Cerebral stroke risk assessment method and system

Abstract

The invention discloses a cerebral apoplexy risk assessment method and system, and the system comprises a preprocessing module which is used for uploading a brain cerebral angiography image of a user, and carrying out the preprocessing and storage of the image; the segmentation module is used for calling the stored brain cerebral angiography image of the user and segmenting the brain cerebral angiography image; the extraction module is used for receiving the segmented brain cerebrovascular angiography image of the user and extracting a cerebrovascular contour image; the analysis module is used for receiving the extracted cerebrovascular contour image, traversing the cerebrovascular contour image and analyzing the cerebrovascular contour mutation tendency; the refreshing module is used for triggering skipping and skipping to the operation stage of the preprocessing module for refreshing operation; and the evaluation module is used for evaluating whether the user has a cerebral apoplexy worsening risk or not. According to the cerebral apoplexy risk assessment method, the cerebral apoplexy contour variation tendency is digitally defined by carrying out filtering processing, graph segmentation and contour image operation on the image, the cerebral apoplexy risk is assessed based on the cerebral apoplexy contour variation tendency which is continuously output, and the cerebral apoplexy risk assessment precision is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of health management, and specifically to a stroke risk assessment method and system. Background Art

[0002] Ischemic stroke is a common cerebrovascular disease. It is mostly caused by cerebral atherosclerosis, thrombosis or embolism, etc., resulting in stenosis or blockage of cerebral blood vessels, causing local cerebral tissue ischemia, hypoxia and necrosis, and triggering symptoms such as hemiplegia, aphasia, and sensory disorders. It is characterized by high incidence, high disability rate and high mortality rate.

[0003] The invention patent application with the application number 201910016245.6 discloses a new type of stroke risk assessment method, including: collecting target biological information, wherein the biological information includes target physiological information and target cerebral blood oxygen signal; generating target evaluation parameters after processing the target biological information, and comparing the target evaluation parameters with standard evaluation parameters to obtain a comparison result; generating an evaluation report according to the comparison result and the target physiological information. This application aims to solve the problem of "how to evaluate the stroke risk for the elderly population in home-based care, fundamentally reduce the stroke rate, actively carry out health care and rehabilitation training, and achieve safe aging in place, which is a major challenge currently faced".

[0004] However, for ischemic stroke patients, the diagnosis and monitoring of their condition is an important continuous task. Most of the existing technologies mainly focus on obtaining information related to the patient's condition, and the diagnosis process is still completed by doctors. Its efficiency is too low, and it mainly relies on the professional knowledge and experience of doctors. This method is limited by the doctor's personal experience and subjective judgment, and it is difficult to make an accurate judgment. Patients cannot get early intervention or cannot get timely treatment.

[0005] Therefore, a stroke risk assessment method and system are proposed. Summary of the Invention

[0006] Aiming at the above-mentioned disadvantages of the existing technology, the present invention provides a stroke risk assessment method and system, which can effectively solve the problems of the existing technology.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions;

[0008] The present invention discloses a stroke risk assessment system, including:

[0009] The preprocessing module is used to upload the cerebral angiography images of the user's brain, preprocess and store the cerebral angiography images of the user's brain; the segmentation module is used to retrieve the cerebral angiography images of the user's brain stored in the preprocessing module and segment the cerebral angiography images of the user's brain; the extraction module is used to receive the segmented cerebral angiography images of the user's brain in the segmentation module and extract the cerebral blood vessel contour images from the cerebral angiography images; the analysis module is used to receive the cerebral blood vessel contour images extracted by the extraction module, traverse the cerebral blood vessel contour images, and analyze the tendency of abnormal changes in the cerebral blood vessel contours; the refresh module is used to trigger a jump and jump to the running stage of the preprocessing module for refreshing; the evaluation module is used to monitor the number of times the refresh module runs, and when the number of times the refresh module runs is greater than or equal to one, use the historical analysis results of the tendency of abnormal changes in the cerebral blood vessel contours to evaluate whether the user has the risk of stroke deterioration.

