Method and device for marking target blood vessel based on intracavity image and angiography image

Abstract

The invention discloses a method and device for marking a target blood vessel based on an intracavity image and an angiography image, and the method for marking the target blood vessel comprises the steps: obtaining an angiography image containing the target blood vessel, obtaining the intracavity image of the target blood vessel, collecting a plurality of cross-sectional images of the target blood vessel based on the intracavity image to obtain a first cross-sectional image set, selecting a target cross section image meeting a preset condition from the first cross section image set and forming a target cross section image set, registering the first cross section image set based on the angiography image to obtain a registration mapping set, and obtaining a target position set based on the target cross section image set and the registration mapping set, and marking the angiography image based on the target position set to obtain the contour and position of the target blood vessel in the angiography image. Therefore, the method for marking the target blood vessel can simplify the calculation process, improve the registration efficiency and improve the registration precision.

Description

Technical Field

[0001] The present disclosure generally relates to the field of medical image processing, and more particularly to a method and apparatus for marking a target blood vessel based on an intracavity image and an angiography image. Background Art

[0002] Intravascular ultrasound (IVUS) instruments can obtain cross-sectional image information of blood vessels (such as coronary arteries) by retracting the ultrasound probe in the blood vessels. However, for target blood vessels (such as narrowed blood vessels), they are generally continuous and have length. Therefore, the operator also needs to combine the playback function and labeling function of the IVUS instrument to better identify the specific length of the target blood vessel, which increases the workload of the operator to integrate information and is not conducive to quickly identifying the specific length of the target blood vessel.

[0003] Digital Subtraction Angiography (DSA) is a technology that uses X-rays to image blood vessels. DSA technology can obtain relatively clear blood vessel contour information (also called angiographic images). Although IVUS and DSA technology can obtain the correspondence between each position in the angiographic image and the cross-sectional image of the blood vessel. However, since the IVUS instrument does not mark the specific length of the target blood vessel on the angiographic image, the operator also needs to do a lot of work to integrate information before finally confirming the specific length and specific position of the target blood vessel on the angiographic image.

[0004] Patent document 1 (Chinese Patent Publication No. CN115359100A) discloses a fusion angiography method, an intracavity image and an angiography image registration method and device, wherein the fusion angiography method comprises: acquiring an angiography image and an intracavity image; based on the correspondence between the contrast marker and the intracavity marker, fusing the contrast image and the intracavity image to obtain a fused angiography result; Patent document 1 links the contrast image and the intracavity image through the contrast marker and the intracavity marker, so that the intracavity image information and the contrast image information can be corresponded and fused to obtain a fused angiography result, thereby overcoming the problem that the traditional angiography image contains a single information.

[0005] However, the method for registering the intracavitary image and the contrast image involved in Patent Document 1 requires the use of an iterative closest point algorithm (ICP), semantic segmentation recognition, and obtaining the contrast image and the cross-sectional image based on the backtest speed and time of the ultrasound probe. The calculation process is complex and the amount of calculation is large. At the same time, these methods require a large number of types and quantities of data. Summary of the invention

[0006] The present disclosure is proposed in view of the above situation, and aims to provide a method and device for marking target blood vessels based on ultrasound images and angiography images, which can simplify the calculation process, improve the registration efficiency and enhance the registration accuracy.

[0007] To this end, the first aspect of the present disclosure provides a method for marking a target blood vessel based on an intracavitary image and an angiography image, which is a method for marking the contour and position of a target blood vessel in an angiography image, comprising: acquiring an angiography image containing a target blood vessel, acquiring an intracavitary image of the target blood vessel, acquiring multiple cross-sectional images of the target blood vessel based on the intracavitary image to obtain a first cross-sectional image set, selecting target cross-sectional images that meet preset conditions from the first cross-sectional image set and forming a target cross-sectional image set, registering the first cross-sectional image set based on the angiography image to obtain a registration mapping set, acquiring a target position set based on the target cross-sectional image set and the registration mapping set, and marking the angiography image based on the target position set to obtain the contour and position of the target blood vessel in the angiography image.

[0008] In the method for marking a target blood vessel involved in the first aspect of the present disclosure, by registering the angiographic image and the first cross-sectional image set, each position of the target blood vessel in the angiographic image can form a one-to-one mapping relationship with the cross-sectional image of each frame number in the first cross-sectional image set, so that after selecting the target cross-sectional image set that meets the preset conditions, the position information of the corresponding target blood vessel can be locked in the angiographic image more accurately, and the contour and position of the target blood vessel can be marked more accurately in the target blood vessel of the angiographic image using the obtained position information. As a result, the target blood vessel can be displayed intuitively and clearly in the target blood vessel of the angiographic image, so that it is convenient for operators to observe and analyze.

[0009] In addition, in the method for marking a target blood vessel involved in the first aspect of the present disclosure, optionally, the step of registering the first cross-sectional image set based on the angiography image to obtain a registration mapping set includes: marking the starting position and the ending position of the target blood vessel in the angiography image and the intracavitary image, obtaining a first blood vessel image feature set for the angiography image from the starting position to the ending position; obtaining a second blood vessel image feature set for the first cross-sectional image set from the starting position to the ending position; and obtaining the registration mapping set based on the intersection of the first blood vessel image feature set and the second blood vessel image feature set. In this way, the registration mapping set can be obtained more accurately and quickly.

