WO2021098768A1

WO2021098768A1 - Method and system for assessing aneurysm rupture risk

Info

Publication number: WO2021098768A1
Application number: PCT/CN2020/130058
Authority: WO
Inventors: 马泽; 宋凌; 印胤; 杨光明; 秦岚
Original assignee: 强联智创（北京）科技有限公司
Priority date: 2019-11-22
Filing date: 2020-11-19
Publication date: 2021-05-27
Also published as: CN111081378A; CN111081378B

Abstract

A method and a system for assessing an aneurysm rupture risk comprising: obtaining data to be processed; inputting 3D triangle grid data and/or second data and/or third data into an aneurysm risk assessment model, to obtain an aneurysm risk assessment result of the data to be processed, wherein the aneurysm risk assessment model is a digital model obtained by means of training on the basis of labeled 3D triangle grid data and/or the second data and/or the third data; and outputting the aneurysm risk assessment result of the data to be processed. The method can eliminate or reduce participation of human factors, reduce time-consumption, achieve simple and quick assessment of an aneurysm rupture risk, and provide objective support for assessment of the aneurysm rupture risk.

Description

A method and system for evaluating aneurysm rupture risk

Technical field

This specification relates to the field of medical imaging and computer technology, and in particular to a method and system for assessing the risk of aneurysm rupture.

Background technique

Intracranial aneurysms are mostly abnormal bulges that occur on the walls of intracranial arteries. According to reports, the prevalence of unruptured intracranial aneurysms in Chinese adults can be as high as 7%. Subarachnoid hemorrhage can occur after rupture. Cause severe disability or death. The treatment of intracranial aneurysm after rupture is mainly to remove intracranial hematoma and prevent blood from continuing to flow into the skull; intracranial unruptured aneurysms need to develop a personalized management plan based on the results of the aneurysm rupture risk assessment and the condition of the observed person, and be conservative Observation or surgical intervention. Therefore, the risk assessment of aneurysm rupture is of great significance.

At present, the observer observes the aneurysm morphology based on the three-dimensional DSA (Digital subtraction angiography) image, MRA (Magnetic Resonance Angiography) or CTA (CT angiography, CT angiography) image, and Combining the condition of the observer to assess the risk of aneurysm rupture, this method often relies on the experience of the observer, is affected by subjective judgment, lacks objective support, and takes a long time.

Therefore, a new method is needed that can eliminate or reduce the participation of human factors, shorten the time-consuming, realize simple and quick assessment of the risk of aneurysm rupture, and provide objective support for the assessment of the risk of aneurysm rupture.

Summary of the invention

The embodiments of this specification provide an aneurysm rupture risk assessment method and system, which are used to solve the following technical problems: in the prior art, the observer uses three-dimensional DSA (Digital subtraction angiography, digital subtraction angiography) angiography images or MRA (Magnetic Subtraction Angiography) images or MRA (Magnetic Subtraction Angiography) images according to the prior art. Resonance Angiography, magnetic resonance angiography) or CTA (CT angiography, CT angiography) image to observe the aneurysm morphology, combined with the condition of the subject to assess the risk of aneurysm rupture, this method often depends on the experience of the observer, subject to subjectivity It takes a long time to judge the impact without objective support.

The embodiment of this specification provides a method for assessing the risk of aneurysm rupture, which includes the following steps:

Acquiring data to be processed, where the data to be processed includes first data and/or second data and/or third data;

Convert the first data into 3D triangle mesh data, and input the aneurysm risk assessment model based on the 3D triangle mesh data and/or the second data and/or the third data to obtain the to-be-processed Data based on the aneurysm risk assessment result, wherein the aneurysm risk assessment model is based on the labeled 3D triangle grid data and/or the second data and/or the digital model obtained by training on the third data, so The aneurysm risk assessment result includes the aneurysm rupture risk probability of the to-be-processed data and/or the vertex classification of the 3D triangular mesh data corresponding to the to-be-processed data, and the value of the 3D triangular mesh data corresponding to the to-be-processed data The vertex is classified as whether the vertex of the 3D triangle mesh data corresponding to the to-be-processed data belongs to a blood vessel or an aneurysm;

Output the aneurysm risk assessment result of the to-be-processed data.