[0010] Furthermore, each time the preprocessing module runs, it uploads the cerebral angiography images of the left, right, upper, and posterior directions of the brain, and the number of cerebral angiography images uploaded in each direction is not less than four. And among the cerebral angiography images belonging to the same direction, the time interval between the earliest acquired cerebral angiography image and the most recently acquired cerebral angiography image is one second;

[0011] Among them, the preprocessing operation of the cerebral angiography images of the user's brain in the preprocessing module is the filtering and denoising operation. When the preprocessing module stores the cerebral angiography images of the user's brain, it stores them separately based on the source direction of the cerebral angiography images.

[0012] Furthermore, the filtering and denoising processing logic of the cerebral angiography images of the user's brain is as follows:

[0013] The original cerebral angiography image of the user's brain is denoted as I(x, y);

[0014]

[0015] In the formula: J(x, y) is the cerebral angiography image after filtering and denoising; Ω is the neighborhood window centered on (x, y); ω(s, t) is the weight function; I(s, t) is the pixel value of the original cerebral angiography image at the coordinate (s, t);

[0016] Among them, the weight function ω(s, t) = ω d (s, t)·ω r (s, t), ω d (s, t), ω r (s, t) are the spatial domain weight and the gray scale domain weight.

[0017] Furthermore, the values of the spatial domain weight and the gray scale domain weight follow:

[0018]

[0019] In the formula: σ d is the standard deviation in the spatial domain; σ r is the standard deviation in the gray scale domain.

[0020] Furthermore, during the operation stage of the segmentation module, it synchronously identifies whether the cerebral angiography image of the user's brain is a gray-scale image. When the recognition result is no, it converts the cerebral angiography image of the user's brain into a gray-scale image and then performs the segmentation operation;

[0021] The segmentation module is provided with an editing unit and a forwarding unit at the lower level. The editing unit is used to provide the editing permission for the segmentation ratio of the cerebral angiography image of the user's brain at the system end, and the forwarding unit is used to forward the sub-cerebral angiography images obtained by the segmentation module to the extraction module;

[0022] Among them, the format of the segmentation ratio of the cerebral angiography image edited in the editing unit is X×Y, where X×Y is the number of sub-cerebral angiography images divided horizontally and vertically of the cerebral angiography image, and the size of each sub-cerebral angiography image is the same. When editing the segmentation ratio of the cerebral angiography image of the user's brain in the editing unit, it follows that the higher the accuracy requirement of the system end user for stroke risk assessment, the more the number of sub-cerebral angiography images obtained by segmentation, and vice versa, the fewer the number.

[0023] Furthermore, the segmentation module and the extraction module continuously obtain and process the cerebral angiography images in the preprocessing module, so that all the cerebral angiography images stored in the preprocessing module are processed by the segmentation module and the extraction module;

[0024] Among them, the target of each operation and processing of the segmentation module and the extraction module is the same cerebral angiography image.

[0025] Furthermore, the preprocessing module to the refresh module runs in a loop to form a closed loop, so as to continuously output the tendency of cerebrovascular contour variation through the analysis module;

[0026] The analysis logic of the tendency of cerebrovascular contour variation in the analysis module is expressed as:

[0027]

[0028] In the formula: Q up is the tendency of cerebrovascular contour variation shown by the cerebrovascular contour image from the upper cerebral angiography image; n is the set of corresponding positions of the sub-cerebral angiography images obtained by segmentation; m i is the total amount of cerebrovascular contour images extracted from the sub-cerebral angiography image corresponding to the i-th position; S j，j+1is the difference between the j-th cerebrovascular contour image and the (j + 1)-th cerebrovascular contour image;

[0029] Among them, represents the average of , and the cerebrovascular contour variation tendency shown by the cerebrovascular contour images from the left, right, and posterior cerebral angiography images has the same calculation logic as Q up , denoted as Q left , Q right , Q respectively , and the result output by the analysis module is the maximum value among the calculation results corresponding to the four directions;

[0030]

[0031] In the formula: A, P, and C are the area, perimeter, and roundness of the cerebrovascular contour image, ω A , ω P , ω C are weights, the sum of the weights is 1, and they are all positive numbers.