[0010] In addition, in the method for marking a target blood vessel involved in the first aspect of the present disclosure, optionally, the first blood vessel image feature set includes a first blood vessel diameter set in the angiography image, the second blood vessel image feature set includes a second blood vessel diameter set in the first cross-sectional image set, and the registration mapping set is acquired based on the intersection of the first blood vessel diameter set and the second blood vessel diameter set; and / or the first blood vessel image feature set includes a first blood vessel branch diameter set in the angiography image, the second blood vessel image feature set includes a second blood vessel branch diameter set in the first cross-sectional image set, and the registration mapping set is acquired based on the intersection of the first blood vessel branch diameter set and the second blood vessel branch diameter set; and / or the first blood vessel image feature set includes a first blood vessel branch position set in the angiography image, the second blood vessel image feature set includes a second blood vessel branch position set in the first cross-sectional image set, and the registration mapping set is acquired based on the intersection of the first blood vessel branch position set and the second blood vessel branch position set. In this case, when registering the angiography image and the first cross-sectional image set, since the vascular diameter is a vascular parameter that is relatively easy to obtain from the angiography image and the cross-sectional image, and is also a relatively obvious feature shared by the angiography image and the cross-sectional image, the vascular diameter can be used to quickly and accurately link each position in the angiography image with each cross-sectional image frame, that is, the angiography image and the cross-sectional image are associated with the vascular diameter as a link. At the same time, through the intersection method, each position in the angiography image can be linked with each cross-sectional image frame (that is, the mapping relationship between each position in the angiography image and each cross-sectional image frame is obtained), which can simplify the calculation process, reduce the difficulty of registration, and thus improve the efficiency of registration. In this way, the registration mapping set can be obtained relatively accurately and quickly. In addition, since the vascular branch has relatively obvious image features in the angiography image and the cross-sectional image, the use of the vascular branch can improve the accuracy of the registration of the angiography image and the first cross-sectional image set at the vascular branch, thereby improving the registration accuracy between the angiography image and the first cross-sectional image set.

[0011] In addition, in the method for marking a target blood vessel involved in the first aspect of the present disclosure, optionally, the method for obtaining an intraluminal image of the target blood vessel includes: obtaining an intraluminal image of the target blood vessel by intravascular ultrasound imaging or obtaining an intraluminal image of the target blood vessel by intravascular optical coherence tomography. In this case, the convenience of obtaining an intraluminal image of the target blood vessel can be improved; in addition, the applicability of the registration method involved in the present disclosure can be improved.

[0012] In addition, in the method for marking a target blood vessel involved in the first aspect of the present disclosure, optionally, the target blood vessel includes blood vessel walls on two opposite sides in the angiography image, the starting position is represented by the intersection of the blood vessel cross section at the starting position and the blood vessel walls on the two opposite sides, and the ending position is represented by the intersection of the blood vessel cross section at the ending position and the blood vessel walls on the two opposite sides. In this case, since the resolution of the angiography image is higher than the resolution of the cross-sectional image, the shape of the target blood vessel can be displayed more clearly through the angiography image, and the blood vessel diameter and axial direction of the target blood vessel can be displayed clearly. By determining the starting position and the ending position based on the intersection of the blood vessel cross section and the blood vessel walls on the two opposite sides, the accuracy of positioning the starting position and the ending position in the angiography image can be improved, and the starting position and the ending position can be made consistent with the target blood vessel diameter displayed in the angiography image to the greatest extent, thereby being conducive to improving the registration accuracy between the angiography image and the first cross-sectional image set.

[0013] In addition, in the method for marking a target blood vessel involved in the first aspect of the present disclosure, optionally, a target cross-sectional image that meets the preset condition is selected by feature recognition, and the feature recognition step includes: obtaining the plaque load of each cross-sectional image in the first cross-sectional image set, the preset condition is that the plaque load is not less than a preset threshold, and in response to the plaque load being not less than the preset threshold, obtaining the corresponding cross-sectional image as the target cross-sectional image. In this case, since the target blood vessel has a continuous length, by obtaining the plaque load of each cross-sectional image in the first cross-sectional image set, the state of the lumen of the target blood vessel can be comprehensively analyzed and confirmed, thereby improving the accuracy of obtaining the target blood vessel. In addition, by taking the plaque load of the blood vessel reflected by the cross-sectional image as not less than the preset threshold as the preset condition, the range of screening the target cross-sectional image can be expanded, and the possibility of missing the target cross-sectional image can be reduced, thereby facilitating the acquisition of the complete target blood vessel.

[0014] In addition, in the method for marking a target blood vessel according to the first aspect of the present disclosure, optionally, in response to the multiple target cross-sectional images being adjacent cross-sectional images, the target cross-sectional image set is obtained based on the multiple target cross-sectional images. In this case, since the target blood vessel has a continuous length, by confirming that the multiple target cross-sectional images are adjacent to each other as the target cross-sectional image set, the possibility of misjudging the target blood vessel can be effectively reduced, thereby improving the accuracy of identifying the target blood vessel.

[0015] In addition, in the method for marking a target vessel involved in the first aspect of the present disclosure, optionally, after obtaining the contour of the target vessel, a secondary adjustment is performed by manual verification, and the manual verification method includes adjusting the contour of the target vessel based on a target longitudinal section image, and the target longitudinal section image is obtained based on the target cross-sectional image set. In this case, since the target longitudinal section image can more intuitively display the axial characteristics of the target vessel, it is convenient for the operator to compare and analyze the target longitudinal section image with the contour of the target vessel, thereby improving the convenience and accuracy of manual verification. As a result, the accuracy of obtaining the contour and position of the target vessel can be further improved.

