel defect, aneurysm defect and the like are easy to occur in the aneurysm segmentation based on the cerebrovascular image data, the problem of low aneurysm segmentation precision exists, the aneurysm segmentation speed is low, and the aneurysm segmentation effect is poor.

Therefore, a new aneurysm segmentation method is needed, which can reduce the problems of hollow, vascular defects, aneurysm defects, and the like, improve the aneurysm segmentation accuracy and the aneurysm segmentation speed, and obtain a good aneurysm segmentation effect.

Disclosure of Invention

The embodiment of the specification provides an aneurysm segmentation method, device and equipment, which are used for solving the following technical problems: the existing aneurysm segmentation technology is easy to have the problems of hollowing, blood vessel defect, aneurysm defect and the like, and has the problems of low aneurysm segmentation precision, low aneurysm segmentation speed and poor aneurysm segmentation effect.

In order to solve the above technical problem, the embodiments of the present specification are implemented as follows:

an embodiment of the present disclosure provides an aneurysm segmentation method, including:

acquiring image data to be processed;

inputting the image data to be processed into an aneurysm segmentation model, segmenting blood vessels and aneurysms in the image data to be processed to obtain blood vessel segmentation results and aneurysm segmentation results, wherein the aneurysm segmentation model takes points on the aneurysms as centers, selects image blocks with different sizes, and trains a model obtained in advance through a convolutional neural network.

Further, the aneurysm segmentation model is a model obtained in advance by selecting image blocks of different sizes and training a convolutional neural network with points on the aneurysm as a center, and specifically comprises:

preprocessing training sample data to obtain preprocessed image data;

labeling the preprocessed image data to obtain first labeled image data;

and randomly selecting three image blocks with different sizes from the first marked image data, and performing model training by adopting a fine adjustment method to obtain an aneurysm segmentation model.

Further, the labeling the preprocessed image data to obtain first labeled image data specifically includes:

labeling the preprocessed image data to obtain second labeled image data;

and post-processing the second marked image data to obtain first marked image data.

Further, post-processing the second labeled image data to obtain first labeled image data specifically includes:

and repairing the second marked image data to obtain first marked image data, wherein the repairing comprises the following steps: repairing unreasonable segmented data points of blood vessels and aneurysms in the second marked image data;

and/or

Filling data points of the vascular defect, the aneurysm defect and the vascular disconnection in the second labeled image data;

and/or

And filling the data points with hollow blood vessels in the second labeled image data.

Further, the randomly selecting three image blocks with different sizes from the first labeled image data, and performing model training by using a fine tuning method to obtain an aneurysm segmentation model specifically includes:

randomly acquiring an aneurysm point from the preprocessed image data as a center, and randomly selecting a first image block on the first marked image data;

inputting the first image block into a 3D full convolution neural network model for training to obtain a first aneurysm segmentation model;

inputting the second image block into a first aneurysm segmentation model, and finely adjusting the first aneurysm segmentation model to obtain a second aneurysm segmentation model;

and inputting the third image block into a second aneurysm segmentation model, and finely adjusting the second aneurysm segmentation model to obtain the aneurysm segmentation model.

Embodiments of the present specification also provide an aneurysm segmentation device, including:

the acquisition module acquires image data to be processed;

and the segmentation module is used for inputting the image data to be processed into an aneurysm segmentation model, segmenting blood vessels and aneurysms in the image data to be processed and obtaining blood vessel segmentation results and aneurysm segmentation results, wherein the aneurysm segmentation model takes points on the aneurysm as a center, selects image blocks with different sizes and trains a model obtained in advance through a convolutional neural network.

preprocessing training sample data to obtain preprocessed image data;

labeling the preprocessed image data to obtain first labeled image data;

labeling the preprocessed image data to obtain second labeled image data;

and/or

An embodiment of the present specification further provides an electronic device, including:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to:

acquiring image data to be processed;

In the method for segmenting the aneurysm provided by the embodiment of the present specification, to-be-processed image data is acquired; inputting the image data to be processed into an aneurysm segmentation model, segmenting blood vessels and aneurysms in the image data to be processed to obtain blood vessel segmentation results and aneurysm segmentation results, wherein the aneurysm segmentation model takes points on the aneurysms as a center, selects image blocks with different sizes, trains a model obtained in advance through a convolutional neural network, can reduce the problems of hollowing, blood vessel defect, aneurysm defect and the like, improves the aneurysm segmentation precision and the aneurysm segmentation speed, and obtains a better aneurysm segmentation effect.