Further, the labeled 3D triangle mesh data is based on the known 3D triangle mesh data, the vertices belonging to the blood vessel in the known 3D triangle mesh data are marked as 0, and the known 3D triangle mesh data is marked as 0. The vertex belonging to the aneurysm in the 3D triangle mesh data is marked as 1.

Further, the labeled 3D triangle mesh data further includes:

The vertices belonging to the blood vessel and/or the vertices belonging to the aneurysm are randomly discarded for the labeled 3D triangle mesh data according to the preset ratio to obtain the coordinates of the vertices belonging to the aneurysm point and the points belonging to the aneurysm. Edge information connected between vertices.

Further, the training of the aneurysm risk assessment model includes:

Based on the labeled 3D triangle mesh data, a one-dimensional vector of the labeled 3D triangle mesh data is obtained through an arithmetic block operation, where the one-dimensional vector represents the labeled 3D triangle mesh The global feature vector of the data;

Based on the one-dimensional vector and the second data and/or third data corresponding to the labeled 3D triangle grid data, the aneurysm risk assessment of the labeled 3D triangle grid data is obtained through a sigmoid function operation As a result, the aneurysm risk assessment model is obtained.

Further, the obtaining a one-dimensional vector of the marked 3D triangular mesh data through an arithmetic block operation based on the marked 3D triangular mesh data specifically includes:

The labeled 3D triangle mesh data is subjected to the operations of the first operation block of (64, 64) dimensions and the second operation block of (64, 128, 1024) dimensions to obtain a matrix of (N, 1024), where N is The number of vertices in the labeled 3D triangle mesh data;

The dimension reduction of the matrix of (N, 1024) through the maximum pooling layer, to obtain the one-dimensional vector of the labeled 3D triangular mesh data.

Further, the second data and/or third data corresponding to the one-dimensional vector and the labeled 3D triangle mesh data are operated by a sigmoid function to obtain the labeled 3D triangle mesh data The results of the aneurysm risk assessment include:

Input the one-dimensional vector and the second data and/or third data corresponding to the labeled 3D triangle grid data into a multi-layer perceptron with a dimension of (512, 256, 2) to obtain fourth data;

Subjecting the fourth data to a sigmoid function operation to obtain the aneurysm rupture risk probability of the marked 3D triangle mesh data;

and / or

A new matrix obtained by passing the one-dimensional vector and the labeled 3D triangle mesh data through the first operation block of (64, 64) dimensions, passing through the third operation block of dimensions (512, 256, 128), and the dimension is ( 128,2) the fourth operation block to obtain the matrix of (N,2);

The matrix of (N, 2) is subjected to a sigmoid function operation to obtain the vertex classification of the labeled 3D triangle mesh data.

An aneurysm rupture risk assessment system provided by an embodiment of this specification includes:

An input module to obtain data to be processed, where the data to be processed includes first data and/or second data and/or third data;

The evaluation module converts the first data into 3D triangle mesh data, and inputs the aneurysm risk assessment model based on the 3D triangle mesh data and/or the second data and/or the third data to obtain all The aneurysm risk assessment result of the to-be-processed data, wherein the aneurysm risk assessment model is based on the marked 3D triangle grid data and/or the second data and/or the number obtained by training on the third data Model, the aneurysm risk assessment result includes the aneurysm rupture risk probability of the data to be processed and/or the vertex classification of the 3D triangle mesh data corresponding to the data to be processed, and the 3D triangle mesh corresponding to the data to be processed The vertex of the grid data is classified as whether the vertex of the 3D triangle mesh data corresponding to the data to be processed belongs to a blood vessel or an aneurysm;

The output module outputs the aneurysm risk assessment result of the to-be-processed data.

Further, the labeled 3D triangle mesh data further includes:

Further, the training of the aneurysm risk assessment model includes:

and / or

The above at least one technical solution adopted in the embodiment of this specification can achieve the following beneficial effects:

The embodiment of this specification acquires data to be processed, wherein the data to be processed includes first data and/or second data and/or third data; the first data is converted into 3D triangle mesh data based on the The 3D triangle grid data and/or the second data and/or the third data are input into an aneurysm risk assessment model to obtain an aneurysm risk assessment result of the to-be-processed data, wherein the aneurysm risk assessment model It is a digital model obtained by training based on the labeled 3D triangle grid data and/or the second data and/or the third data, and the aneurysm risk assessment result includes the aneurysm rupture risk of the data to be processed Probability and/or the vertex classification of the 3D triangle mesh data corresponding to the data to be processed, the vertex of the 3D triangle mesh data corresponding to the data to be processed is classified as the vertex of the 3D triangle mesh data corresponding to the data to be processed It belongs to a blood vessel or an aneurysm; outputting the aneurysm risk assessment result of the data to be processed can eliminate or reduce the participation of human factors, shorten time-consuming, and realize simple and quick assessment of aneurysm rupture risk, which is the risk of aneurysm rupture The evaluation provides objective support.