[0016] In addition, in the method for marking a target blood vessel according to the first aspect of the present disclosure, optionally, a target length is obtained based on the contour of the target blood vessel, and the target length is compared with the length of the target longitudinal section image to evaluate the accuracy of the contour of the target blood vessel. In this case, the target length corresponding to the length of the target blood vessel can be obtained relatively quickly through the angiography image. In addition, the target length can be compared with the length of the target longitudinal section image relatively conveniently, and the accuracy of the comparison result can be improved.

[0017] According to a second aspect of the present disclosure, there is provided a device for marking a target blood vessel based on an intracavitary image and an angiography image, comprising an image acquisition module, a registration module, an identification module, a marking module and a human-computer interaction module; the image acquisition module comprises a first acquisition unit for acquiring an angiography image of the target blood vessel and a second acquisition unit for acquiring multiple cross-sectional images of the target blood vessel, the registration module is configured to register the multiple cross-sectional images based on the angiography image to acquire a registration mapping set, the identification module is configured to select target cross-sectional images that meet preset conditions from the multiple cross-sectional images and form a target cross-sectional image set, the marking module is configured to acquire a target position set based on the target cross-sectional image set and the registration mapping set, and to mark the angiography image based on the target position set to acquire the contour and position of the target blood vessel, the human-computer interaction module comprises a display unit for displaying the angiography image, the cross-sectional image, and the longitudinal cross-sectional image of the target blood vessel, and an input unit for adjusting the contour and position of the target blood vessel.

[0018] In the device for marking the target blood vessel involved in the second aspect of the present disclosure, by registering the angiographic image and the first cross-sectional image set, each position in the angiographic image can form a one-to-one mapping relationship with each frame of the cross-sectional image in the first cross-sectional image set, so that after selecting the target cross-sectional image set that meets the preset conditions, the corresponding position information can be locked in the angiographic image more accurately, and the contour and position of the target blood vessel can be marked more accurately in the angiographic image using the obtained position information. As a result, the target blood vessel can be displayed intuitively and clearly in the angiographic image, so that it is convenient for operators to observe and analyze. In addition, the human-computer interaction module can facilitate operators to observe and analyze the angiographic image, the cross-sectional image and the longitudinal section image of the target blood vessel more intuitively. In addition, the input unit can provide a human-computer interaction function, so that the convenience of adjusting the contour and position of the target blood vessel by manual verification can be improved.

[0019] According to the present disclosure, a method and device for marking a target blood vessel based on an intracavity image and an angiography image can be provided, which can simplify the calculation process, improve the registration efficiency and enhance the registration accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present disclosure will now be explained in further detail, by way of example only, with reference to the accompanying drawings.

[0021] Figure 1 is a flow chart showing a method for marking a target blood vessel according to an example of the present disclosure.

[0022] Figure 2 is a schematic diagram showing the acquisition of a cross-sectional image of a target blood vessel by using an ultrasound probe according to an example of the present disclosure.

[0023] Figure 3 is a cross-sectional image showing a target blood vessel involved in the examples of the present disclosure.

[0024] Figure 4 is a diagram showing an angiography image including a target blood vessel according to an example of the present disclosure.

[0025] Figure 5A is a schematic diagram showing a target blood vessel involved in the examples of the present disclosure.

[0026] Figure 5B is a schematic diagram showing target blood vessels and blood vessel branches involved in the examples of the present disclosure.

[0027] Fig. 6A is a schematic diagram showing a cross section of a target blood vessel involved in an example of the present disclosure.

[0028] Figure 6Bis a schematic diagram showing the lumen of a target blood vessel involved in examples of the present disclosure.

[0029] Figure 7 is a schematic diagram showing the starting position and the ending position of the target blood vessel involved in the example of the present disclosure.

[0030] Figure 8 is a flow chart showing the process of selecting a target cross-sectional image through feature recognition according to an example of the present disclosure.

[0031] Fig. 9 is a flow chart showing a registration method involved in the examples of the present disclosure.

[0032] Fig.10 is a schematic diagram showing the outline and position of marking a target blood vessel in a target blood vessel according to an example of the present disclosure.

[0033] Fig.11 is a flow chart showing the manual verification involved in the examples of the present disclosure.

[0034] Fig.12 is a schematic diagram showing a longitudinal cross-sectional image of a target involved in the example of the present disclosure.

[0035] Fig.13 is a structural block diagram showing an apparatus for marking a target blood vessel according to an example of the present disclosure.

[0036] Description of reference numerals:

[0037] Target vessel…1, target vessel…10, starting position…A, ending position…B, ultrasound probe…2, vessel branch…12, lumen…14, external elastic membrane…16, target direction…D, vessel centerline…L, marking device…100, image acquisition module…110, registration module…120, recognition module…130, marking module…140, entry module…150. DETAILED DESCRIPTION

[0038] Hereinafter, with reference to the accompanying drawings, the preferred embodiments of the present disclosure are described in detail. In the following description, the same symbols are given to the same parts, and repeated descriptions are omitted. In addition, the accompanying drawings are only schematic diagrams, and the ratio of the sizes of the parts to each other or the shapes of the parts, etc. may be different from the actual ones. It should be noted that the terms "including" and "having" in the present disclosure and any variations thereof, such as the processes, methods, systems, products or devices of a series of steps or units included or possessed are not necessarily limited to those steps or units clearly listed, but may include or have other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. All methods described in the present disclosure may be performed in any suitable order, unless otherwise indicated herein or clearly contradictory to the context.

[0039] The first aspect of the present disclosure relates to a method for marking a target blood vessel based on an intracavitary image and an angiographic image, which is a method for marking the contour and position of a target blood vessel in an angiographic image. The second aspect of the present disclosure relates to a device for marking a target blood vessel based on an intracavitary image and an angiographic image, which can use the method for marking a target blood vessel involved in the present disclosure to mark the contour and position of the target blood vessel in an angiographic image.

