Risk assessment method and assessment device for aneurysm

Abstract

The invention provides an aneurysm risk assessment method and device, and the method comprises the steps: obtaining image data corresponding to a patient end based on a risk assessment model, determining a detection region in the image data, carrying out the external risk analysis of the detection region according to an external detection mode, and obtaining external risk data corresponding to the detection region, the method comprises the following steps: acquiring external risk data in a detection area, acquiring a marker in the detection area, performing internal risk analysis on the marker according to an internal detection mode to obtain internal risk data corresponding to the detection area, determining risk assessment data of the detection area according to the external risk data and the internal risk data, and sending the risk assessment data to a physician end.

Description

Technical Field

[0001] The present invention relates to image processing technology, and in particular to a risk assessment method and device for an aneurysm. Background Art

[0002] In clinical medicine, aneurysm is a serious vascular disease, and the rupture risk assessment is crucial for the selection of treatment options for patients.

[0003] In the prior art, the risk assessment of aneurysm rupture is often based on medical staff's viewing of angiographic images. However, different medical staff may have different observation results of the same angiographic image due to differences in experience, skill level and professional knowledge, resulting in deviations in the assessment of aneurysm rupture risk, thereby affecting the accuracy of the assessment reference data.

[0004] Therefore, how to provide more comprehensive and accurate reference data for aneurysm rupture risk assessment has become an urgent problem to be solved. Summary of the invention

[0005] The present invention provides an aneurysm risk assessment method and an assessment device, which can provide more comprehensive and accurate reference data for aneurysm rupture risk assessment.

[0006] A first aspect of the present invention provides a method for assessing the risk of an aneurysm, comprising: Acquire image data corresponding to the patient end based on the risk assessment model, and determine a detection area in the image data; Performing external risk analysis on the detection area according to the external detection mode to obtain external risk data corresponding to the detection area; Acquire the marker in the detection area, perform internal risk analysis on the marker according to the internal detection mode, and obtain internal risk data corresponding to the detection area; The risk assessment data of the detection area is determined according to the external risk data and the internal risk data, and the risk assessment data is sent to the physician end.

[0007] Optionally, in a possible implementation manner of the first aspect, performing external risk analysis on the detection area according to the external detection mode to obtain external risk data corresponding to the detection area includes: Acquire a plurality of historical image data based on the external detection mode, and acquire a historical detection area in each of the historical image data; Arrange the historical detection areas in chronological order to obtain a detection sequence, and sequentially obtain area differences between adjacent historical detection areas in the detection sequence; When the area difference of the detected number of regions is greater than or equal to the external change threshold, determining that the change trend of the detection area is an increasing trend; When the area difference of the region where there is no detection number is greater than or equal to the external change threshold, determining that the change trend of the detection area is a stable trend; The detection area of ​​the increasing trend is determined as the external abnormal area, and the detection area of ​​the stable trend is determined as the external normal area, and external risk data is obtained according to the external abnormal area or the external normal area.

[0008] Optionally, in a possible implementation manner of the first aspect, acquiring a marker in the detection area, performing internal risk analysis on the marker according to an internal detection mode, and obtaining internal risk data corresponding to the detection area includes: Determine a pixel point in the detection area that is within the marked pixel interval as a marked pixel point, and obtain a marked contour corresponding to the marker according to adjacent marked pixel points; Acquiring the distribution position of the marker contour, and determining an evaluation mode of the marker according to the distribution position of the marker, wherein the evaluation mode includes an attached evaluation mode and a free evaluation mode; An internal risk analysis is performed on the marker according to the evaluation mode to obtain internal risk data corresponding to the detection area.

[0009] Optionally, in a possible implementation manner of the first aspect, the distribution position of the marker contour is obtained, and an evaluation mode of the marker is determined according to the distribution position of the marker, the evaluation mode including an attached evaluation mode and a free evaluation mode, including: Acquire contour points connecting the detection area and the blood vessel area in the image data, and generate an opening contour according to adjacent contour points; Determine the distribution position corresponding to the mark contour in contact with the edge of the detection area as the tube wall position, obtain the closest distance between the mark contour and the opening contour, and determine the distribution position corresponding to the mark contour whose closest distance is less than the opening distance threshold as the opening position; Acquire the distribution position corresponding to the marking contour that is not in contact with the edge of the detection area and whose closest distance is greater than or equal to the opening distance threshold as the intracavitary position; The markers that determine the opening position and edge position correspond to the attachment evaluation mode, and the markers that determine the intracavitary position correspond to the free evaluation mode.

[0010] Optionally, in a possible implementation of the first aspect, performing internal risk analysis on the marker according to the evaluation mode to obtain internal risk data corresponding to the detection area includes: Based on the attachment evaluation mode, a type of evaluation data of the marker is obtained, wherein the type of evaluation data at least includes the marker density, the marker area, and the contact area between the marker and the detection area, and the risk coefficient of the marker is determined according to the type of evaluation data; or Acquire the second type of evaluation data of the marker according to the free evaluation mode, wherein the second type of evaluation data at least includes the area of ​​the marker, and determine the risk coefficient of the marker according to the second type of evaluation data; Based on the risk coefficient of each marker in the detection area, the detection area is subjected to risk assessment to obtain an internal abnormal area or an internal normal area, and internal risk data is obtained based on the internal abnormal area or the internal normal area.

[0011] Optionally, in a possible implementation manner of the first aspect, obtaining a type of evaluation data of the marker based on the attachment evaluation mode, the type of evaluation data at least including marker density, marker area, and contact area between the marker and the detection area, and determining the risk coefficient of the marker according to the type of evaluation data, comprising: Obtaining a pixel mean value corresponding to the marking contour of the marker, and traversing multiple pixel intervals based on the pixel mean value; Determine a preset density corresponding to a pixel interval where the pixel mean value is located as the marker density, and obtain a density assessment coefficient according to a ratio of the marker density to a reference density; Counting the number of pixels of the marker contour, determining the marker area corresponding to the marker according to the number of pixels, and obtaining an area evaluation coefficient according to the ratio of the marker area to the total area of ​​the detection zone; Obtaining the number of intersecting pixels between the marker contour and the edge of the detection area, obtaining the contact area according to the number of intersecting pixels, and obtaining a contact surface evaluation coefficient based on a ratio of the contact area to a reference contact area; The risk coefficient is obtained according to the sum of the density assessment coefficient, the area assessment coefficient and the contact surface assessment coefficient.

[0012] Optionally, in a possible implementation manner of the first aspect, obtaining second-category evaluation data of the marker according to the free evaluation mode, the second-category evaluation data at least including the marker area, and determining the risk coefficient of the marker according to the second-category evaluation data, comprising: Counting the number of pixels corresponding to the marking contour of the marker, and determining the area of ​​the marker according to the number of pixels; The risk coefficient corresponding to the marker whose area is smaller than the abnormal area threshold is obtained as a preset low risk coefficient; Determine that the marker whose area is greater than or equal to the abnormal area threshold is an abnormal object, count the abnormal areas of all abnormal objects, and obtain a first abnormal coefficient according to the ratio of the abnormal area to the total area of ​​the detection area; Acquire a plurality of historical image data, and determine a historical anomaly corresponding to a free evaluation mode and having a marked object area greater than or equal to an abnormal area threshold in each of the historical image data; Counting the historical abnormal areas corresponding to the historical abnormal objects in each of the historical image data, and obtaining the abnormal differences between adjacent historical abnormal areas in chronological order; The number of abnormalities corresponding to the abnormal difference value greater than the abnormal difference threshold is obtained, a second abnormality coefficient is obtained according to the ratio of the abnormal number to the benchmark abnormal number, and the risk coefficient is obtained according to the sum of the first abnormality coefficient and the second abnormality coefficient.