Description of the drawings

In order to more clearly describe the technical solutions in the embodiments of this specification 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 described in this specification. For those of ordinary skill in the art, other drawings can be obtained from these drawings without creative labor.

Fig. 1 is a schematic flow chart of a method for assessing the risk of aneurysm rupture according to an embodiment of this specification;

Figure 2 is a schematic diagram of the construction process of the aneurysm risk assessment model provided by the embodiments of this specification;

Fig. 3 is a schematic diagram of an aneurysm rupture risk assessment system provided by an embodiment of this specification.

Detailed ways

The morphological parameters of intracranial aneurysms are of great significance for the diagnosis of intracranial aneurysms. The morphological parameters of intracranial aneurysms can be obtained through medical imaging, including but not limited to DSA, MRA, and CTA.

The basic principle of DSA is to digitally input two frames of X-ray images taken before and after the injection of the contrast agent into the image computer. Through the process of subtraction, enhancement and re-imaging, a clear pure blood vessel image is obtained, and the vascular shadow is displayed in real time. DSA has the advantages of high contrast resolution, short examination time, low contrast agent consumption, low concentration, significantly reduced patient X-ray absorption, and film saving. It is of great significance in the clinical diagnosis of vascular diseases. DSA has become the gold standard for the diagnosis of intracranial arterial malformations and aneurysms due to its imaging characteristics.

The basic principle of MRA is based on saturation effect, inflow enhancement effect, and flow dephasing effect. In MRA, the pre-saturation zone is placed at the head end of the 3D slab to saturate the venous blood flow. The arterial blood flowing in the reverse direction enters the 3D slab and generates MR signals because it is not saturated. During scanning, a thicker volume is divided into multiple thin layers for excitation, reducing the thickness of the excitation volume to reduce the inflow saturation effect, and can ensure the scanning volume range, obtain several layers of adjacent layers of thin-layer images, make the image clear, and the blood vessels are subtle The structure is displayed well and the spatial resolution is improved. Due to its high-quality imaging characteristics, MRA is gradually used in the diagnosis of intracranial aneurysms.

The basic principle of CTA is that after intravenous injection of iodine-containing contrast agent, spiral CT or electron beam CT is used to continuously perform thin-slice scanning at the peak of the blood vessel filled with the contrast agent to quickly obtain a large number of thin-layer superimposed sections, which are reconstructed after computer image processing Stereoscopic images of blood vessels, clearly showing the anterior cerebral artery, middle artery, posterior artery and its main branches, Wi11is arterial ring, etc. After intravenous injection of iodine-containing contrast agent, spiral CT or electron beam CT is used to continuously perform thin-layer scanning at the peak of the blood vessel filled with the contrast agent to quickly obtain a large number of thin-layer superimposed sections, and reconstruct the three-dimensional image of the blood vessel after computer image processing. Show the anterior cerebral artery, middle artery, posterior artery and its main branches, Wi11is arterial ring, etc. Because CTA has the characteristics of non-invasive, fast, simple operation and low price, it can mostly replace DSA in the diagnosis of clinical intracranial aneurysms.

In this application, the original image based on DSA and/or CTA and/or MRA can be digitally processed into 3D triangular grid data. Based on the morphological parameters of the aneurysm of the observed person and the information parameters of the observed person, the estimated value can be realized Assessment of the risk of aneurysm rupture.

In order to enable those skilled in the art to better understand the technical solutions in this specification, the following will clearly and completely describe the technical solutions in the embodiments of this specification in conjunction with the drawings in the embodiments of this specification. Obviously, the described The embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

The technical solutions provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

Fig. 1 is a schematic flow chart of a method for assessing the risk of aneurysm rupture according to an embodiment of this specification. The method specifically includes the following steps:

Step S101: Obtain data to be processed, where the data to be processed includes first data and/or second data and/or third data.