[0040] In some examples, the method for marking a target blood vessel based on an intracavity image and an angiography image involved in the present disclosure may be referred to as a method for marking a target blood vessel or a marking method. The device for marking a target blood vessel based on an intracavity image and an angiography image involved in the present disclosure may be referred to as a device for marking a target blood vessel or a marking device.

[0041] Hereinafter, the marking method involved in the present disclosure and the marking device using the marking method will be described in detail with reference to the accompanying drawings.

[0042] Figure 1 is a flow chart showing a method of marking a target blood vessel 10 according to an example of the present disclosure. Figure 2 1 is a schematic diagram showing a cross-sectional image of a target blood vessel 1 obtained by using an ultrasonic probe 2 according to an example of the present disclosure. Figure 3 is a cross-sectional image showing a target blood vessel 1 according to the example of the present disclosure. Figure 4 is a diagram showing a blood vessel angiography image including a target blood vessel 1 according to an example of the present disclosure. Figure 5A is a schematic diagram showing a target blood vessel 1 according to an example of the present disclosure. Figure 5B is a schematic diagram showing a target blood vessel 1 and a blood vessel branch 12 involved in an example of the present disclosure. Fig. 6A is a schematic diagram showing a cross section of a target blood vessel 1 according to an example of the present disclosure. Figure 6B is a schematic diagram showing the lumen 14 of the target blood vessel 1 according to the example of the present disclosure. Figure 7 1 is a schematic diagram showing a starting position A and an ending position B of a target blood vessel 1 according to an example of the present disclosure and marking a target blood vessel 10 .

[0043] For some examples, see Figure 1 The method for marking the target blood vessel 10 may include: acquiring an angiography image including the target blood vessel 1, a first cross-sectional image set, and a target cross-sectional image set (step S100); acquiring a registration mapping set based on the angiography image and the first cross-sectional image set (step S200); acquiring a target position set based on the target cross-sectional image set and the registration mapping set (step S300); and acquiring a contour and position of the target blood vessel 10 based on the target position set and the angiography image (step S400).

[0044] In some examples, in step S100 , an angiography image including the target blood vessel 1 , a first cross-sectional image set, and a target cross-sectional image set may be acquired.

[0045] In some examples, the target blood vessel 1 may represent a blood vessel to be treated, for example, the target blood vessel 1 may be a coronary artery that needs to be observed and analyzed.

[0046] In some examples, the target vessel 10 may be a vessel located at a distance in the target vessel 1, that is, the target vessel 10 may have a continuous length. Figure 2 The target vessel 10 may represent an abnormal vessel at a distance in the target vessel 1 (for example, the target vessel 10 may represent a vessel at a distance in the coronary artery where stenosis occurs).

[0047] In some examples, the angiographic image may be an image of a blood vessel obtained by introducing a contrast agent (such as a contrast agent or a tracer) into the blood vessel and then obtaining the image of the blood vessel by X-ray imaging technology. The angiographic image involved in the present disclosure may be an angiographic image ( Figure 4 The target blood vessel 1 is schematically shown in the position of the angiography image. The marking method of the present disclosure can be used to mark a target blood vessel 10 that meets preset conditions in an angiography image (hereinafter referred to as an angiography image) containing the target blood vessel 1.

[0048] In some examples, the digital subtraction angiography (DSA) technology can be used to better eliminate the image of redundant tissue in the angiography image, and mainly retain the image of the target blood vessel 1. In this case, the angiography image can display the structure of the target blood vessel 1 in more detail.

[0049] In some examples, the first cross-sectional image set may be a set of multiple cross-sectional images representing the target blood vessel 1 , wherein the cross-sectional images of the target blood vessel 1 may reflect intraluminal information of the target blood vessel 1 .

[0050] In some examples, in step S100 , an intraluminal image of the target blood vessel may be acquired, and multiple cross-sectional images of the target blood vessel may be acquired based on the intraluminal image to acquire a first set of cross-sectional images.

[0051] Specifically, the method for acquiring an intraluminal image of a target blood vessel may include: acquiring an intraluminal image of the target blood vessel 1 through an intravascular ultrasound (IVUS) imaging system. In some examples, the IVUS imaging system may include a guide catheter, a guide wire, an IVUS catheter, and an ultrasonic transducer (referred to as ultrasound probe 2) located in the IVUS catheter. The process of acquiring an intraluminal image through the IVUS imaging system may include: moving the guide catheter to a preset position of the target blood vessel 1 (e.g., the coronary ostium of a coronary artery), moving the guide wire along the guide catheter into the target blood vessel 1 (e.g., into a coronary artery), moving the IVUS catheter along the guide wire into the target blood vessel 1, and performing retraction and rotational movements along the target direction D by the ultrasound probe 12 to acquire an intraluminal image of the target blood vessel 1 (see Figure 2 ).

[0052] In other examples, the method for acquiring the intraluminal image of the target blood vessel may further include acquiring the intraluminal image of the target blood vessel 1 by intravascular optical coherence tomography. Specifically, an optical coherence tomography (OCT) system may be used to scan and image the lumen of the target blood vessel 1, thereby acquiring the intraluminal image of the target blood vessel 1. In this case, the convenience of acquiring the intraluminal image of the target blood vessel can be improved; in addition, the applicability of the registration method involved in the present disclosure can be improved.