[0013] Optionally, in a possible implementation of the first aspect, based on the risk coefficient of each of the markers in the detection area, performing risk determination on the detection area to obtain an internal abnormal area or an internal normal area includes: According to the distribution position corresponding to each of the markers, a corresponding preset risk threshold is retrieved, and a detection area where the risk coefficient corresponding to each of the markers is less than the preset risk threshold is determined as an internal normal area; Determine the detection area where the risk coefficient corresponding to the marker is greater than or equal to the preset risk threshold as an internal abnormal area; The risk coefficient greater than or equal to the preset risk threshold is obtained as the abnormal coefficient, and the total abnormal difference between each abnormal coefficient and the corresponding preset risk threshold is counted; A plurality of abnormal range intervals are traversed according to the abnormal total difference value, and it is determined that the abnormal level corresponding to the abnormal range interval where the abnormal total difference value is located is the internal risk level corresponding to the internal abnormal area.

[0014] Optionally, in a possible implementation manner of the first aspect, determining risk assessment data of the detection area according to the external risk data and the internal risk data, and sending the risk assessment data to a physician end, comprises: Determine that the detection area where the external risk data is the external normal area and the internal risk data is the internal normal area is a normal area; Acquire the detection area where the external risk data is the external abnormal area and / or the internal risk data is the internal abnormal area as the abnormal area; The abnormal icons corresponding to the external abnormal area and / or the internal abnormal area are retrieved, and the abnormal slots corresponding to the detection area in the image data are updated according to the abnormal icons to obtain risk assessment data.

[0015] A second aspect of the present invention provides an aneurysm risk assessment device, comprising: A determination module, configured to obtain image data corresponding to the patient end based on the risk assessment model, and determine a detection area in the image data; An analysis module, used for performing external risk analysis on the detection area according to the external detection mode to obtain external risk data corresponding to the detection area; an acquisition module, used to acquire the markers in the detection area, perform internal risk analysis on the markers according to the internal detection mode, and obtain internal risk data corresponding to the detection area; The sending module is used to determine the risk assessment data of the detection area according to the external risk data and the internal risk data, and send the risk assessment data to the physician end.

[0016] The beneficial effects of the present invention are as follows: 1. The present invention can combine the imaging data of the aneurysm after embolization, and predict the rupture risk of the aneurysm through data from multiple dimensions, so as to obtain more comprehensive risk assessment data and provide more comprehensive assessment reference data for physicians. The present invention can use an external detection mode to perform external risk prediction on the detection area corresponding to the aneurysm image in the image, determine the area change trend of the detection area, and analyze the rupture risk of the aneurysm according to the area change trend of the detection area, thereby providing more accurate reference data for physicians. The present invention can obtain the markers existing inside the aneurysm image according to the imaging data, and perform internal risk analysis on the markers through the internal detection mode, and can predict the rupture risk of the aneurysm according to the size and position of the markers in the image, so as to obtain more comprehensive reference data.

[0017] 2. When the present invention predicts the rupture risk of an aneurysm according to an external detection mode, it can obtain historical image data corresponding to multiple different time periods, and identify the historical detection area corresponding to the detection area from the historical image data. By arranging these historical detection areas in chronological order, it is possible to obtain the regional area difference corresponding to the adjacent historical detection areas. The regional area difference can be used to more accurately judge the area change trend of the historical detection area, so that the rupture risk of the aneurysm can be predicted through the image data of different time periods. When the present invention analyzes the rupture risk of an aneurysm according to an internal detection mode, it can configure a corresponding evaluation mode for the marker according to the distribution position of the marker. After obtaining the evaluation mode corresponding to the marker, the rupture risk of the aneurysm can be predicted in different ways according to the density, area size, and distribution position of the marker. The rupture risk of the aneurysm can be predicted through data from multiple dimensions, and more comprehensive reference data can be obtained.

[0018] 3. After acquiring external risk data and internal risk data, the present invention can obtain risk assessment data corresponding to the detection area, and by sending the risk assessment data to the physician end, it can provide physicians with assessment reference data. The present invention can pre-configure a series of abnormal icons corresponding to abnormal types, and display the abnormal types of abnormal areas through abnormal icons, so that physicians can quickly view the abnormal types and take targeted treatment measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flow chart of a method for risk assessment of an aneurysm provided by an embodiment of the present invention; Figure 2 is a schematic diagram of a detection sequence provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of a distribution position corresponding to a marking contour provided by an embodiment of the present invention being a pipe wall position; Figure 4 It is a schematic diagram of a distribution position corresponding to a marking contour provided by an embodiment of the present invention as a position in a cavity; Figure 5 It is a schematic structural diagram of an aneurysm risk assessment device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0022] See also Figure 1 , is a schematic diagram of a method for risk assessment of an aneurysm provided by an embodiment of the present invention, Figure 1The execution subject of the method shown may be a software and / or hardware device. The execution subject of the present application may include but is not limited to at least one of the following: user equipment, network equipment, etc. Among them, user equipment may include but is not limited to computers, smart phones, personal digital assistants (PDA) and the electronic devices mentioned above. Network equipment may include but is not limited to a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers, wherein cloud computing is a type of distributed computing, a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. It includes steps S1 to S4, as follows: S1, acquiring image data corresponding to the patient end based on the risk assessment model, and determining a detection area in the image data.

[0023] It is understandable that if the coil is not packed tightly enough after aneurysm coil embolization, tiny channels or gaps may be left, allowing blood to still flow into the aneurysm, which may form blood clots inside the aneurysm and may cause the aneurysm to rupture. Therefore, a risk assessment of aneurysm rupture after surgery can promptly remind physicians to take appropriate treatment measures.

[0024] This solution will combine imaging data to evaluate the rupture of aneurysms after embolization, and provide doctors with corresponding reference data so that doctors can make corresponding treatments based on the obtained evaluation data.

[0025] Specifically, image data of angiographic examination of the patient's embolized aneurysm can be obtained, for example, DSA image data obtained through digital subtraction angiography. In the DSA image, the morphological changes of the aneurysm after embolization can be intuitively observed, such as the filling condition of the embolic material, changes in blood flow status, etc. The image area where the embolized aneurysm is located, i.e., the detection area, can be determined based on the DSA image data, and the detection area can be marked. Subsequently, the data in the detection area can be used for risk assessment and prediction through multiple dimensions.

[0026] The risk assessment model refers to a model for predicting rupture risk in combination with the imaging data of the aneurysm, the imaging data refers to the image data after angiography, and the detection area refers to the image area where the aneurysm after embolization is located.

[0027] S2, performing external risk analysis on the detection area according to the external detection mode to obtain external risk data corresponding to the detection area.

[0028] In actual applications, after aneurysm embolization, if the treatment effect is not good or there are other factors that promote aneurysm growth, the aneurysm may continue to grow, resulting in an increased risk of aneurysm rupture. By using imaging data, aneurysm images at different time periods can be analyzed to determine whether there is a trend of growth, and then predict its risk based on the growth trend.