Drawings

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

Fig. 1 is a schematic diagram of an aneurysm segmentation method provided in an embodiment of the present disclosure;

fig. 2 is a flowchart of training an aneurysm segmentation model provided in an embodiment of the present specification;

fig. 3 is a schematic diagram illustrating an aneurysm segmentation effect provided by an embodiment of the present disclosure;

fig. 4 is a schematic view of an aneurysm segmentation device provided in an embodiment of the present disclosure.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any inventive step based on the embodiments of the present disclosure, shall fall within the scope of protection of the present application.

Fig. 1 is a schematic diagram of an aneurysm segmentation method provided in an embodiment of the present disclosure, as shown in fig. 1, the method includes:

step S101: and acquiring image data to be processed.

In the embodiment of the present specification, the image data to be processed is three-dimensional intracranial image data, and specifically, may be 3D-CTA image data, 3D-DSA image data, or 3D-MRA image data.

If the image data to be processed is 3D-CTA image data or 3D-DSA image data, digital silhouette of the 3D-CTA or 3D-DSA image data is required. It should be noted that, if the image data to be processed is 3D-CTA image data, before being input into the aneurysm segmentation model, preprocessing is required, specifically, image data is selected from the 3D-CTA image data as the image data to be processed according to a preset gray scale range to be input into the aneurysm segmentation model. In a specific embodiment, the preset gray scale range is 5% to 95% of the gray scale values, and the specific value of the preset gray scale range does not constitute a specific limitation of the present application.

Step S103: inputting the image data to be processed into an aneurysm segmentation model, segmenting blood vessels and aneurysms in the image data to be processed to obtain blood vessel segmentation results and aneurysm segmentation results, wherein the aneurysm segmentation model takes points on the aneurysms as centers, selects image blocks with different sizes, and trains a model obtained in advance through a convolutional neural network.

By adopting the segmentation method provided by the embodiment of the specification, the blood vessel segmentation result and the aneurysm segmentation result are obtained, and the segmentation of the aneurysm image can be realized for subsequent auxiliary treatment.

In an embodiment of the present disclosure, the aneurysm segmentation model is a model obtained by selecting image blocks of different sizes and training the image blocks by a convolutional neural network, and includes:

preprocessing training sample data to obtain preprocessed image data;

labeling the preprocessed image data to obtain first labeled image data;

In an embodiment of the present specification, the preprocessing the training sample data to obtain preprocessed image data specifically includes: and carrying out normalization processing on the training sample data to obtain preprocessed image data. In the embodiment of the present disclosure, the training sample data is 3D-CTA image data, 3D-DSA image data or 3D-MRA image data, and may also be aneurysm segmentation image data obtained based on the 3D-CTA image data, 3D-DSA image data or 3D-MRA image data. In the embodiment of the present specification, the normalization processing method is as follows: the mean value of the gray levels of the image data to be processed is subtracted from the gray levels of the image data to be processed, and then the mean value of the gray levels is divided by the variance of the gray levels.

In an embodiment of the present specification, the labeling the preprocessed image data to obtain first labeled image data specifically includes:

labeling the preprocessed image data to obtain second labeled image data;

In an embodiment of the present specification, the preprocessed image data is labeled to obtain second labeled image data, and specifically, if the training sample data is 3D-CTA image data, 3D-DSA image data, or 3D-MRA image data, aneurysm segmentation is performed on the training sample data to obtain aneurysm segmentation image data. In the embodiments of the present application, the method of segmenting an aneurysm by using training sample data may be performed by using a threshold segmentation method, and the method of segmenting an aneurysm by using training sample data is not limited to the present application. And further carrying out manual labeling on the aneurysm segmentation image data to obtain second labeled image data.

Because the aneurysm segmentation image obtained by the method of threshold segmentation and the like may have the problems of hollow-out, defect and the like, the obtained second annotation image data needs to be further post-processed, so that the shape of the aneurysm segmentation image is more complete.

In an embodiment of the present specification, labeling the preprocessed image data includes: aneurysm, blood vessel and background.