[0032] Furthermore, the evaluation logic for whether the user has the risk of stroke deterioration in the evaluation module is:

[0033] Q g (MAX) > Q g-1 (MAX) > Q g-2 (MAX);

[0034] In the formula: Q g (MAX) is the cerebrovascular contour variation tendency output by the latest operation of the analysis module; Q g-1 (MAX) is the cerebrovascular contour variation tendency output by the analysis module compared to the previous output of Q g (MAX); Q g-2 (MAX) is the cerebrovascular contour variation tendency output by the analysis module compared to the previous output of Q g-1 (MAX);

[0035] Among them, if the above formula holds, it means that the user has the risk of stroke deterioration, otherwise, it means that the user does not have the risk of stroke deterioration.

[0036] Furthermore, the preprocessing module is interconnected with the segmentation module and the extraction module through a wireless network. The lower level of the segmentation module is interconnected with the editing unit and the forwarding unit through a wireless network. The extraction module is interconnected with the analysis module and the refresh module through a wireless network. The refresh module is interconnected with the evaluation module through a wireless network.

[0037] A stroke risk assessment method includes:

[0038] Upload the user's cerebral angiography images, and perform filtering and denoising processing on the cerebral angiography images; segment and extract the contours of the cerebral angiography images that have completed filtering and denoising processing; analyze the tendency of cerebral blood vessel contour abnormalities based on the cerebral blood vessel contour images; continuously upload the user's cerebral angiography images, and continuously perform the operation of analyzing the tendency of cerebral blood vessel contour abnormalities; evaluate whether the user has a risk of stroke deterioration based on the continuous analysis results of the tendency of cerebral blood vessel contour abnormalities.

[0039] Adopting the technical solution provided by the present invention, compared with the known prior art, it has the following beneficial effects:

[0040] By uploading the user's cerebral angiography images and performing filtering processing on the images, the present invention greatly improves the accuracy of stroke risk assessment. Through graphic segmentation and contour image operations, the accuracy of stroke risk assessment is further improved. Furthermore, through the comparison of cerebral blood vessel contour images, the tendency of cerebral blood vessel contour abnormalities is digitally defined. Then, by continuously uploading cerebral angiography images, the tendency of cerebral blood vessel contour abnormalities is continuously output. Finally, based on the continuously output tendency of cerebral blood vessel contour abnormalities, the stroke risk is evaluated, bringing continuous, effective, and more rapid and accurate stroke risk assessment services to stroke patients, enabling them to understand the disease risk and deterioration risk earlier and take intervention measures, providing better protection for the life safety of stroke patients. Brief Description of the Drawings

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0042] Figure 1 It is a structural schematic diagram of a stroke risk assessment system;

[0043] Figure 2 It is a flowchart of a stroke risk assessment method. Detailed Embodiments

[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0045] The following further describes the present invention with reference to the embodiments.

[0046] Example 1:

[0047] A stroke risk assessment system according to this example, as Figure 1 shown, includes:

[0048] A preprocessing module 1, used to upload the user's cerebral angiography image, preprocess and store the user's cerebral angiography image;

[0049] The preprocessing module 1 runs each time to upload the cerebral angiography images in the left, right, upper, and posterior directions, and the number of cerebral angiography images uploaded in each direction is not less than four. And among the cerebral angiography images belonging to the same direction, the time interval between the earliest acquired cerebral angiography image and the most recently acquired cerebral angiography image is one second;

[0050] Among them, the preprocessing operation of the user's cerebral angiography image in the preprocessing module 1 is the filtering and denoising operation. When the preprocessing module 1 stores the user's cerebral angiography image, it is stored separately based on the source direction of the cerebral angiography image;

[0051] The filtering and denoising processing logic of the user's cerebral angiography image is:

[0052] The original user's cerebral angiography image is denoted as I(x, y);

[0053]

[0054] In the formula: J(x, y) is the cerebral angiography image after filtering and denoising; Ω is the neighborhood window centered on x and y; ω(s, t) is the weight function; I(s, t) is the pixel value of the original cerebral angiography image at the coordinate (s, t);

[0055] Among them, the weight function ω(s, t) = ω d (s, t)·ω r (s, t), ω d (s, t), ω r (s, t) are the spatial domain weight and the gray domain weight;

[0056] The values of the spatial domain weight and the gray domain weight follow:

[0057]

[0058] In the formula: σ d is the spatial domain standard deviation; σr is the gray domain standard deviation;

[0059] Filter and denoise the user's cerebral angiography image through the above logical formula, so as to improve the quality of the user's cerebral angiography image, provide a more accurate image for the subsequent modules of the system in this embodiment, and improve the accuracy of the system operation output result.