[0029] Among them, the external detection mode refers to a mode of risk prediction through the area change trend of the detection area. When conducting external risk analysis, the area of ​​the detection area at different times can be obtained, and risk prediction can be carried out based on the area change trend at different times. The external risk data refers to the reference data obtained by risk prediction based on the area change trend of the detection area.

[0030] Based on the above embodiment, the specific implementation of step S2 may be: S21, acquiring a plurality of historical image data based on the external detection mode, and acquiring a historical detection area in each of the historical image data.

[0031] In actual applications, aneurysms may still rupture after embolization treatment. In order to monitor the embolization status of aneurysms in real time, multiple DSA examinations can be performed at different time periods after embolization to obtain historical imaging data corresponding to multiple different time periods. The imaging data can be used to provide physicians with imaging reference data of the embolized aneurysm at different time periods.

[0032] Since there may be multiple aneurysms, there may also be multiple areas corresponding to aneurysms in the image data. In order to distinguish different aneurysm areas, the image area corresponding to each aneurysm can be marked. For the same aneurysm area, the same marking data can be used, so that the area changes of each aneurysm area in different time periods can be determined through historical image data.

[0033] For example, on a DSA image, medical personnel can use specific annotation symbols to mark an aneurysm on the image, and during the marking process, the same aneurysm area can be marked with the same annotation symbol.

[0034] The historical image data refers to the image data corresponding to the patient at different time points in the past, for example, it may be DSA images at multiple different time points, and the historical detection area refers to the area corresponding to the detection area in the historical image data.

[0035] S22, arranging the historical detection areas in chronological order to obtain a detection sequence, and sequentially obtaining area differences between adjacent historical detection areas in the detection sequence.

[0036] When predicting the expansion trend of an aneurysm through image data, multiple image data may be arranged in chronological order, and whether the aneurysm has an expansion trend may be determined by the change in the area of ​​the region corresponding to the aneurysm in the image data in the arranged sequence.

[0037] Specifically, since the area of ​​the historical detection zone corresponding to different time points may be different, the area corresponding to each historical detection zone can be obtained, and the area corresponding to adjacent historical detection zones in the detection sequence are calculated in turn to obtain the area difference between adjacent historical detection zones. The changing trend of the historical detection zone can be judged by the area difference.

[0038] See also Figure 2 , is a schematic diagram of a detection sequence provided by an embodiment of the present invention, such as Figure 2 As shown in , in the historical image data, the aneurysm area in contact with the arterial blood vessel area is the historical detection area. It can be seen from the figure that, from left to right, the area of ​​the historical detection area in the second historical image data is larger than the area of ​​the historical detection area in the first historical image data, the third is larger than the second, and the fourth is larger than the third. Therefore, it can be seen that the area of ​​the historical detection area in the detection sequence shows an increasing trend.

[0039] The detection sequence refers to the sequence obtained by arranging the historical detection areas in chronological order, and the regional area difference refers to the area difference between historical detection areas adjacent in time.

[0040] S23, when the area difference of the detection number is greater than or equal to the external change threshold, determine that the change trend of the detection area is an increasing trend.

[0041] In practical applications, there may be errors in the area differences between adjacent detection areas. Therefore, the area change trend of the detection area may not be accurately judged by the area difference of one area. When the area differences of a number of detection areas are greater than or equal to the external change threshold, it means that the area of ​​a certain number of previous detection areas is greater than the area of ​​the next detection area. It can be considered that the area of ​​the detection area has an increasing trend. Therefore, the change trend of the detection area pair can be determined as an increasing trend.

[0042] Among them, the number of detections refers to the number of area differences that can be used to determine the trend of regional area changes. For example, when the number of detections is 3, it means that when the area differences between three adjacent detection areas are greater than or equal to the external change threshold, it can be considered that the area of ​​the detection area has a trend of increasing. The external change threshold refers to the maximum error value of the area difference under normal conditions.

[0043] S24, when the area difference of the region where there is no detection number is greater than or equal to the external change threshold, determine that the change trend of the detection area is a stable trend.

[0044] When there are no consecutive area differences greater than or equal to the external change threshold, it can be considered that the change between the area of ​​the detection area is not obvious, and its area may be in a stable state without a trend of expansion, so the change trend of the detection area pair can be determined as a stable trend. The stable trend refers to the trend that the area of ​​the detection area changes relatively steadily.

[0045] S25, determining the detection area of ​​the increasing trend as the external abnormal area, determining the detection area of ​​the stable trend as the external normal area, and obtaining external risk data according to the external abnormal area or the external normal area.

[0046] When the area change trend of the detection zone is an increasing trend, the risk of aneurysm rupture may increase, so the detection zone with an increasing trend can be determined as an external abnormal area. When the area change trend of the detection zone is a stable trend, it can be considered that the risk of aneurysm rupture is small, so the detection zone with a stable trend can be determined as an external normal area. External risk data can be obtained based on the external abnormal area and the external normal area.

[0047] The external abnormal area refers to an area where the area change trend of the aneurysm region is abnormal, and the external normal area refers to an area where the area change trend of the aneurysm region is normal.

[0048] Through the above implementation, the efficiency of data processing can be improved.

[0049] S3, obtaining the markers in the detection area, performing internal risk analysis on the markers according to the internal detection mode, and obtaining internal risk data corresponding to the detection area.

[0050] It is understandable that the aneurysm after embolization may cause blood to bypass the spring coil and continue to flow into the aneurysm due to the spring coil not being placed tightly enough or being in the wrong position, thereby forming a blood clot inside the aneurysm. When the blood clot in the aneurysm is too large, it may cause the aneurysm to rupture after embolization, and if the blood clot in the aneurysm contacts the aneurysm wall, it may cause a certain impact on the aneurysm wall, thereby causing the aneurysm to rupture. Therefore, the size and location of the blood clot may increase the risk of aneurysm rupture, and the different locations and sizes of blood clots have different effects on the risk of aneurysm rupture.

[0051] This solution can obtain the blood clot image, i.e., marker, inside the aneurysm based on the imaging data. Since there may be multiple blood clot images, there may be multiple markers. When the internal risk analysis of the marker is performed through the internal detection mode, the rupture risk of the aneurysm can be predicted based on the size of the marker and the position in the detection area to obtain the corresponding internal risk data.

[0052] When acquiring the marker, an iodine-containing contrast agent can be injected into the blood vessel through the angiography catheter. The contrast agent can make the blood vessel and its blood flow more clearly displayed, so that the morphology, position and relationship of the aneurysm area with the surrounding blood vessel area can be clearly displayed in the DSA angiography image, and the distribution of the blood clot image inside the aneurysm area can be obtained through the blood flow in the DSA angiography image.

[0053] Among them, the marker refers to the blood clot image inside the aneurysm area after embolization in the imaging data, the internal detection mode refers to the mode of risk prediction through the position and size of the marker inside the detection area, when conducting internal risk analysis, the risk of the detection area can be predicted according to the distribution of the marker in the detection area and the size of the area, and the internal risk data refers to the reference data obtained by risk prediction through the position and size of the marker inside the detection area.

[0054] Based on the above embodiment, the specific implementation of step S3 may be: S31, determining a pixel point in the detection area that is located in the marked pixel interval as a marked pixel point, and obtaining a marked contour corresponding to the marker according to adjacent marked pixel points.