In this application, the first data and/or the second data and/or the third data are the data corresponding to each aneurysm sample

In this application, the first data is 3D image data, including but not limited to DSA image data, CTA image data, and MRA image data.

In this application, the second data includes but is not limited to: the age and/or gender of the observed person and/or whether it is a patient with multiple aneurysms and/or history of drinking and/or smoking and/or family history and/or arteries The location and/or symptoms of the tumor in the vascular segment.

In this application, the third data is the morphological parameters of the aneurysm of the observed person, including but not limited to: aneurysm volume, average diameter of tumor-bearing blood vessels, ratio of tumor body length to tumor neck diameter (SR), tumor body length and Ratio of aneurysm neck width (AR), aneurysm length, aneurysm height, aneurysm width, aneurysm neck width, inflow angle, aneurysm neck area, aneurysm neck diameter, tumor-carrying vessel length, fluctuation index (UI), non- Spherical Index (NSI).

In the embodiment of the present application, the second data and the third data are stored in the form of a table for subsequent use.

In an embodiment of the present application, based on the first data, the morphological parameters of the aneurysm of the observed person are obtained. Specifically, the method of threshold segmentation and region growth is adopted to convert the first data into a binary image; the Marching Cube algorithm is used to construct 3D triangular mesh data; based on the constructed 3D triangular mesh data, the observed aneurysm is obtained Morphological parameters. The 3D triangle mesh data is a topological structure composed of multiple triangles in a Cartesian coordinate system in space, where each triangle contains vertices and edges.

In an embodiment of the present application, the 3D triangular mesh data adopts STL format data, specifically, it may be a binary text or a text file.

Since the observed person may have one or more aneurysms, the data to be processed includes data corresponding to one or more aneurysms.

In this application, the observed person is the subject of the risk assessment of aneurysm rupture in this application.

Step S103: Convert the first data into 3D triangular mesh data, and input the aneurysm risk assessment model based on the 3D triangular mesh data and/or the second data and/or the third data to obtain all The aneurysm risk assessment result of the to-be-processed data, wherein the aneurysm risk assessment model is based on the marked 3D triangle grid data and/or the second data and/or the number obtained by training on the third data Model, the aneurysm risk assessment result includes the aneurysm rupture risk probability of the data to be processed and/or the vertex classification of the 3D triangle mesh data corresponding to the data to be processed, and the 3D triangle mesh corresponding to the data to be processed The vertex of the grid data is classified as whether the vertex of the 3D triangle mesh data corresponding to the data to be processed belongs to a blood vessel or an aneurysm.

In this application, the 3D triangle mesh data corresponding to the data to be processed is the method of threshold segmentation and region growth for the data to be processed, and the first data is converted into a binary image; the 3D triangle mesh data obtained by the Marching Cube algorithm .

In this application, the aneurysm risk assessment model is a digital model obtained by training based on labeled 3D triangle mesh data. In order to understand the acquisition of an aneurysm risk assessment model, FIG. 2 is a schematic diagram of the acquisition process of an aneurysm risk assessment model provided in an embodiment of this specification, which specifically includes:

Step S201: Annotate the 3D triangle mesh data to obtain the annotated 3D triangle mesh data.

Based on the known 3D triangle mesh data, the vertices belonging to the blood vessel in the known 3D triangle mesh data are marked as 0, and the vertices belonging to the aneurysm in the known 3D triangular mesh data are marked as 1 . In this application, the known 3D triangle mesh data is based on the corresponding 3D triangle mesh data obtained as the first data of the training set

In the specific implementation process, the labeled 3D triangle mesh data is further: randomly discarded the vertices belonging to the blood vessel and/or the aneurysm according to the preset ratio of the labeled 3D triangle mesh data, so as to obtain the vertices belonging to the aneurysm. The coordinates of the vertices around the aneurysm point and the edge information connected between the vertices belonging to the aneurysm point. It should be particularly noted that the preset ratio is comprehensively determined according to the amount of 3D triangle mesh data that has been marked and/or the number of vertex coordinates.

In an embodiment of the present application, the labeled 3D triangle mesh data is further: randomly discarded vertices belonging to blood vessels and/or vertices belonging to aneurysms for the labeled 3D triangle mesh data according to a preset ratio. Obtain the coordinates of 1024 or 2048 vertices around the aneurysm point and the edge information connected between the 1024 or 2048 vertices around the aneurysm point. The specific number of vertices 1024 or 2048 is only an exemplary description of this application, and does not constitute a limitation of this application.