In an embodiment of the present specification, the post-processing the second annotated image data to obtain the first annotated image data specifically includes:

and/or

The obtained first labeled image data can be used for training the aneurysm segmentation model. Because the existing aneurysm segmentation method cannot achieve the aneurysms with different sizes, or the aneurysm segmentation needs to be achieved by achieving the aneurysm segmentation and the aneurysm classification at the same time when the aneurysm is segmented, the processing speed is low. Therefore, in the embodiments of the present specification, the processing speed and the sizes of different aneurysms are considered in training the aneurysm segmentation model. The image size corresponding to the training sample data or the first labeled image data is relatively large, so that the training sample data or the first labeled image data is not suitable for being directly input into a 3D convolutional neural network for training. In the embodiment of the present specification, an image corresponding to training sample data is cut into image blocks in a manner of cutting the image; and inputting the image blocks into a 3D convolutional neural network for training, and splicing prediction results of the image blocks for outputting to obtain an aneurysm segmentation result.

It should be particularly noted that in this specification, three image blocks with different sizes are selected, and a fine tuning method is adopted to perform model training, so as to obtain an aneurysm segmentation model.

In an embodiment of the present disclosure, the randomly selecting three image blocks with different sizes from the first labeled image data, and performing model training by using a fine tuning method to obtain an aneurysm segmentation model specifically includes:

In one embodiment of the present description, the image patches are randomly selected from the aneurysm, and the size of the image patches may be 32 × 32, 64 × 64, 128 × 128. In one embodiment of the present description, during training, 32 × 32 image blocks are first used for training to obtain a first aneurysm segmentation model, then fine tuning (fine-tuning) is performed on the basis of the first aneurysm segmentation model, training is performed using 64 × 64 image blocks to obtain a second aneurysm segmentation model, fine tuning is performed on the basis of the second aneurysm segmentation model, and finally training is performed using 128 × 128 image blocks to obtain a final aneurysm segmentation model.

In the model training process, in order to ensure that a better aneurysm segmentation model is obtained, a data enhancement method is used in the process of training and verifying the aneurysm segmentation model. Specifically, the data enhancement method is to randomly translate, rotate and/or flip the data left and right. In the embodiment of the present specification, in the training process of the aneurysm segmentation model, the loss functions used may be CE loss (cross entropy loss) and Dice coeffient loss, and the specific type of the loss function does not constitute a limitation of the present application.

In order to further understand the training process of the aneurysm segmentation model provided in the embodiments of the present disclosure, the following description will be given with reference to specific embodiments. Taking 3D-DSA image data as an example, fig. 2 is a flowchart illustrating a training process of an aneurysm segmentation model according to an embodiment of the present disclosure. The training of the aneurysm segmentation model comprises:

step S201: and preprocessing the training sample data to obtain the preprocessed training sample data.

In the embodiments of the present specification, a normalization process is performed on training sample data to obtain preprocessed training sample data. The normalization processing method comprises the following steps: the mean of the gray levels of the training sample data is subtracted by the variance.

Step S203: and marking the preprocessed training sample data to obtain marked sample data.

In the embodiment of the present specification, after performing aneurysm segmentation based on preprocessed training sample data, obtaining a first segmentation result; and manually modifying the first segmentation result, wherein the modified aneurysm segmentation result is a labeling result, namely labeled sample data. In an embodiment of the present specification, a modification method for manually modifying a first segmentation result includes: repairing unreasonable data points of aortic aneurysm segmentation, and/or filling data points of blood vessel defect, aneurysm defect and blood vessel disconnection; and/or filling the vascular void data points.

Step S205: and inputting the labeled sample data into a 3D convolutional neural network for training to obtain an aneurysm segmentation model.

In the embodiment of the present specification, 80% of labeled sample data is randomly selected as training set data, and the remaining 20% of labeled sample data is selected as verification set data. The training set data is used for training the aneurysm segmentation model, the verification set data is used for verifying the obtained aneurysm segmentation model, and parameters of the aneurysm segmentation model are further adjusted to obtain the optimal aneurysm segmentation model.

In one embodiment of the present description, a point selection is performed on an aneurysm, and then an image patch is selected centered on the selected point, which may be of any size. In one embodiment of the present specification, the size of the default selected image block is 96 × 96. Then, the image block is input into a trained model for full convolution, and a prediction result is output. Of course, the size of the image block may specify other sizes as well. For example, if the aneurysm is too large, which may result in 96 × 96 image blocks not containing the entire aneurysm, the size of the image block may be modified to 128 × 128, and the model may also obtain better segmentation results on larger image blocks. Similarly, if the size of the aneurysm is relatively small, 64 × 64 image patches may be selected to reduce the non-target area (aneurysm) to achieve better segmentation.

Fig. 3 is a schematic diagram illustrating an aneurysm segmentation effect provided in an embodiment of the present disclosure. As shown in fig. 3, the segmentation of the aneurysm image is realized, and a better segmentation effect is obtained.