[0060] The segmentation module 2 is used to retrieve the user's cerebral angiography image stored in the preprocessing module 1 and segment the user's cerebral angiography image;

[0061] During the operation stage of the segmentation module 2, synchronously identify whether the user's cerebral angiography image is a grayscale image. When the recognition result is no, convert the user's cerebral angiography image into a grayscale image and then perform the segmentation operation;

[0062] The segmentation module 2 is subordinate to an editing unit 21 and a forwarding unit 22. The editing unit 21 is used to provide the system-side user with the editing permission for the segmentation ratio of the cerebral angiography image, and the forwarding unit 22 is used to forward the segmented sub-cerebral angiography image obtained by the segmentation module 2 to the extraction module;

[0063] Among them, the format of the segmentation ratio of the cerebral angiography image edited in the editing unit 21 is X×Y. X×Y is the number of segmented sub-cerebral angiography images in the horizontal and vertical directions of the cerebral angiography image, and the size of each sub-cerebral angiography image is the same. When editing the segmentation ratio of the user's cerebral angiography image in the editing unit 21, it follows that the higher the system-side user's requirement for the accuracy of stroke risk assessment, the more the number of segmented sub-user cerebral angiography images, and vice versa, the fewer the number;

[0064] The extraction module 3 is used to receive the segmented user's cerebral angiography image in the segmentation module 2 and extract the cerebral blood vessel contour image from the cerebral angiography image;

[0065] The segmentation module 2 and the extraction module 3 continuously acquire the cerebral angiography image in the preprocessing module 1 for processing, so that all the cerebral angiography images stored in the preprocessing module 1 are processed by the segmentation module 2 and the extraction module 3;

[0066] Among them, the processing target of the segmentation module 2 and the extraction module 3 each time is the same cerebral angiography image;

[0067] The analysis module 4 is used to receive the cerebral blood vessel contour image extracted by the extraction module 3 during operation, traverse the cerebral blood vessel contour image, and analyze the tendency of abnormal changes in the cerebral blood vessel contour;

[0068] The preprocessing module 1 to the refresh module 5 run in a loop to form a closed loop, so as to continuously output the tendency of abnormal changes in the cerebral blood vessel contour through the analysis module 4;

[0069] The analysis logic for the tendency of cerebrovascular contour variation in the analysis module 4 is expressed as:

[0070]

[0071] In the formula: Q up is the tendency of cerebrovascular contour variation in the cerebrovascular contour image representation sourced from the cerebral angiography image of the upper brain; n is the set of corresponding positions in the sub-brain cerebral angiography images obtained by segmentation; m i is the total number of cerebrovascular contour images extracted from the sub-brain cerebral angiography image corresponding to the i-th position; S j，j+1 is the difference between the j-th cerebrovascular contour image and the (j + 1)-th cerebrovascular contour image;

[0072] Among them, represents the averaging of . The tendency of cerebrovascular contour variation in the cerebrovascular contour image representation sourced from the left, right, and posterior cerebral angiography images has the same calculation logic as Q up and is denoted as Q left , Q right , Q respectively . The result output by the analysis module 4 is the maximum value among the calculation results corresponding to the four directions;

[0073]

[0074] In the formula: A, P, and C are the area, perimeter, and roundness of the cerebrovascular contour image, ω A , ω P , ω C are weights, the sum of the weights is 1, and they are all positive numbers;

[0075] It should be noted that the roundness here represents the geometric feature of the cerebrovascular contour. The closer its value is to 1, the closer the cerebrovascular contour is to a circle. The units of area and perimeter should be consistent during the operation of this formula;

[0076] Through the above logical formula calculation, the tendency of cerebrovascular contour variation is represented in a digital form, thus providing data support for the operation of the evaluation module 6 in this embodiment of the system.