In the embodiment of the present application, the known 3D triangle mesh data can be input into the corresponding labeling software, so that the mesh data can be visualized, so that manual labeling can be performed point by point. In order to ensure the accuracy of the labeling results, multi-person labeling can be used, and the number of labeling personnel shall be at least 3 people. The person making the labeling should have professional experience to ensure the accuracy of the labeling.

In the embodiment of this application, the 3D triangle mesh data is in the form of a N*F matrix, where N is the number of vertices (for example, N=1024 means there are 1024 vertices), and F is the number of features (for example, F=3 Indicates that there are features in the three directions of x, y, and z).

Step S203: Based on the labeled 3D triangular mesh data, a one-dimensional vector of the labeled 3D triangular mesh data is obtained through an arithmetic block operation, where the one-dimensional vector represents the labeled 3D The global feature vector of the triangle mesh data.

In the embodiment of the present application, the labeled 3D triangle mesh data is subjected to the operations of the first operation block of (64, 64) dimensions and the second operation block of (64, 128, 1024) dimensions to obtain (N, 1024) ), where N is the number of vertices in the labeled 3D triangle mesh data;

Step S205: Based on the one-dimensional vector and the second data and/or third data corresponding to the labeled 3D triangle mesh data, perform a sigmoid function operation to obtain the artery of the labeled 3D triangle mesh data An aneurysm risk assessment result to obtain the aneurysm risk assessment model.

In the embodiment of the present application, the second data and/or third data corresponding to the one-dimensional vector and the labeled 3D triangular mesh data are input into a multi-layer perceptron with a dimension of (512, 256, 2) to obtain Fourth data

and / or

In the embodiments of this application, the sigmoid function is also called the Logistic function. In this application, after the sigmoid classification function is processed, it can be ensured that the output data is within (0,1), and the risk of aneurysm rupture can be assessed. The sigmoid function can also realize data classification to realize the vertex classification of the labeled 3D triangle mesh data.

During the establishment of the aneurysm risk assessment model, the data used for the establishment of the aneurysm risk assessment model includes: training set and test set. The training set is used to train the aneurysm risk assessment model, and the test set is used to test the effect of the obtained aneurysm risk assessment model.

The aneurysm risk assessment model provided by the embodiments of this specification can automatically learn the local features of each vertex and surrounding vertices in the labeled 3D triangle mesh data, and further integrate the second data and/or third data of the observed person , So as to achieve a comprehensive and efficient prediction of aneurysm rupture risk and vertex classification.

Step S105: Output the aneurysm risk assessment result of the to-be-processed data.

In this application, the output of the aneurysm risk assessment result includes: the aneurysm rupture risk probability of the to-be-processed data and/or the vertex classification of the 3D triangle mesh corresponding to the to-be-processed data.

In one embodiment of the present application, using the aneurysm risk assessment method provided in the present application, inputting the data to be processed, the probability of obtaining the aneurysm rupture risk of the observed person is 70%, which can be used as an auxiliary reference for subsequent diagnosis and treatment to determine Whether to perform surgical treatment.

The method provided in the embodiments of the present application can realize the input of one or more sets of aneurysm data to obtain an assessment result of the risk of aneurysm rupture. In this application, a set of aneurysm data is data corresponding to an aneurysm, including the first data and/or the second data and/or the third data.

The method provided in the embodiments of this specification can eliminate or reduce the participation of human factors, shorten time-consuming, realize simple and quick assessment of the risk of aneurysm rupture, and provide objective support for the assessment of the risk of aneurysm rupture.

The evaluation method provided in the embodiments of this specification can be packaged as software in actual application to assist observers in making decisions about treatment of unruptured aneurysms to quickly obtain reasonable and reliable prediction results and/or vertex classification.