[0055] Specifically, after contrast agent is injected, the pixel value corresponding to the area where the marker inside the aneurysm is located is different from the pixel values ​​corresponding to other areas. The pixel value interval where the pixel point corresponding to the marker is located can be pre-configured, that is, the marker pixel interval. After obtaining the pixel values ​​corresponding to multiple pixel points corresponding to the detection area, the pixel values ​​corresponding to the multiple pixel points are compared with the marker pixel interval respectively. If the pixel value corresponding to the pixel point is within the marker pixel interval, the corresponding pixel point can be considered to be the pixel point of the marker, that is, the marker pixel point. By obtaining multiple adjacent marker pixel points, the marker contour corresponding to the marker can be obtained.

[0056] The marked pixel value interval refers to the pixel value interval corresponding to the pixel points of the blood clot image in the image, and the marked contour refers to the contour corresponding to the marker.

[0057] S32, acquiring the distribution position of the marker contour, and determining an evaluation mode of the marker according to the distribution position of the marker, wherein the evaluation mode includes an attached evaluation mode and a free evaluation mode.

[0058] It is understandable that the distribution position of the blood clot image located inside the aneurysm area in the image may be different. The distribution position of the blood clot image is different, and the corresponding risk prediction method of the present scheme is also different. Specifically, the present scheme can divide the blood clot image into free type blood clot image and attached type blood clot image according to the different positions of the blood clot image, and can configure a corresponding attachment assessment mode for the attached type blood clot image, so that the risk of aneurysm rupture can be predicted by analyzing the blood clot image for the free type blood clot image.

[0059] Among them, a free type of blood clot image refers to a blood clot image whose distribution position is located in the aneurysm cavity and is not in direct contact with the aneurysm wall, an attached type of blood clot image refers to a blood clot image whose distribution position is located at the aneurysm wall or opening of the aneurysm and is in direct contact with the aneurysm wall, an attachment evaluation mode refers to a mode for evaluating attachment type blood clot images, and a free evaluation mode refers to a mode for evaluating free type blood clot images.

[0060] In some embodiments, the evaluation pattern of a marker may be determined by: S321, acquiring contour points connecting the detection area and the blood vessel area in the image data, and generating an opening contour according to adjacent contour points.

[0061] In practical applications, an aneurysm is formed due to local bulging of the arterial wall, so the aneurysm is connected to the arterial blood vessel, and therefore the detection area where the aneurysm is located in the image is connected to the arterial blood vessel area. Therefore, there are intersecting contour points between the pixel points corresponding to the detection area and the pixel points corresponding to the blood vessel area. The opening contour corresponding to the aneurysm opening area in the image can be determined through adjacent contour points. When determining the distribution position of the blood clot, the position of the blood clot in the aneurysm can be determined based on the distance between the blood clot and the opening contour.

[0062] The blood vessel region refers to the arterial blood vessel region including the aneurysm region in the image data, the contour point refers to the point where the pixel point of the detection area intersects with the pixel point of the blood vessel region, and the opening contour refers to the contour corresponding to the opening area of ​​the aneurysm region.

[0063] S322, determining that the distribution position corresponding to the mark contour in contact with the edge of the detection area is the tube wall position, obtaining the closest distance between the mark contour and the opening contour, and determining that the distribution position corresponding to the mark contour whose closest distance is less than the opening distance threshold is the opening position.

[0064] See also Figure 3 , which is a schematic diagram of a distribution position corresponding to a marking contour provided by an embodiment of the present invention as a pipe wall position, such as Figure 3As shown in , the marker contour is in contact with the edge of the detection area, and it can be considered that the blood clot image inside the aneurysm area is attached to the aneurysm wall contour of the aneurysm area. At this time, the distribution position corresponding to the marker contour can be determined as the tube wall position. It can be understood that in actual applications, when the hemodynamics in the aneurysm changes, such as a sharp increase in blood pressure, the blood clot attached to the aneurysm wall may cause a certain impact on the aneurysm wall, which may cause the aneurysm to rupture. Therefore, the image data can be used to determine whether the blood clot is attached to the aneurysm wall, and the corresponding rupture risk prediction can be combined with the image data.

[0065] In addition, since the aneurysm opening position is the key point for blood flow to enter the aneurysm cavity and is also the area where blood flow impact is most concentrated, when there is a blood clot at this position, the blood flow may be forced to change direction or speed, which may increase the pressure on the aneurysm wall and make it more likely to rupture. Therefore, the closest distance between the marked contour and the opening contour in the image data can be obtained. When the closest distance is less than the opening distance threshold, it means that the distance between the marked contour and the opening contour in the image data is too close. It can be considered that the blood clot image inside the aneurysm area is located at the opening area of ​​the aneurysm area. At this time, the distribution position of the marked contour in the image data can be determined to be the opening position, and when the closest distance between the marked contour and the opening contour is less than the opening distance threshold, even if the marked contour contacts the edge of the detection area, it can be considered that the blood clot image inside the aneurysm area is located at the opening area of ​​the aneurysm area. Therefore, it can also be determined that the distribution position corresponding to the marked contour is at the opening position.

[0066] The wall position refers to the position corresponding to the wall contour of the aneurysm region in the image data, and the opening position refers to the position where the aneurysm region is connected to the arterial blood vessel region in the image data.

[0067] S323, acquiring a distribution position corresponding to a marking contour that is not in contact with the edge of the detection area and whose closest distance is greater than or equal to the opening distance threshold as the intracavitary position.

[0068] See also Figure 4 , which is a schematic diagram of a distribution position corresponding to a marking contour provided by an embodiment of the present invention as a position in a cavity, such as Figure 4 As shown in , when the marker contour does not touch the edge of the detection area, and the closest distance between the marker contour and the opening contour is greater than or equal to the opening distance threshold, it can be explained that the distribution position of the marker in the graphic data is not located on the aneurysm wall contour of the aneurysm area, and is too far away from the opening position. Therefore, it can be considered that the distribution position of the marker in the image data is very likely to be in the intraluminal position of the aneurysm area. The intraluminal position refers to the position inside the aneurysm area in the image data.

[0069] S324, determining that the markers at the opening position and the tube wall position correspond to an attached evaluation mode, and determining that the markers at the intracavitary position correspond to a free evaluation mode.

[0070] Specifically, since in the image data, the distance between the marker located at the opening position and the aneurysm wall contour of the aneurysm area is very close, it can be considered to be attached to the aneurysm wall contour of the aneurysm area to a certain extent. Therefore, the evaluation mode corresponding to the marker located at the opening position and the tube wall position can be determined as the attached evaluation mode, and the evaluation mode corresponding to the marker located in the intracavitary position can be determined as the free evaluation mode.

[0071] S33, performing internal risk analysis on the marker according to the evaluation mode to obtain internal risk data corresponding to the detection area.

[0072] Specifically, after obtaining the evaluation mode corresponding to the marker, the density, size, and distribution position of the marker can be combined for analysis, and the risk of aneurysm rupture can be predicted through multiple dimensions, thereby providing doctors with more comprehensive evaluation reference data.

[0073] In some embodiments, step S33 includes S331 to S333, which are specifically as follows: S331, obtaining a type of evaluation data of the marker based on the attachment evaluation mode, wherein the type of evaluation data at least includes marker density, marker area, and contact area between the marker and the detection area, and determining the risk coefficient of the marker according to the type of evaluation data.