[0077] The refresh module 5 is used to trigger a jump and jump to the preprocessing module 1 for refreshed operation during the running stage;

[0078] The evaluation module 6 is used to monitor the number of times the refresh module 5 runs. When the number of times the refresh module 5 runs is greater than or equal to one, the historical analysis results of the tendency of cerebrovascular contour variation are used to evaluate whether the user has a risk of stroke deterioration;

[0079] The evaluation logic for whether the user has a risk of stroke deterioration in the evaluation module 6 is:

[0080] Q g (MAX) > Q g-1 (MAX) > Q g-2 (MAX);

[0081] Where: Q g (MAX) is the tendency of abnormal changes in the cerebrovascular contour output by the analysis module 4 in its latest operation; Q g-1 (MAX) is compared with Q by the analysis module 4 g (MAX) is the tendency of abnormal changes in the cerebrovascular contour output in the previous time; Q g-2 (MAX) is compared with Q by the analysis module 4 g-1 (MAX) is the tendency of abnormal changes in the cerebrovascular contour output in the previous time;

[0082] If the above formula holds, it means that the user has a risk of stroke deterioration; otherwise, it means that the user does not have a risk of stroke deterioration;

[0083] The preprocessing module 1 is interconnected with the segmentation module 2 and the extraction module 3 through a wireless network. The lower level of the segmentation module 2 is interconnected with the editing unit 21 and the forwarding unit 22 through a wireless network. The extraction module 3 is interconnected with the analysis module 4 and the refresh module 5 through a wireless network. The refresh module 5 is interconnected with the evaluation module 6 through a wireless network.

[0084] In this embodiment, the preprocessing module 1 runs to upload the user's cerebral angiography image, preprocesses and stores the user's cerebral angiography image. The segmentation module 2 runs later to retrieve the user's cerebral angiography image stored in the preprocessing module 1 and segment the user's cerebral angiography image. The editing unit 21 synchronously provides the system-side user with the editing permission for the segmentation ratio of the cerebral angiography image. The forwarding unit 22 forwards the sub-cerebral angiography image obtained by the segmentation module 2 after segmentation to the extraction module in real time. The extraction module 3 further receives the user's cerebral angiography image segmented by the segmentation module 2, extracts the cerebrovascular contour image from the cerebral angiography image, and then the analysis module 4 receives the cerebrovascular contour image extracted by the extraction module 3, traverses the cerebrovascular contour image, analyzes the tendency of abnormal changes in the cerebrovascular contour. The refresh module 5 runs to trigger a jump and jumps to the running stage of the preprocessing module 1 to refresh and run. Finally, the evaluation module 6 monitors the running times of the refresh module 5. When the running times of the refresh module 5 are greater than or equal to one, the historical analysis results of the tendency of abnormal changes in the cerebrovascular contour are used to evaluate whether the user has a risk of stroke deterioration.

[0085] Through the operation of the system in the above embodiments, it brings effective continuous condition assessment services for stroke patients, greatly reduces the dependence on medical staff for condition diagnosis, ensures that stroke patients can more conveniently understand their own condition, and thus cooperate with medical staff to take corresponding treatment measures to ensure the stability and recovery of the condition.

[0086] Embodiment 2:

[0087] At the specific implementation level, on the basis of Embodiment 1, this embodiment refers to Figure 2 to further specifically describe a stroke risk assessment system in Embodiment 1:

[0088] A stroke risk assessment method includes:

[0089] Upload the cerebral angiography image of the user, and perform filtering and denoising processing on the cerebral angiography image of the brain; segment and extract the contour of the cerebral angiography image of the brain after the filtering and denoising processing is completed; analyze the tendency of abnormal changes in the cerebral blood vessel contour according to the cerebral blood vessel contour image; continuously upload the cerebral angiography image of the user, and continuously perform the operation of analyzing the tendency of abnormal changes in the cerebral blood vessel contour; evaluate whether the user has the risk of stroke deterioration based on the continuous analysis results of the tendency of abnormal changes in the cerebral blood vessel contour.