Based on the same idea, the embodiment of this specification also provides an aneurysm rupture risk assessment system. FIG. 3 is a schematic diagram of an aneurysm rupture risk assessment system provided by the embodiment of this specification, and the system includes:

The input module 301 obtains data to be processed, where the data to be processed includes first data and/or second data and/or third data;

The evaluation module 303 converts the first data into 3D triangle mesh data, and inputs the aneurysm risk assessment model based on the 3D triangle mesh data and/or the second data and/or the third data to obtain The aneurysm risk assessment result of the to-be-processed data, wherein the aneurysm risk assessment model is obtained by training based on the marked 3D triangle grid data and/or the second data and/or the third data A digital model, where the aneurysm risk assessment result includes the aneurysm rupture risk probability of the to-be-processed data and/or the vertex classification of the 3D triangle mesh data corresponding to the to-be-processed data, and the 3D triangle corresponding to the to-be-processed data The vertex of the mesh data is classified as whether the vertex of the 3D triangle mesh data corresponding to the data to be processed belongs to a blood vessel or an aneurysm;

The output module 305 outputs the aneurysm risk assessment result of the to-be-processed data.

Further, the labeled 3D triangle mesh data further includes:

Further, the training of the aneurysm risk assessment model includes:

and / or

The matrix of (N, 2) is subjected to a sigmoid function operation to obtain the vertex classification of the labeled 3D triangular mesh data.

The foregoing describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than in the embodiments and still achieve desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown in order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, electronic equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and for related parts, please refer to the part of the description of the method embodiments.

The device, electronic device, non-volatile computer storage medium and method provided in the embodiments of this specification correspond to each other. Therefore, the device, electronic device, and non-volatile computer storage medium also have beneficial technical effects similar to the corresponding method. The beneficial technical effects of the method have been described in detail above, therefore, the beneficial technical effects of the corresponding device, electronic equipment, and non-volatile computer storage medium will not be repeated here.

In the 1990s, the improvement of a technology can be clearly distinguished between hardware improvements (for example, improvements in circuit structures such as diodes, transistors, switches, etc.) or software improvements (improvements in method flow). However, with the development of technology, the improvement of many methods and processes of today can be regarded as a direct improvement of the hardware circuit structure. Designers almost always get the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that the improvement of a method flow cannot be realized by the hardware entity module. For example, a programmable logic device (Programmable Logic Device, PLD) (for example, a Field Programmable Gate Array (Field Programmable Gate Array, FPGA)) is such an integrated circuit whose logic function is determined by the user's programming of the device. It is programmed by the designer to "integrate" a digital system on a PLD, without requiring the chip manufacturer to design and manufacture a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly realized with "logic compiler" software, which is similar to the software compiler used in program development and writing, but before compilation The original code must also be written in a specific programming language, which is called Hardware Description Language (HDL), and there is not only one type of HDL, but many types, such as ABEL (Advanced Boolean Expression Language) , AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description), etc., currently most commonly used It is VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. It should also be clear to those skilled in the art that only a little logic programming of the method flow in the above-mentioned hardware description languages and programming into an integrated circuit can easily obtain the hardware circuit that implements the logic method flow.

The controller can be implemented in any suitable manner. For example, the controller can take the form of, for example, a microprocessor or a processor and a computer-readable medium storing computer-readable program codes (such as software or firmware) executable by the (micro)processor. , Logic gates, switches, application specific integrated circuits (ASICs), programmable logic controllers and embedded microcontrollers. Examples of controllers include but are not limited to the following microcontrollers: ARC625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicon Labs C8051F320, the memory controller can also be implemented as part of the memory control logic. Those skilled in the art also know that, in addition to implementing the controller in a purely computer-readable program code manner, it is entirely possible to program the method steps to make the controller use logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded logic. The same function can be realized in the form of a microcontroller or the like. Therefore, such a controller can be regarded as a hardware component, and the devices included in it for realizing various functions can also be regarded as a structure within the hardware component. Or even, the device for realizing various functions can be regarded as both a software module for realizing the method and a structure within a hardware component.

The systems, devices, modules, or units explained in the above embodiments may be implemented by computer chips or entities, or implemented by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cell phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or Any combination of these devices.

For the convenience of description, when describing the above device, the functions are divided into various units and described separately. Of course, when implementing one or more embodiments of this specification, the functions of each unit may be implemented in the same one or more software and/or hardware.

Those skilled in the art should understand that the embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, the embodiments of this specification may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of this specification may adopt the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

This specification is described with reference to flowcharts and/or block diagrams of methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and/or block in the flowchart and/or block diagram, and the combination of processes and/or blocks in the flowchart and/or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing equipment to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing equipment are generated It is a device that realizes the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing equipment to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction device. The device implements the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

These computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operation steps are executed on the computer or other programmable equipment to produce computer-implemented processing, so as to execute on the computer or other programmable equipment. The instructions provide steps for implementing the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

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

The memory may include non-permanent 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 computer readable media.