[0053] For some examples, see Figure 2 According to the intracavitary image acquired by the IVUS imaging system, multiple cross-sectional images of the target blood vessel 1 can be acquired to obtain a first cross-sectional image set. According to the process of the ultrasound probe 2 performing a retraction movement in the target blood vessel 1, the acquisition order of each cross-sectional image in the first cross-sectional image set can be acquired.

[0054] In some examples, a cross-sectional image may represent a frame of cross-sectional images acquired by the ultrasound probe 2 at a retracted position in the target blood vessel 1, and multiple frames of cross-sectional images (i.e., cross-sectional image sequence frames) may be acquired according to the acquisition order of each cross-sectional image. In some examples, a cross-sectional image sequence frame may represent a cross-sectional image set.

[0055] In some examples, the target cross-sectional image set may represent a set of multiple target cross-sectional images, wherein the target cross-sectional image may be a cross-sectional image in the first cross-sectional image set that meets a preset condition.

[0056] In some examples, specifically, a target cross-sectional image satisfying a preset condition may be selected from the first cross-sectional image set to form a target cross-sectional image set.

[0057] Figure 8is a flow chart showing the process of selecting a target cross-sectional image through feature recognition according to an example of the present disclosure.

[0058] In some examples, a target cross-sectional image that meets a preset condition may be selected from the first cross-sectional image set by feature recognition. Figure 8 The feature recognition step may include: obtaining the plaque load of each cross-sectional image in the first cross-sectional image set (step S110); setting a preset threshold related to the plaque load (step S130); and obtaining a target cross-sectional image in response to the plaque load being not less than the preset threshold (step S150).

[0059] In some examples, in step S110, the plaque load of each cross-sectional image in the first cross-sectional image set may be obtained. The plaque load may be a ratio of the plaque area on a cross-sectional image to the area of ​​the lumen 14. In this case, since the target blood vessel 10 has a continuous length, by obtaining the plaque load of each cross-sectional image in the first cross-sectional image set, the state of the lumen of the target blood vessel 1 can be comprehensively analyzed and confirmed, thereby improving the accuracy of obtaining the target blood vessel 10.

[0060] In some examples, the plaque load can be used to characterize the degree of stenosis of a blood vessel. For example, if the plaque load of a certain distance in a blood vessel is greater than a preset threshold, it can indicate that the blood vessel in that distance has a relatively serious stenosis, and a vascular stent needs to be installed to dredge the blood vessel.

[0061] In some examples, the cross-sectional image may contain information on multiple blood vessel parameters, and the multiple blood vessel parameters may be acquired by analyzing and processing the cross-sectional image.

[0062] For some examples, see Fig. 6A , the IVUS imaging system can identify the lumen 14 and the external elastic membrane 16 of the blood vessel according to the cross-sectional image, and the plaque load corresponding to the current cross-sectional image can be calculated using the lumen 14 and the external elastic membrane 16. In this case, the plaque load corresponding to each cross-sectional image in the first cross-sectional image set can be calculated relatively conveniently through the lumen 14 and the external elastic membrane 16. Thus, the convenience of obtaining the plaque load of the cross-sectional image can be improved.

[0063] In some examples, in step S130, a preset threshold value related to plaque load may be set. In some examples, the preset threshold value may be a fixed value set based on years of clinical experience (e.g., the preset threshold value may be derived from past experience and set by medical staff). Thus, the practicality of the preset threshold value may be improved.

[0064] In some examples, the preset condition may be that the plaque load is not less than a preset threshold. For example, the target cross-sectional image may be a cross-sectional image selected from the first cross-sectional image set and having a plaque load greater than a preset threshold.

[0065] In some examples, in step S150, in response to the plaque load being not less than a preset threshold, a target cross-sectional image may be acquired. In this case, by using the plaque load of the blood vessel reflected by the cross-sectional image being not less than a preset threshold as a preset condition, the range of screening the target cross-sectional image can be expanded, and the possibility of missing the target cross-sectional image can be reduced, thereby facilitating the acquisition of a complete target blood vessel 10.

[0066] In some examples, in response to the plaque load being greater than a preset threshold, a corresponding cross-sectional image may be acquired as a target cross-sectional image.

[0067] In other examples, in response to the plaque load being equal to a preset threshold, a corresponding cross-sectional image may be acquired as a target cross-sectional image.

[0068] In some examples, a target cross-sectional image set may be formed based on the multiple target cross-sectional images acquired in step S150. Specifically, in response to the multiple target cross-sectional images being adjacent cross-sectional images, a target cross-sectional image set may be obtained based on the multiple target cross-sectional images. In this case, since the target blood vessel 10 has a continuous length, by confirming that the multiple target cross-sectional images that are adjacent to each other are the target cross-sectional image set, the possibility of misjudging the target blood vessel 10 can be effectively reduced, thereby improving the accuracy of identifying the target blood vessel 10.

[0069] Return to reference Figure 1 In some examples, in step S200, a registration mapping set may be acquired based on the angiography image and the first cross-sectional image set. Specifically, the first cross-sectional image set may be registered based on the angiography image of the target vessel 1 to acquire the registration mapping set.

[0070] Fig. 9 is a flow chart showing a registration method involved in the examples of the present disclosure.

[0071] In some examples, the registration (also referred to as the registration method) involved in the present disclosure may represent a one-to-one mapping relationship between multiple cross-sectional images of the target blood vessel 1 and various positions of the target blood vessel 1 in the angiography image. For example, the retraction position of the ultrasound probe 2 in the target blood vessel 1 corresponding to each frame of the cross-sectional image may be found in the angiography image through a blood vessel parameter (e.g., blood vessel diameter).