[0090] In summary, the system in the above embodiments greatly improves the accuracy of stroke risk assessment by uploading the cerebral angiography image of the user and performing filtering processing on the image, and further improves the accuracy of stroke risk assessment through graphic segmentation and contour image operations. Furthermore, through the comparison of cerebral blood vessel contour images, the tendency of abnormal changes in the cerebral blood vessel contour is digitally defined, and then continuous cerebral angiography images are uploaded to continuously output the tendency of abnormal changes in the cerebral blood vessel contour. Finally, the stroke risk is evaluated based on the continuously output tendency of abnormal changes in the cerebral blood vessel contour, bringing continuous, effective and more rapid and accurate stroke risk assessment services for stroke patients, and providing better protection for the life safety of stroke patients.

[0091] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements will not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A stroke risk assessment system, characterized in that, Including: A preprocessing module (1) for uploading the cerebral angiography images of the user's brain, preprocessing and storing the cerebral angiography images of the user's brain; A segmentation module (2) for retrieving the cerebral angiography images of the user's brain stored in the preprocessing module (1) and segmenting the cerebral angiography images of the user's brain; An extraction module (3) for receiving the segmented cerebral angiography images of the user's brain in the segmentation module (2) and extracting the cerebral blood vessel contour images from the cerebral angiography images; An analysis module (4) for receiving the cerebral blood vessel contour images extracted by the extraction module (3) during operation, traversing the cerebral blood vessel contour images, and analyzing the tendency of abnormal changes in the cerebral blood vessel contours; A refresh module (5) for triggering a jump and jumping to the operation stage of the preprocessing module (1) for refreshing operation; An evaluation module (6) for monitoring the number of times the refresh module (5) operates. When the number of times the refresh module (5) operates is greater than or equal to one, the historical analysis results of the tendency of abnormal changes in the cerebral blood vessel contours are used to evaluate whether the user has a risk of stroke deterioration.

2. The stroke risk assessment system according to claim 1, wherein Each time the preprocessing module (1) operates, it uploads the cerebral angiography images of the brain in the left, right, upper, and posterior directions, and the number of cerebral angiography images uploaded in each direction is not less than four. Moreover, among the cerebral angiography images belonging to the same direction, the time interval between the earliest acquired cerebral angiography image and the most recently acquired cerebral angiography image is one second; Among them, the preprocessing operation on the cerebral angiography images of the user's brain in the preprocessing module (1) is a filtering and denoising operation. When the preprocessing module (1) stores the cerebral angiography images of the user's brain, it stores them separately based on the source direction of the cerebral angiography images.

3. The stroke risk assessment system according to claim 2, wherein The filtering and denoising processing logic of the cerebral angiography images of the user's brain is as follows: The original cerebral angiography image of the user's brain is denoted as I(x, y); In the formula: J(x, y) is the cerebral angiography image after filtering and denoising processing; Ω is the neighborhood window centered on (x, y); ω(s, t) is the weight function; I(s, t) is the pixel value of the original cerebral angiography image at the coordinate (s, t); Among them, the weight function ω(s,t) = ω d (s,t)·ω r (s,t), ω d (s,t), ω r (s,t) are the spatial domain weight and the gray-scale domain weight.

4. The stroke risk assessment system according to claim 3, characterized in that, The values of the spatial domain weight and the gray domain weight follow: where: σ d is the standard deviation in the spatial domain; σr is the standard deviation in the gray-scale domain.

5. The stroke risk assessment system according to claim 1, characterized in that, During the operation stage of the segmentation module (2), it synchronously identifies whether the cerebral angiography image of the user's brain is a gray-scale image. When the identification result is no, it converts the cerebral angiography image of the user's brain into a gray-scale image and then performs the segmentation operation; The segmentation module (2) is subordinate to an editing unit (21) and a forwarding unit (22). The editing unit (21) is used to provide the editing permission for the segmentation ratio of the cerebral angiography images of the user's brain at the system end, and the forwarding unit (22) is used to forward the sub-cerebral angiography images obtained by the segmentation module (2) after segmentation to the extraction module; Among them, the segmentation ratio format of the brain cerebral angiography image edited in the editing unit (21) is X×Y. X×Y is the number of segmented sub-brain cerebral angiography images in the horizontal and vertical directions of the brain cerebral angiography image, and the size of each sub-brain cerebral angiography image is the same. When editing the segmentation ratio of the user's brain cerebral angiography image in the editing unit (21), it follows that the higher the accuracy requirement of the system-end user for stroke risk assessment, the more the number of segmented sub-user brain cerebral angiography images, and vice versa, the fewer the number.