Computer-readable media include permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology. The information can be computer-readable instructions, data structures, program modules, 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, CD-ROM, digital versatile disc (DVD) or other optical storage, Magnetic cassettes, magnetic tape storage or other magnetic storage devices or any other non-transmission media can be used to store information that can be accessed by computing devices. According to the definition in this article, computer-readable media does not include transitory media, such as modulated data signals and carrier waves.

It should also be noted that the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or equipment including a series of elements not only includes those elements, but also includes Other elements that are not explicitly listed, or they also include elements inherent to such processes, methods, commodities, or equipment. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, commodity, or equipment that includes the element.

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

The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, as for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for related parts, please refer to the part of the description of the method embodiment.

The above descriptions are only examples of this specification, and are not intended to limit this application. For those skilled in the art, this application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included in the scope of the claims of this application.

Claims

An aneurysm rupture risk assessment method, characterized in that the assessment method includes:

Acquiring data to be processed, where the data to be processed includes first data and/or second data and/or third data;

Convert the first data into 3D triangle mesh data, and input the aneurysm risk assessment model based on the 3D triangle mesh data and/or the second data and/or the third data to obtain the to-be-processed Data based on the aneurysm risk assessment result, wherein the aneurysm risk assessment model is based on the labeled 3D triangle grid data and/or the second data and/or the digital model obtained by training on the third data, so The aneurysm risk assessment result includes the aneurysm rupture risk probability of the to-be-processed data and/or the vertex classification of the 3D triangular mesh data corresponding to the to-be-processed data, and the value of the 3D triangular mesh data corresponding to the to-be-processed data The vertex is classified as whether the vertex of the 3D triangle mesh data corresponding to the to-be-processed data belongs to a blood vessel or an aneurysm;

Output the aneurysm risk assessment result of the to-be-processed data.
The method of claim 1, wherein the labeled 3D triangle mesh data is based on known 3D triangle mesh data, and the vertices belonging to the blood vessel in the known 3D triangle mesh data It is marked as 0, and the vertex belonging to the aneurysm in the known 3D triangle mesh data is marked as 1.
3. The method of claim 2, wherein the labeled 3D triangle mesh data further comprises:

The vertices belonging to the blood vessel and/or the vertices belonging to the aneurysm are randomly discarded for the labeled 3D triangle mesh data according to the preset ratio to obtain the coordinates of the vertices belonging to the aneurysm point and the points belonging to the aneurysm. Edge information connected between vertices.
The method of claim 1, wherein the training of the aneurysm risk assessment model comprises:

Based on the labeled 3D triangle mesh data, a one-dimensional vector of the labeled 3D triangle mesh data is obtained through an arithmetic block operation, where the one-dimensional vector represents the labeled 3D triangle mesh The global feature vector of the data;

Based on the one-dimensional vector and the second data and/or third data corresponding to the labeled 3D triangle grid data, the aneurysm risk assessment of the labeled 3D triangle grid data is obtained through a sigmoid function operation As a result, the aneurysm risk assessment model is obtained.
The method according to claim 4, wherein the obtaining a one-dimensional vector of the marked 3D triangular mesh data through an arithmetic block operation based on the marked 3D triangular mesh data specifically includes :

The labeled 3D triangle mesh data is subjected to the operations of the first operation block of (64, 64) dimensions and the second operation block of (64, 128, 1024) dimensions to obtain a matrix of (N, 1024), where N is The number of vertices in the labeled 3D triangle mesh data;

The dimension reduction of the matrix of (N, 1024) through the maximum pooling layer, to obtain the one-dimensional vector of the labeled 3D triangular mesh data.
The method according to claim 4, wherein the second data and/or third data corresponding to the one-dimensional vector and the labeled 3D triangle mesh data are obtained through a sigmoid function operation. The aneurysm risk assessment result of the marked 3D triangle grid data specifically includes:

Input the one-dimensional vector and the second data and/or third data corresponding to the labeled 3D triangle grid data into a multi-layer perceptron with a dimension of (512, 256, 2) to obtain fourth data;