By adopting the aneurysm segmentation method provided by the embodiment of the specification, the problems of hollow-out, blood vessel defect, aneurysm defect and the like can be reduced, the aneurysm segmentation precision and the aneurysm segmentation speed are improved, and a better aneurysm segmentation effect is obtained.

The above details describe a method for dividing an aneurysm, and accordingly, the present specification also provides an aneurysm dividing apparatus, as shown in fig. 4. Fig. 4 is a schematic diagram of an aneurysm segmentation device provided in an embodiment of the present disclosure, the device including:

an obtaining module 401, configured to obtain image data to be processed;

the segmentation module 403 is configured to input the image data to be processed into an aneurysm segmentation model, segment a blood vessel and an aneurysm in the image data to be processed, and obtain a blood vessel segmentation result and an aneurysm segmentation result, where the aneurysm segmentation model is obtained by selecting image blocks of different sizes with a point on the aneurysm as a center, and training the image blocks through a convolutional neural network to obtain a model in advance.

preprocessing training sample data to obtain preprocessed image data;

labeling the preprocessed image data to obtain first labeled image data;

labeling the preprocessed image data to obtain second labeled image data;

and/or

at least one processor; and the number of the first and second groups,

acquiring image data to be processed;

preprocessing training sample data to obtain preprocessed image data;

labeling the preprocessed image data to obtain first labeled image data;

labeling the preprocessed image data to obtain second labeled image data;

and/or

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the embodiments of the apparatus, the electronic device, and the nonvolatile computer storage medium, since they are substantially similar to the embodiments of the method, the description is simple, and the relevant points can be referred to the partial description of the embodiments of the method.

The apparatus, the electronic device, the nonvolatile computer storage medium and the method provided in the embodiments of the present description correspond to each other, and therefore, the apparatus, the electronic device, and the nonvolatile computer storage medium also have similar advantageous technical effects to the corresponding method.

In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Hardware Description Language), traffic, pl (core universal Programming Language), HDCal (jhdware Description Language), lang, Lola, HDL, laspam, hardward Description Language (vhr Description Language), vhal (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the various elements may be implemented in the same one or more software and/or hardware implementations in implementing one or more embodiments of the present description.

As will be appreciated by one skilled in the art, the present specification embodiments may be provided as a method, system, or computer program product. Accordingly, embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present description may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.

The description has been presented with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the description. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). 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 (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 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. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

This description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

The above description is only an example of the present specification, and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An aneurysm segmentation method, the method comprising:

acquiring image data to be processed;

2. The method of claim 1, wherein the aneurysm segmentation model is a model obtained by selecting image blocks of different sizes centered on a point on an aneurysm and training the image blocks by a convolutional neural network, and specifically comprises:

preprocessing training sample data to obtain preprocessed image data;

labeling the preprocessed image data to obtain first labeled image data;

3. The method of claim 2, wherein said labeling the preprocessed image data to obtain first labeled image data, comprises:

labeling the preprocessed image data to obtain second labeled image data;

4. The method of claim 3, wherein post-processing the second annotated image data to obtain first annotated image data comprises:

and repairing the second marked image data to obtain first marked image data, wherein the repairing comprises the following steps: repairing data points in the second labeled image data, wherein the aortic aneurysm is unreasonably segmented;

and/or

5. The method of claim 2, wherein the randomly selecting three image blocks with different sizes from the first labeled image data, and performing model training by using a fine tuning method to obtain an aneurysm segmentation model specifically comprises:

6. An aneurysm segmentation device, the device comprising:

the acquisition module acquires image data to be processed;

7. The apparatus of claim 6, wherein the aneurysm segmentation model is a model obtained by selecting image blocks of different sizes centered on a point on the aneurysm and training the image blocks by a convolutional neural network, and specifically comprises:

preprocessing training sample data to obtain preprocessed image data;

labeling the preprocessed image data to obtain first labeled image data;

8. The apparatus of claim 7, wherein said labeling the preprocessed image data to obtain first labeled image data comprises:

labeling the preprocessed image data to obtain second labeled image data;

9. The apparatus of claim 8, wherein post-processing the second annotated image data to obtain first annotated image data comprises:

and/or

10. The apparatus of claim 7, wherein the randomly selecting three image blocks with different sizes from the first labeled image data, and performing model training by using a fine tuning method to obtain an aneurysm segmentation model, specifically comprises:

11. An electronic device, comprising:

at least one processor; and the number of the first and second groups,

acquiring image data to be processed;