[0074] Uneven distribution of blood clot density may lead to uneven pressure distribution on the aneurysm wall. In some areas, when the blood clot density is high, the pressure may be relatively high, while in other areas with lower blood clot density, the pressure may be relatively low. This pressure difference may cause different parts of the aneurysm wall to bear different stresses, thereby increasing the risk of aneurysm rupture, and the size of the blood clot may also affect the stability of the aneurysm wall. For example, a large blood clot may compress the aneurysm wall and increase the risk of aneurysm rupture. In addition, when the contact area between the blood clot and the aneurysm wall increases, the blood clot may exert greater pressure on the aneurysm wall, resulting in an increase in local pressure on the aneurysm wall. This increase in pressure may make the aneurysm wall more susceptible to damage, thereby increasing the risk of rupture. Therefore, the density, area and contact area of ​​the marker with the detection area can be used as reference data for risk assessment.

[0075] Among them, marker density refers to density reference data related to the marker obtained by combining the pixel value of the blood clot image in the image data, marker area refers to the area of ​​the blood clot image in the image data, contact area refers to the contact area between the blood clot image in the image data and the aneurysm wall contour of the aneurysm area, a type of evaluation data refers to the parameters related to risk prediction obtained through the marker in the attachment evaluation mode, and the risk coefficient refers to the coefficient for risk prediction combined with a type of evaluation data. The larger the risk coefficient, the greater the possibility of aneurysm rupture.

[0076] Based on the above embodiment, the specific implementation of step S331 may be: S3311, obtaining a pixel mean value corresponding to the marking contour of the marker, and traversing multiple pixel intervals based on the pixel mean value.

[0077] In the DSA angiography image, the pixel value of the marker can represent the amount of X-ray transmission at the position corresponding to the marker. Since the amount of X-ray transmission is directly related to the density of the marker, the pixel value of the marker can indirectly reflect the density of the marker. For example, the density corresponding to a darker area may be higher, while the density corresponding to a brighter area may be lower.

[0078] When determining the density corresponding to the marker, the present solution may pre-configure multiple pixel value intervals, each interval may be configured with a corresponding preset density, and the density corresponding to the marker may be determined based on the pixel mean corresponding to the marker contour in the DSA angiography image.

[0079] Specifically, the marking contour of the marker corresponds to multiple pixel points. By calculating the mean of the pixel values ​​corresponding to the multiple pixel points, the pixel mean corresponding to the marking contour can be obtained. By traversing multiple pixel intervals according to the pixel mean, it can be determined in which pixel interval the pixel mean is located. Subsequently, the density corresponding to the marker can be determined according to the pixel interval.

[0080] S3312, determining that the preset density corresponding to the pixel interval where the pixel mean value is located is the marker density, and obtaining a density assessment coefficient according to the ratio of the marker density to the reference density.

[0081] After traversing multiple pixel intervals according to the pixel mean, the pixel interval where the pixel mean is located can be determined, and then the corresponding preset density can be determined. The preset density corresponding to the pixel mean can be determined as the marker density. By calculating the ratio of the marker density to the reference density, the density assessment coefficient corresponding to the marker can be obtained. The larger the marker density, the larger the density assessment coefficient.

[0082] Among them, the preset density refers to the pre-configured density corresponding to the pixel interval, the reference density refers to the pre-set reference density related to the density, and the density assessment coefficient refers to the coefficient obtained by assessing the rupture risk in combination with the density data.

[0083] S3313, counting the number of pixels of the marker outline, determining the marker area corresponding to the marker according to the number of pixels, and obtaining an area evaluation coefficient according to the ratio of the marker area to the total area of ​​the detection zone.

[0084] It can be understood that in image processing, since the image is composed of discrete pixels, each pixel represents a minimum unit area in the image, so the area can be determined according to the number of pixels. The larger the number of pixels, the larger the corresponding area. Specifically, the pixels corresponding to the marker contour can be obtained, and the total number of pixels corresponding to the marker contour in the angiography image can be determined. The marker area corresponding to the marker can be obtained through the number of pixels. After obtaining the marker area, the ratio can be calculated with the total area of ​​the detection zone. The larger the ratio, the greater the proportion of the marker in the corresponding detection zone, and the larger the calculated area assessment coefficient, and the greater the corresponding aneurysm rupture risk.

[0085] The total region area refers to the total area corresponding to the entire detection area, and the area assessment coefficient refers to the coefficient obtained by assessing the risk of aneurysm rupture based on the area of ​​the marker.

[0086] S3314, obtaining the number of intersecting pixels between the mark outline and the edge of the detection area, obtaining the contact area according to the number of intersecting pixels, and obtaining a contact surface evaluation coefficient based on the ratio of the contact area to a reference contact area.

[0087] Specifically, the pixel points where the marked outline and the edge of the detection area intersect can be obtained. The total number of pixel points where the marked outline and the edge of the detection area intersect in the contrast image can be determined. The number of intersecting pixel points can indirectly represent the contact area of ​​the portion where the marked outline and the edge of the detection area in the image are in contact. After obtaining the contact area, it can be compared with the reference contact area, so that the contact surface evaluation coefficient can be calculated.

[0088] Among them, the number of intersecting pixels refers to the number of pixels where the marking contour and the edge of the detection area intersect, the reference contact area refers to a pre-set reference area related to the contact area, and the contact surface evaluation coefficient refers to the coefficient obtained by evaluating the rupture risk in combination with the contact area data.

[0089] S3315, obtaining the risk coefficient according to the sum of the density assessment coefficient, the area assessment coefficient and the contact surface assessment coefficient.

[0090] In some embodiments, the various coefficients for assessing the risk of aneurysm rupture may have different degrees of influence. For example, when the area of ​​the marker is the main factor causing aneurysm rupture, and the density and contact area are secondary factors causing aneurysm rupture, the area assessment coefficient has a higher degree of influence, and the density assessment coefficient and the contact surface assessment coefficient have second highest degree of influence. Therefore, different weights can be assigned to different assessment coefficients, with higher weights assigned to coefficients with higher degrees of influence and lower weights assigned to coefficients with lower degrees of influence, so that the weights of various reference factors for assessing the risk of aneurysm rupture can be adjusted.

[0091] Through the above implementation, the risk of aneurysm rupture can be predicted through data in multiple dimensions, thereby providing physicians with more accurate evaluation reference data.

[0092] S332, or, obtaining second-category evaluation data of the marker according to the free evaluation mode, wherein the second-category evaluation data at least includes the area of ​​the marker, and determining the risk coefficient of the marker according to the second-category evaluation data.

[0093] In practical applications, when the blood clot is located in the lumen of the aneurysm and is too large, the blood clot occupies space in the lumen of the aneurysm, which will exert direct physical pressure on the aneurysm wall. This pressure may make the aneurysm wall weaker, thereby increasing its risk of rupture. When the blood clot gradually increases, the pressure on the aneurysm wall may gradually accumulate, and eventually the aneurysm wall cannot withstand and ruptures. Therefore, when combining the blood clot image located in the lumen to predict the risk, the rupture risk can be predicted according to the area of ​​the marker in the image data. Specifically, the larger the area of ​​the marker in the free state, the higher the rupture risk of the aneurysm may be. According to the free evaluation mode, the influence of the area size of the marker in the image data on the rupture risk of the aneurysm can be analyzed, and the second type of evaluation data corresponding to the marker located in the lumen can be obtained. Since the area of ​​the marker in the aneurysm lumen is more important to the rupture risk caused by the aneurysm, the second type of evaluation data must at least include the evaluation data corresponding to the area of ​​the marker. According to the second type of evaluation coefficient, the risk coefficient corresponding to the rupture risk caused by the marker to the aneurysm can be determined.

[0094] Among them, the second type of evaluation data refers to the reference data for evaluating the rupture risk of the marker based on the free evaluation mode.