6. The stroke risk assessment system according to claim 1, characterized in that The segmentation module (2) and the extraction module (3) continuously acquire the brain cerebral angiography images in the preprocessing module (1) for processing, so that all the brain cerebral angiography images stored in the preprocessing module (1) are processed by the segmentation module (2) and the extraction module (3); Among them, the processing target of the segmentation module (2) and the extraction module (3) each time is the same brain cerebral angiography image.

7. The stroke risk assessment system according to claim 1, wherein The preprocessing module (1) to the refresh module (5) run in a loop to form a closed loop, so as to continuously output the tendency of cerebrovascular contour variation through the analysis module (4); The analysis logic of the tendency of cerebrovascular contour variation in the analysis module (4) is expressed as: Where: Q up is the tendency of cerebrovascular contour variation in the cerebrovascular contour image representation from the upper brain cerebral angiogram image; n is the set of corresponding positions of the sub-brain cerebral angiogram images obtained by segmentation; m i is the total amount of cerebrovascular contour images extracted from the sub-brain cerebral angiogram image corresponding to the i-th position; S j，j+1 is the difference between the j-th cerebrovascular contour image and the j+1-th cerebrovascular contour image; Among them, represents the averaging of , and the tendency of cerebrovascular contour variation in the cerebrovascular contour image obtained from the cerebrovascular angiography images of the left, right, and posterior brain parts is the same as the calculation logic of Q up , denoted as Q left , Q right , Q respectively . The result output by the analysis module (4) is the maximum value among the calculation results corresponding to the four directions. Where: A, P, and C are the area, perimeter, and circularity of the cerebrovascular contour image, and ω A , ω P , ω C are weights, the sum of the weights is 1, and all are positive numbers.

8. The stroke risk assessment system according to claim 1, wherein, The evaluation logic of whether the user has the risk of stroke deterioration in the evaluation module (6) is: Q g (MAX) > Q g-1 (MAX) > Q g-2 (MAX); Where: Q g (MAX) is the tendency of abnormal changes in the cerebrovascular contour output by the analysis module (4) in the latest operation; Q g-1 (MAX) is compared with Q by the analysis module (4) g (MAX) is the tendency of abnormal changes in the cerebrovascular contour output in the previous time; Q g-2 (MAX) is compared with Q by the analysis module (4) g-1 (MAX) is the tendency of abnormal changes in the cerebrovascular contour output in the previous time; Among them, if the above formula holds, it means that the user has the risk of stroke deterioration, and vice versa, it means that the user does not have the risk of stroke deterioration.

9. The stroke risk assessment system according to claim 1, characterized in that The preprocessing module (1) is interconnected with the segmentation module (2) and the extraction module (3) through a wireless network. The lower level of the segmentation module (2) is interconnected with the editing unit (21) and the forwarding unit (22) through a wireless network. The extraction module (3) is interconnected with the analysis module (4) and the refresh module (5) through a wireless network. The refresh module (5) is interconnected with the evaluation module (6) through a wireless network.

10. A method for evaluating the risk of stroke, which is an implementation method of a stroke risk assessment system as described in any one of claims 1-9, characterized in that, Including: Upload the user's brain cerebral angiography image and perform filtering and denoising processing on the brain cerebral angiography image; Segment and extract the contour of the brain cerebral angiography image after the filtering and denoising processing is completed; Analyze the tendency of cerebrovascular contour variation according to the cerebrovascular contour image; Continuously upload the user's brain cerebral angiography image and continuously perform the operation of analyzing the tendency of cerebrovascular contour variation; Evaluate whether the user has the risk of stroke deterioration based on the continuous analysis results of the tendency of cerebrovascular contour variation.

Images