[0072] For some examples, see Fig. 9The steps of the registration method may include: marking the starting position A and the ending position B of the target blood vessel 1 in the angiography image and the intracavity image (step S210); acquiring a first blood vessel image feature set of the angiography image between the starting position A and the ending position B (step S230); acquiring a second blood vessel image feature set of the first cross-sectional image set (step S250); and acquiring a registration mapping set based on the intersection of the first blood vessel image feature set and the second blood vessel image feature set (step S270).

[0073] In some examples, in step S210, the starting position A and the ending position B of the target blood vessel 1 may be marked in the angiography image. Specifically, by taking the target cross-sectional image set acquired in step S150 as a reference, the starting position A and the ending position B of the target blood vessel 1 may be marked in the angiography image (see Figure 5A In this case, by marking the starting position A and the ending position B of the target blood vessel 1 based on the target cross-sectional image set as a reference, the length of the target blood vessel 10 can be better adapted.

[0074] In some examples, in the angiography image, the start position A and the end position B may be manually marked by the user. In other examples, in the angiography image, the start position A and the end position B may be automatically recommended based on artificial intelligence (AI). Thus, the start position A and the end position B may be obtained more conveniently.

[0075] In some examples, the target blood vessel 1 may include blood vessel walls on two opposite sides in the angiography image. Specifically, the angiography image may show the outline of the target blood vessel 1, and the outline of the target blood vessel 1 may include blood vessel walls on two opposite sides.

[0076] In some examples, in the angiography image, the starting position A of the target blood vessel 1 may be represented by the intersection of the blood vessel cross section at the starting position A and the blood vessel walls on opposite sides. For example, at the starting position A of the target blood vessel 1, the intersection of the blood vessel cross section and the blood vessel walls on opposite sides may include a first intersection point a1 and a second intersection point a2 ( Figure 7 It is schematically shown that the starting position A is determined by the line between the first intersection point a1 and the second intersection point a2). In this case, by determining the starting position A at the position where the blood vessel cross section intersects with the blood vessel walls on opposite sides, the accuracy of positioning the starting position A in the angiographic image can be improved, and the starting position A can be made consistent with the diameter of the target blood vessel 1 displayed in the angiographic image to the greatest extent, thereby facilitating the improvement of the registration accuracy between the target blood vessel 1 in the angiographic image and the first cross-sectional image set of the target blood vessel 1.

[0077] In some examples, in the angiography image, the end position B of the target blood vessel 1 may be represented by the intersection of the blood vessel cross section at the end position B and the blood vessel walls on opposite sides. For example, at the end position B of the target blood vessel 1, the intersection of the blood vessel cross section and the blood vessel walls on opposite sides may include a third intersection point b1 and a fourth intersection point b2 ( Figure 7 It is schematically shown that the end position B is determined by the line between the third intersection point b1 and the fourth intersection point b2).

[0078] In some examples, in step S230, a first blood vessel image feature set of the angiography image between the starting position A and the ending position B may be obtained. Specifically, in the angiography image, a first blood vessel image feature set for the angiography image between the starting position and the ending position may be obtained.

[0079] In some examples, the first set of vessel image features may include a first set of vessel diameters.

[0080] In some examples, the first vessel diameter set may include multiple positions from the starting position A to the ending position B of the target vessel 1 and multiple target vessel diameters corresponding to the multiple positions. Specifically, the first vessel diameter set may include multiple first vessel diameter elements, each of which may include a vessel position (i.e., a target vessel position) and a first vessel diameter corresponding to the target vessel position.

[0081] In some examples, in the angiography image, there may be multiple target vessel positions between the starting position A and the ending position B of the target vessel 1, and each target vessel position may correspond to a first vessel diameter, so that multiple target vessel positions between the starting position A and the ending position B and multiple first vessel diameters corresponding to the multiple target vessel positions may be acquired to form a first vessel diameter set.

[0082] For some examples, see Figure 5A , the multiple target blood vessel positions can be determined in the following manner: in the angiography image, by calculating all pixels distributed along the center line L of the target blood vessel 1, the specific number of the multiple target blood vessel positions can be determined in a manner that one pixel corresponds to one target blood vessel position. In this case, the convenience of obtaining the multiple target blood vessel positions can be improved, and the exact number of the multiple target blood vessel positions can be determined easily, so that the multiple first blood vessel diameters can be obtained more conveniently.

[0083] In some examples, the first blood vessel diameter can be determined by the target blood vessel position corresponding to the first blood vessel diameter. Specifically, in the angiography image, the target blood vessel position can be represented by the intersection of the blood vessel cross section located at the target blood vessel position and the blood vessel walls on opposite sides (i.e., the first intersection and the second intersection), so that the first blood vessel diameter can be determined by the straight-line distance between the first intersection and the second intersection. In this way, the first blood vessel diameter corresponding to the target blood vessel position can be obtained relatively accurately.

[0084] In some examples, the first vessel image feature also includes a first vessel branch position set. In some examples, in the angiography image, features related to each vessel branch 12 from the starting position A to the ending position B of the target vessel 1 can be obtained to form a first vessel branch position set. Specifically, the first vessel branch position set can be a set of distances between each vessel branch 12 and the starting position A from the starting position A to the ending position B of the target vessel 1. In other examples, the first vessel branch position set can be a set of distances between each adjacent vessel branch 12 from the starting position A to the ending position B of the target vessel 1.