[0095] Based on the above embodiment, the specific implementation of step S332 may be: S3321, counting the number of pixels corresponding to the marking contour of the marker, and determining the area of ​​the marker according to the number of pixels.

[0096] Specifically, the pixel points corresponding to the marker contour can be obtained, and the total number of the pixel points corresponding to the marker contour in the contrast image can be determined by traversing the pixel points, and the marker area corresponding to the marker can be determined by the number of pixel points.

[0097] S3322, obtaining a risk coefficient corresponding to a marker whose marker area is smaller than an abnormal area threshold as a preset low risk coefficient.

[0098] It is understandable that since there may still be tiny gaps inside the aneurysm after embolization, blood clots may form in the tiny gaps. The size of the blood clots varies, and the risk of aneurysm rupture is also different. Specifically, when the marker area is smaller than the abnormal area threshold, it can be considered that the marker is small, resulting in a smaller possibility of aneurysm rupture. Therefore, the preset low risk factor can be determined as the risk factor corresponding to the marker.

[0099] The abnormal area threshold refers to the preset minimum value of the marker area when abnormal, and the low risk coefficient refers to the coefficient corresponding to the low rupture risk of the aneurysm caused by the preset reference factors.

[0100] S3323, determining that the marker whose area is greater than or equal to the abnormal area threshold is an abnormal object, counting the abnormal areas of all abnormal objects, and obtaining a first abnormal coefficient according to the ratio of the abnormal area to the total area of ​​the detection zone.

[0101] When the area of ​​the marker is greater than or equal to the abnormal area threshold, the marker can be considered abnormal. At this time, the corresponding marker can be determined as an abnormal object. Since there may be multiple free abnormal objects in the detection area, the sum of the areas of all abnormal objects can be counted to obtain the corresponding abnormal area. The ratio of the abnormal area to the total area of ​​the detection area can be calculated to obtain the corresponding first abnormal coefficient. The larger the abnormal area, the greater the proportion of the abnormal area may be, and the larger the corresponding first abnormal coefficient is, the greater the risk of aneurysm rupture.

[0102] Among them, an abnormal object refers to a marker whose area is greater than the abnormal area threshold, the abnormal area refers to the sum of the areas corresponding to all abnormal objects, and the first abnormal coefficient refers to the relevant risk coefficient obtained by combining the area reference data.

[0103] S3324, acquiring a plurality of historical image data, and determining the historical anomalies in each of the historical image data that correspond to the free evaluation mode and whose marked area is greater than or equal to the anomaly area threshold.

[0104] It is understandable that when the shape of the blood clot located in the cavity is unstable, for example, it may continue to grow, resulting in increased pressure on the aneurysm wall, which may further cause the aneurysm to rupture. Therefore, in order to determine whether there is an unstable marker in the detection area, multiple historical anomalies can be obtained from multiple historical image data. Subsequently, it can be determined whether there are unstable blood clots in the multiple historical anomalies based on the area changes of the historical anomalies, and the risk of aneurysm rupture can be predicted, wherein the historical anomaly refers to an anomaly in the historical image data that is in a free state and the marker area is greater than or equal to the abnormal area threshold.

[0105] S3325, counting the historical abnormal areas corresponding to the historical abnormal objects in each of the historical image data, and obtaining abnormal differences between adjacent historical abnormal areas in chronological order.

[0106] It is understandable that since the area size of historical anomalies may be unstable and may change at any time, the area corresponding to the historical anomalies in the historical image data corresponding to different time points may be different. The historical abnormal area corresponding to the historical anomaly in each historical image data can be obtained. In order to obtain the changing trend of the historical abnormal area corresponding to the historical anomaly and predict the rupture risk of the aneurysm based on the changing trend of the historical abnormal area, multiple historical image data can be arranged in chronological order, and the historical abnormal areas corresponding to the historical anomalies in adjacent historical image data are calculated in turn for difference, so as to obtain the abnormal difference between adjacent historical abnormal areas.

[0107] Among them, the historical anomaly area refers to the total area corresponding to the historical anomalies, and the anomaly difference refers to the difference between adjacent historical anomaly areas.

[0108] S3326, obtaining the number of abnormalities corresponding to the abnormal difference value greater than the abnormal difference threshold, obtaining a second abnormality coefficient according to the ratio of the abnormal number to the benchmark abnormal number, and obtaining the risk coefficient according to the sum of the first abnormality coefficient and the second abnormality coefficient.

[0109] When the abnormal difference is greater than the abnormal difference threshold, it can be considered that the previous historical abnormal area is greater than the next historical abnormal area. The number of abnormal differences greater than the abnormal threshold can be counted. The abnormal difference can indicate that the difference between adjacent historical abnormal areas has a tendency to expand. The more abnormalities there are, the stronger the expansion trend between adjacent historical abnormal areas can be considered, and thus the more obvious the trend of increase in historical abnormal areas can be considered. The corresponding second abnormal coefficient can be obtained by calculating the ratio of the number of abnormalities to the benchmark abnormal number. The more abnormalities there are, the larger the second abnormal coefficient is. In addition, the risk coefficient corresponding to the marker can be obtained by adding the first abnormal coefficient to the second abnormal coefficient.

[0110] Among them, the abnormal difference threshold refers to the maximum area difference under normal conditions, the abnormal number refers to the number of abnormal differences counted, the benchmark abnormal number refers to the pre-set standard abnormal number, and the second abnormal coefficient refers to the relevant risk data obtained by combining the area changes of historical abnormal objects.

[0111] S333, based on the risk coefficient of each marker in the detection area, perform risk assessment on the detection area to obtain an internal abnormal area or an internal normal area, and obtain internal risk data based on the internal abnormal area or the internal normal area.

[0112] Specifically, different markers in the detection area correspond to different risk factors. Some markers have lower corresponding risk factors, which will hardly cause aneurysm rupture. Some markers have higher corresponding risk factors, which are very likely to cause aneurysm rupture. Therefore, after obtaining the risk coefficients corresponding to each marker in the detection area, the risk of the detection area can be determined, and it can be determined whether the corresponding detection area is an internal abnormal area or an internal normal area, and then the corresponding internal risk data can be obtained.

[0113] Among them, the internal abnormal area refers to the detection area where the risk coefficient of the marker is in an abnormal state, and the risk of aneurysm rupture corresponding to the internal abnormal area is relatively high. The internal normal area refers to the detection area where the risk coefficients corresponding to all markers are in a normal state, and the risk of aneurysm rupture corresponding to the internal normal area is relatively low.

[0114] In some embodiments, the step S333 of “based on the risk coefficient of each marker in the detection area, performing risk determination on the detection area to obtain an internal abnormal area or an internal normal area” includes the following steps: S3331, retrieve the corresponding preset risk threshold according to the distribution position corresponding to each of the markers, and determine that the detection area where the risk coefficient corresponding to each of the markers is less than the preset risk threshold is the internal normal area.

[0115] It can be understood that, within the detection area, the risk levels of aneurysm rupture corresponding to markers located at different distribution positions are different. Different risk thresholds can be pre-configured for different distribution positions. According to the distribution position of the marker, the corresponding preset risk threshold is retrieved, and the risk coefficient corresponding to the marker is compared with the retrieved preset risk threshold. If the risk coefficient of the marker is less than the preset risk threshold at the corresponding distribution position, it can be considered that the risk coefficient corresponding to the marker is in a normal state. When the risk coefficients corresponding to all markers in the detection area are less than the preset risk threshold corresponding to the distribution position of the marker, it can be considered that the risk coefficients corresponding to all markers in the detection area are in a normal state. Therefore, the corresponding detection area can be determined as an internal normal area.