Subjecting the fourth data to a sigmoid function operation to obtain the aneurysm rupture risk probability of the marked 3D triangle mesh data;

and / or

A new matrix obtained by passing the one-dimensional vector and the labeled 3D triangle mesh data through the first operation block of (64, 64) dimensions, passing through the third operation block of dimensions (512, 256, 128), and the dimension is ( 128,2) the fourth operation block to obtain the matrix of (N,2);

The matrix of (N, 2) is subjected to a sigmoid function operation to obtain the vertex classification of the labeled 3D triangle mesh data.
An aneurysm rupture risk assessment system, characterized in that the assessment system includes:

An input module to obtain data to be processed, where the data to be processed includes first data and/or second data and/or third data;

The evaluation module converts the first data into 3D triangle mesh data, and inputs the aneurysm risk assessment model based on the 3D triangle mesh data and/or the second data and/or the third data to obtain all The aneurysm risk assessment result of the to-be-processed data, wherein the aneurysm risk assessment model is based on the marked 3D triangle grid data and/or the second data and/or the number obtained by training on the third data Model, the aneurysm risk assessment result includes the aneurysm rupture risk probability of the data to be processed and/or the vertex classification of the 3D triangle mesh data corresponding to the data to be processed, and the 3D triangle mesh corresponding to the data to be processed The vertex of the grid data is classified as whether the vertex of the 3D triangle mesh data corresponding to the data to be processed belongs to a blood vessel or an aneurysm;

The output module outputs the aneurysm risk assessment result of the to-be-processed data.
The system of claim 7, wherein the labeled 3D triangle mesh data is based on known 3D triangle mesh data, and the vertices belonging to the blood vessel in the known 3D triangle mesh data It is marked as 0, and the vertex belonging to the aneurysm in the known 3D triangle mesh data is marked as 1.
The system of claim 8, wherein the labeled 3D triangle mesh data further comprises:

The vertices belonging to the blood vessel and/or the vertices belonging to the aneurysm are randomly discarded for the labeled 3D triangle mesh data according to the preset ratio to obtain the coordinates of the vertices belonging to the aneurysm point and the points belonging to the aneurysm. Edge information connected between vertices.
The system of claim 7, wherein the training of the aneurysm risk assessment model comprises:

Based on the labeled 3D triangle mesh data, a one-dimensional vector of the labeled 3D triangle mesh data is obtained through an arithmetic block operation, where the one-dimensional vector represents the labeled 3D triangle mesh The global feature vector of the data;

Based on the one-dimensional vector and the second data and/or third data corresponding to the labeled 3D triangle grid data, the aneurysm risk assessment of the labeled 3D triangle grid data is obtained through a sigmoid function operation As a result, the aneurysm risk assessment model is obtained.
The system according to claim 10, wherein the one-dimensional vector of the marked 3D triangular mesh data is obtained through an arithmetic block operation based on the marked 3D triangular mesh data, which specifically includes :

The labeled 3D triangle mesh data is subjected to the operations of the first operation block of (64, 64) dimensions and the second operation block of (64, 128, 1024) dimensions to obtain a matrix of (N, 1024), where N is The number of vertices in the labeled 3D triangle mesh data;

The dimension reduction of the matrix of (N, 1024) through the maximum pooling layer, to obtain the one-dimensional vector of the labeled 3D triangular mesh data.
The system according to claim 10, wherein the second data and/or third data corresponding to the one-dimensional vector and the labeled 3D triangle mesh data are obtained through a sigmoid function operation. The aneurysm risk assessment result of the marked 3D triangle grid data specifically includes:

Input the one-dimensional vector and the second data and/or third data corresponding to the labeled 3D triangle grid data into a multi-layer perceptron with a dimension of (512, 256, 2) to obtain fourth data;

Subjecting the fourth data to a sigmoid function operation to obtain the aneurysm rupture risk probability of the marked 3D triangle mesh data;

and / or

A new matrix obtained by passing the one-dimensional vector and the labeled 3D triangle mesh data through the first operation block of (64, 64) dimensions, passing through the third operation block of dimensions (512, 256, 128), and the dimension is ( 128,2) the fourth operation block to obtain the matrix of (N,2);

The matrix of (N, 2) is subjected to a sigmoid function operation to obtain the vertex classification of the labeled 3D triangle mesh data.