[0085] In the present disclosure, the target blood vessel 1 may be a trunk blood vessel, and the blood vessel branch 12 may represent a blood vessel connected to the trunk blood vessel so that blood flows into (or out of) the trunk blood vessel. In some examples, the trunk blood vessel may be connected to multiple blood vessel branches 12, that is, in the angiography image, multiple blood vessel branches 12 connected to the target blood vessel 1 may be distributed between the starting position A and the ending position B of the target blood vessel 1 ( Figure 5A One of the multiple blood vessel branches 12 is schematically shown.

[0086] In some examples, the position of the blood vessel branch 12 may represent the position where the blood vessel branch 12 is connected to the target blood vessel 1 ( Figure 5B The schematic diagram shows a position C where the blood vessel branch 12 is connected to the target blood vessel 1). In some examples, in the angiography image, the position of the blood vessel branch 12 can be determined by the image distance of the blood vessel branch 12 from the starting position A, wherein the image distance can be determined by the pixel size (i.e., the number of pixels) between the blood vessel branch 12 and the starting position A. In this case, the position of the blood vessel branch 12 can be determined in the angiography image more conveniently and accurately.

[0087] In other examples, in angiography images, for example, if the target blood vessel 1 has multiple blood vessel branches 12 from the starting position A to the ending position B, the positions of the blood vessel branches 12 can be represented by the image distances between any two adjacent blood vessel branches 12 .

[0088] In some examples, the first blood vessel image feature further includes a first blood vessel branch diameter set. In some examples, the first blood vessel branch diameter of the blood vessel branch 12 connected to the target blood vessel 1 in the angiography image can be directly calculated using a Picture Archiving and Communication System (PACS).

[0089] In some examples, in the angiography image, the physical size of the pixel can be calculated based on the characteristic image of the reference object (e.g., the guiding catheter of the IVUS imaging system) and the inherent size of the guiding catheter, and the first vessel branch diameter of the vessel branch 12 can be calculated by calculating the pixel size of the first vessel branch diameter of the vessel branch 12. In this case, the accuracy of obtaining the first vessel branch diameter of the vessel branch 12 can be improved.

[0090] Return to reference Fig. 9 In step S250, a second blood vessel image feature set of the first cross-sectional image set may be obtained. Specifically, in the intracavitary image, a second blood vessel image feature set for the first cross-sectional image set from the starting position A' to the ending position B' may be obtained. The starting position A' may correspond to the starting position of the ultrasound probe 2 withdrawn in the target blood vessel 1, and the ending position B' may correspond to the ending position of the ultrasound probe 2 withdrawn in the target blood vessel 1.

[0091] In some examples, the second set of vessel image features may include a second set of vessel diameters.

[0092] In some examples, the second blood vessel diameter set may include the acquisition order of each cross-sectional image and the second blood vessel diameter of each cross-sectional image. Specifically, the acquisition order of each cross-sectional image in the first cross-sectional image set and the second blood vessel diameter of the cross-sectional image matching the acquisition order may be acquired to form the second blood vessel diameter set. The second blood vessel diameter may represent the diameter of the lumen 14 reflected from the blood vessel cross-sectional image.

[0093] In some examples, the second blood vessel diameter set may include multiple second blood vessel diameter elements, each of which may include a frame number of a cross-sectional image and the second blood vessel diameter of the cross-sectional image, wherein the frame number of the cross-sectional image may represent the sequence number of the cross-sectional image in the cross-sectional image sequence frame.

[0094] In some examples, during the process of the ultrasound probe 2 withdrawing from the starting point (i.e., the starting position A') in the lumen of the target blood vessel 1 along the target direction D at a preset speed to the end point (i.e., the end position B'), the acquisition order of each cross-sectional image can be obtained by determining the acquisition frequency (i.e., the frame rate) of the ultrasound probe 2.

[0095] In some examples, the first cross-sectional image sequence frames can be acquired according to the acquisition order of each cross-sectional image in the first cross-sectional image set, that is, the first cross-sectional image sequence frames can represent multiple cross-sectional images that match the acquisition order. In other words, in the first cross-sectional image set, the first cross-sectional image sequence frames can include a cross-sectional image of an initial frame number, a cross-sectional image of an end frame number, and multiple cross-sectional images located between the initial frame number and the end frame number and sorted by frame number. In this case, the acquisition order of each cross-sectional image in the cross-sectional image set can be acquired more accurately through the frame number of the first cross-sectional image sequence frames. In some examples, the cross-sectional image of the initial frame number can represent the cross-sectional image of the blood vessel acquired by the ultrasound probe 2 at the starting position A'; the cross-sectional image of the end frame number can represent the cross-sectional image of the blood vessel acquired by the ultrasound probe 2 at the end position B'.

[0096] In some examples, the IVUS imaging system can be used to directly calculate the second blood vessel diameter of the cross-sectional image of each frame number in the first cross-sectional image sequence frame, thereby obtaining the second blood vessel diameter of the cross-sectional image matching the acquisition sequence.

[0097] In some examples, in the first cross-sectional image sequence frame, for the cross-sectional image of any frame number, the second blood vessel diameter of the cross-sectional image of the frame number can be obtained by calculating the average value of the diameter along each direction through the center of the lumen 14. Thus, the second blood vessel diameter of the cross-sectional image matching the acquisition sequence can be obtained. Figure 6B In the example shown, the second blood vessel diameter of the cross-sectional image of the frame number can be obtained by calculating the average of the first diameter d1 , the second diameter d2 , and the third diameter d3 passing through the center O of the lumen 14 .

[0098] In some examples, the second vessel image feature set also includes a second vessel branch position set. The second vessel branch position set may be a set of distances between each vessel branch 12 and the starting position A' from the starting position A' (i.e., the position corresponding to the cross-sectional image of the initial frame number) to the ending position B' (i.e., the position corresponding to the cross-sectional image of the ending frame number) of the target vessel 1 and the starting position A'.