[0116] Among them, the preset risk threshold refers to the threshold of the risk coefficient under normal conditions.

[0117] S3332, determining that the detection area where the risk coefficient corresponding to the marker is greater than or equal to the preset risk threshold is an internal abnormal area.

[0118] When the risk coefficient corresponding to the marker in the detection area is greater than or equal to the preset risk threshold, it can be considered that there is a marker in the detection area that increases the risk of aneurysm rupture. At this time, the detection area corresponding to the marker can be determined as an internal abnormal area.

[0119] S3333, obtaining a risk coefficient that is greater than or equal to the preset risk threshold as an abnormality coefficient, and counting the total abnormal difference between each abnormality coefficient and the corresponding preset risk threshold.

[0120] After obtaining the internal abnormal area, when the risk coefficient corresponding to the internal abnormal area is greater than the preset risk threshold, the risk coefficient corresponding to the internal abnormal area can be determined as the abnormal coefficient. If the difference between the abnormal coefficient and the preset risk coefficient is greater, it means that the abnormal degree of the internal abnormal area is greater. The difference between each abnormal coefficient and the preset risk threshold can be calculated. Since there may be multiple abnormal coefficients, the difference between multiple abnormal coefficients and the preset risk threshold can be obtained. The total abnormal difference corresponding to the internal abnormal area can be obtained by adding the multiple differences. The larger the total abnormal difference, the higher the rupture risk degree corresponding to the internal abnormal area. Among them, the abnormal coefficient refers to the risk coefficient greater than or equal to the preset risk threshold, and the total abnormal difference refers to the sum of the differences between multiple abnormal coefficients and the preset risk threshold.

[0121] S3334, traverse multiple abnormal range intervals according to the abnormal total difference value, and determine that the abnormal level corresponding to the abnormal range interval where the abnormal total difference value is located is the internal risk level corresponding to the internal abnormal area.

[0122] It can be understood that the larger the total abnormal difference is, the higher the rupture risk degree corresponding to the internal abnormal area is. Therefore, different total abnormal difference values ​​correspond to different rupture risk degrees. This scheme can pre-configure multiple abnormal range intervals, and each abnormal range interval can be configured with a corresponding abnormal level. By traversing multiple abnormal range intervals according to the total abnormal difference, it can be determined in which abnormal range interval the total abnormal difference corresponding to the internal abnormal area is, and then the corresponding abnormal level can be determined, and the corresponding abnormal level can be used as the risk level corresponding to the internal abnormal area.

[0123] Among them, the abnormal range interval refers to the range interval corresponding to the abnormal total difference, the abnormal level refers to the risk level corresponding to the abnormal range interval, and the internal risk level can be used to indicate the rupture risk degree of the aneurysm corresponding to the internal abnormal area. The higher the internal risk level, the greater the rupture risk.

[0124] S4, determining risk assessment data of the detection area according to the external risk data and the internal risk data, and sending the risk assessment data to the physician end.

[0125] The risk assessment data refers to reference data obtained after evaluating the rupture risk corresponding to the detection area. After the risk assessment data is obtained, the risk assessment data can be sent to the physician end, thereby providing the physician with assessment reference data.

[0126] Based on the above embodiment, the specific implementation of step S4 may be: S41, determining that the detection area whose external risk data is the external normal area and whose internal risk data is the internal normal area is a normal area.

[0127] Specifically, when the external risk data corresponding to the detection area is an external normal area, and the corresponding internal risk data is an internal normal area, it can be considered that the rupture risk of the aneurysm corresponding to the detection area is low, and the detection area can be determined as a normal area. The normal area refers to the detection area corresponding to the aneurysm with a low rupture risk.

[0128] S42, obtaining a detection area where the external risk data is the external abnormal area and / or the internal risk data is the internal abnormal area as an abnormal area.

[0129] Among them, the abnormal area refers to a detection area where risk data is an abnormal area, for example, it can be a detection area where external risk data is an external abnormal area or a detection area where internal risk data is an internal abnormal area, or it can be a detection area where external risk data is an external abnormal area and internal risk data is an internal abnormal area.

[0130] S43, retrieve the abnormal icons corresponding to the external abnormal area and / or the internal abnormal area, and update the abnormal slots corresponding to the detection area in the image data according to the abnormal icons to obtain risk assessment data.

[0131] In actual applications, a series of exception icons can be pre-configured, and the configured exception icons can correspond to the exception types. The exception types in this scheme include external exceptions and internal exceptions. The exception type of the exception area can be displayed by the exception icon, and when configuring the exception icon corresponding to the internal type, exception icons with different pixel values ​​can be configured for different internal risk levels. For example, when the internal risk level is low, a blue exception icon can be configured, and when the internal risk level is high, a red exception icon can be configured.

[0132] Specifically, after determining the abnormal area, the abnormal icon corresponding to the external abnormal area and the abnormal icon corresponding to the internal abnormal area can be retrieved, and the abnormal slot corresponding to the abnormal area in the image data can be updated according to the abnormal icon, and the risk assessment data corresponding to the detection area can be obtained. The abnormal type corresponding to the abnormal area can be quickly determined through the abnormal icon, and the doctor can subsequently take corresponding treatment measures according to the abnormal type.

[0133] Among them, the exception icon refers to an icon that can be used to intuitively display the exception type, and the exception slot refers to a slot in the exception area used to display the exception icon.

[0134] Through the above implementation, the abnormal type corresponding to the abnormal area can be quickly determined through the abnormal icon, so that the doctor can take corresponding treatment measures according to the abnormal type, and can provide the doctor with more comprehensive and intuitive reference data.

[0135] See also Figure 5 , is a schematic diagram of the structure of an aneurysm risk assessment device provided by an embodiment of the present invention. The data processing system based on the aneurysm risk assessment device includes: A determination module, configured to obtain image data corresponding to the patient end based on the risk assessment model, and determine a detection area in the image data; An analysis module, used for performing external risk analysis on the detection area according to the external detection mode to obtain external risk data corresponding to the detection area; an acquisition module, used to acquire the markers in the detection area, perform internal risk analysis on the markers according to the internal detection mode, and obtain internal risk data corresponding to the detection area; The sending module is used to determine the risk assessment data of the detection area according to the external risk data and the internal risk data, and send the risk assessment data to the physician end.

[0136] Figure 5 The apparatus of the embodiment shown can be used to perform Figure 1 The implementation principles and technical effects of the steps in the method embodiment shown are similar and will not be repeated here.

[0137] Finally, it should be noted that 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for assessing the risk of an aneurysm, characterized in that: include: Acquire image data corresponding to the patient end based on the risk assessment model, and determine a detection area in the image data; Performing external risk analysis on the detection area according to the external detection mode to obtain external risk data corresponding to the detection area; Acquire the marker in the detection area, perform internal risk analysis on the marker according to the internal detection mode, and obtain internal risk data corresponding to the detection area; The risk assessment data of the detection area is determined according to the external risk data and the internal risk data, and the risk assessment data is sent to the physician end.

2. The method according to claim 1, characterized in that Performing external risk analysis on the detection area according to the external detection mode to obtain external risk data corresponding to the detection area includes: Acquire a plurality of historical image data based on the external detection mode, and acquire a historical detection area in each of the historical image data; Arrange the historical detection areas in chronological order to obtain a detection sequence, and sequentially obtain area differences between adjacent historical detection areas in the detection sequence; When the area difference of the detected number of regions is greater than or equal to the external change threshold, determining that the change trend of the detection area is an increasing trend; When the area difference of the region where there is no detection number is greater than or equal to the external change threshold, determining that the change trend of the detection area is a stable trend; The detection area of ​​the increasing trend is determined as the external abnormal area, and the detection area of ​​the stable trend is determined as the external normal area, and external risk data is obtained according to the external abnormal area or the external normal area.

3. The method according to claim 1, characterized in that Acquiring the marker in the detection area, performing internal risk analysis on the marker according to the internal detection mode, and obtaining internal risk data corresponding to the detection area, including: Determine a pixel point in the detection area that is within the marked pixel interval as a marked pixel point, and obtain a marked contour corresponding to the marker according to adjacent marked pixel points; Acquiring the distribution position of the marker contour, and determining an evaluation mode of the marker according to the distribution position of the marker, wherein the evaluation mode includes an attached evaluation mode and a free evaluation mode; An internal risk analysis is performed on the marker according to the evaluation mode to obtain internal risk data corresponding to the detection area.

4. The method according to claim 3, characterized in that The distribution position of the marker contour is obtained, and an evaluation mode of the marker is determined according to the distribution position of the marker, wherein the evaluation mode includes an attached evaluation mode and a free evaluation mode, including: Acquire contour points connecting the detection area and the blood vessel area in the image data, and generate an opening contour according to adjacent contour points; Determine the distribution position corresponding to the mark contour in contact with the edge of the detection area as the tube wall position, obtain the closest distance between the mark contour and the opening contour, and determine the distribution position corresponding to the mark contour whose closest distance is less than the opening distance threshold as the opening position; Acquire the distribution position corresponding to the marking contour that is not in contact with the edge of the detection area and whose closest distance is greater than or equal to the opening distance threshold as the intracavitary position; The markers that determine the opening position and edge position correspond to the attachment evaluation mode, and the markers that determine the intracavitary position correspond to the free evaluation mode.

5. The method according to claim 4, characterized in that Performing internal risk analysis on the marker according to the evaluation mode to obtain internal risk data corresponding to the detection area includes: Based on the attachment evaluation mode, a type of evaluation data of the marker is obtained, wherein the type of evaluation data at least includes the marker density, the marker area, and the contact area between the marker and the detection area, and the risk coefficient of the marker is determined according to the type of evaluation data; or Acquire the second type of evaluation data of the marker according to the free evaluation mode, wherein the second type of evaluation data at least includes the area of ​​the marker, and determine the risk coefficient of the marker according to the second type of evaluation data; Based on the risk coefficient of each marker in the detection area, the detection area is subjected to risk assessment to obtain an internal abnormal area or an internal normal area, and internal risk data is obtained based on the internal abnormal area or the internal normal area.

6. The method according to claim 5, characterized in that Acquiring a type of evaluation data of the marker based on the attachment evaluation mode, the type of evaluation data at least including the marker density, the marker area, and the contact area between the marker and the detection area, and determining the risk coefficient of the marker according to the type of evaluation data, including: Obtaining a pixel mean value corresponding to the marking contour of the marker, and traversing multiple pixel intervals based on the pixel mean value; Determine a preset density corresponding to a pixel interval where the pixel mean value is located as the marker density, and obtain a density assessment coefficient according to a ratio of the marker density to a reference density; Counting the number of pixels of the marker contour, determining the marker area corresponding to the marker according to the number of pixels, and obtaining an area evaluation coefficient according to the ratio of the marker area to the total area of ​​the detection zone; Obtaining the number of intersecting pixels between the marker contour and the edge of the detection area, obtaining the contact area according to the number of intersecting pixels, and obtaining a contact surface evaluation coefficient based on a ratio of the contact area to a reference contact area; The risk coefficient is obtained according to the sum of the density assessment coefficient, the area assessment coefficient and the contact surface assessment coefficient.

7. The method according to claim 5, characterized in that Acquiring second-category evaluation data of the marker according to the free evaluation mode, wherein the second-category evaluation data at least includes the area of ​​the marker, and determining the risk coefficient of the marker according to the second-category evaluation data, including: Counting the number of pixels corresponding to the marking contour of the marker, and determining the area of ​​the marker according to the number of pixels; The risk coefficient corresponding to the marker whose area is smaller than the abnormal area threshold is obtained as a preset low risk coefficient; Determine that the marker whose area is greater than or equal to the abnormal area threshold is an abnormal object, count the abnormal areas of all abnormal objects, and obtain a first abnormal coefficient according to the ratio of the abnormal area to the total area of ​​the detection area; Acquire a plurality of historical image data, and determine a historical anomaly corresponding to a free evaluation mode and having a marked object area greater than or equal to an abnormal area threshold in each of the historical image data; Counting the historical abnormal areas corresponding to the historical abnormal objects in each of the historical image data, and obtaining the abnormal differences between adjacent historical abnormal areas in chronological order; The number of abnormalities corresponding to the abnormal difference value greater than the abnormal difference threshold is obtained, a second abnormality coefficient is obtained according to the ratio of the abnormal number to the benchmark abnormal number, and the risk coefficient is obtained according to the sum of the first abnormality coefficient and the second abnormality coefficient.

8. The method according to claim 5, characterized in that Based on the risk coefficient of each marker in the detection area, the detection area is subjected to risk determination to obtain an internal abnormal area or an internal normal area, including: According to the distribution position corresponding to each of the markers, a corresponding preset risk threshold is retrieved, and a detection area where the risk coefficient corresponding to each of the markers is less than the preset risk threshold is determined as an internal normal area; Determine the detection area where the risk coefficient corresponding to the marker is greater than or equal to the preset risk threshold as an internal abnormal area; The risk coefficient greater than or equal to the preset risk threshold is obtained as the abnormal coefficient, and the total abnormal difference between each abnormal coefficient and the corresponding preset risk threshold is counted; A plurality of abnormal range intervals are traversed according to the abnormal total difference value, and it is determined that the abnormal level corresponding to the abnormal range interval where the abnormal total difference value is located is the internal risk level corresponding to the internal abnormal area.

9. The method according to claim 2 or 5, characterized in that: Determining risk assessment data of the detection area according to the external risk data and the internal risk data, and sending the risk assessment data to a physician end, comprises: Determine that the detection area where the external risk data is the external normal area and the internal risk data is the internal normal area is a normal area; Acquire the detection area where the external risk data is the external abnormal area and / or the internal risk data is the internal abnormal area as the abnormal area; The abnormal icons corresponding to the external abnormal area and / or the internal abnormal area are retrieved, and the abnormal slots corresponding to the detection area in the image data are updated according to the abnormal icons to obtain risk assessment data.

10. An aneurysm risk assessment device, characterized in that: include: A determination module, configured to obtain image data corresponding to the patient end based on the risk assessment model, and determine a detection area in the image data; An analysis module, used for performing external risk analysis on the detection area according to the external detection mode to obtain external risk data corresponding to the detection area; an acquisition module, used to acquire the markers in the detection area, perform internal risk analysis on the markers according to the internal detection mode, and obtain internal risk data corresponding to the detection area; The sending module is used to determine the risk assessment data of the detection area according to the external risk data and the internal risk data, and send the risk assessment data to the physician end.

Images