[0099] In other examples, the second vascular branch position set may also be a set of distances between adjacent vascular branches 12 from the starting position A' to the ending position B' of the target vessel 1. In other words, in the intraluminal image of the target vessel 1, the positions of the vascular branches 12 may be represented by the distances between two adjacent vascular branches 12.

[0100] In some examples, the second vessel image feature set also includes a second vessel branch diameter set. The second vessel branch diameter set may be a set of vessel diameters of each vessel branch 12 between the starting position A' and the ending position B' of the target vessel 1. The second vessel branch diameter may represent the diameter of the vessel branch 12 reflected in the intraluminal image of the target vessel 1.

[0101] In some examples, the diameter of each vascular branch 12 in the intraluminal image of the target blood vessel 1 can be directly calculated using the IVUS imaging system, and the diameters of each vascular branch 12 can collectively constitute a second vascular branch diameter set.

[0102] It should be noted that the calculation of the first and second vessel diameters of the target vessel 1, the first and second vessel branch positions of the vessel branch 12, and the first and second vessel branch diameters of the vessel branch 12 belong to the prior art. For example, in the prior art, the vessel image (including the angiography image and the vessel cross-sectional image) can be processed by a software algorithm to obtain the vessel features in the vessel image, and the first and second vessel diameters, the first and second vessel branch positions, and the first and second vessel branch diameters can be obtained by calculation based on the vessel features. For another example, the vessel features in the vessel image can be obtained by deep learning, such as based on a trained AI model for vessel diameter and vessel branch position, and the vessel image can be processed using the trained AI model to obtain the vessel diameter of the vessel in the vessel image (such as the vessel diameter of the target vessel 1 and the vessel diameter of the vessel branch 12) and the position information of the vessel branch.

[0103] Return to reference Fig. 9 In step S270, a registration mapping set may be acquired based on the intersection of the first vascular image feature set and the second vascular image feature set.

[0104] In some examples, a registration mapping set can be obtained based on the intersection of the first blood vessel diameter set and the second blood vessel diameter set. Specifically, for obtaining the registration mapping set, only the blood vessel diameter of the target blood vessel 1 can be used for registration (that is, the first blood vessel diameter set and the second blood vessel diameter set are intersected) to obtain the registration mapping set. In this case, since the blood vessel diameter is a blood vessel parameter that is relatively easy to obtain from the angiography image and the cross-sectional image, and is also a relatively obvious feature shared by the angiography image and the cross-sectional image, the blood vessel diameter can be used to quickly link each position in the angiography image with each frame of the cross-sectional image; in addition, the mapping relationship between each position in the angiography image and each frame of the cross-sectional image can be obtained by the intersection method, which can simplify the calculation process, reduce the difficulty of registration, and thus improve the efficiency of registration.

[0105] In some examples, a registration mapping set may be obtained based on the intersection of the first vessel branch diameter set and the second vessel branch diameter set. Specifically, for obtaining the registration mapping set, only the vessel branch diameter of the vessel branch 12 may be used for registration (i.e., the first vessel branch diameter set and the second vessel branch diameter set are intersected) to obtain the registration mapping set. In this case, since the vessel branch has relatively obvious image features in both the angiographic image and the cross-sectional image, the vessel branch diameter of the vessel branch 12 is used as a link to associate the angiographic image and the cross-sectional image, which can improve the accuracy of the registration of the angiographic image and the first cross-sectional image set at the vessel branch.

[0106] In some examples, a registration mapping set may be obtained based on the intersection of the first vascular branch position set and the second vascular branch position set. Specifically, for obtaining the registration mapping set, only the vascular branch position of the vascular branch 12 may be used for registration (i.e., the first vascular branch position set and the second vascular branch position set are intersected) to obtain the registration mapping set. In this case, since the vascular branch 12 has relatively obvious image features in both the angiographic image and the cross-sectional image, registration using the position information of the vascular branch 12 can improve the accuracy of registration of the angiographic image and the first cross-sectional image set at the vascular branch, thereby improving the registration accuracy between the angiographic image and the first cross-sectional image set.

[0107] However, the present disclosure is not limited thereto. In order to more accurately obtain the registration mapping set, the angiography image and the first cross-sectional image set may be registered multiple times.

[0108] In some examples, for obtaining the registration mapping set, the vessel diameter of the target vessel 1 may be used for the first registration, and then the branch position of the vessel branch 12 may be used for the second registration, and the first registration may be calibrated by the second registration. Thus, the registration mapping set may be obtained more accurately.

[0109] In addition, in some examples, for obtaining the registration mapping set, the vessel diameter of the target vessel 1 may be used for the first registration, and then the vessel branch diameter of the vessel branch 12 may be used for the second registration, and the first registration may be calibrated by the second registration. Thus, the accuracy of obtaining the registration mapping set may be improved.

[0110] It should be noted that the combination of using the vessel diameter of the target vessel 1, the vessel branch diameter of the vessel branch 12, and the vessel branch position of the vessel branch 12 for registration to obtain the registration mapping set can be combined according to actual usage.

[0111] The following describes the process of intersecting the first blood vessel diameter set and the second blood vessel diameter set by way of example.

[0112] As mentioned above, the first vascular diameter set may include multiple first vascular diameter elements, and the second vascular diameter set may include multiple second vascular diameter elements. When the first vascular diameter set and the second vascular diameter set are intersected, the first vascular diameter elements and the second vascular diameter elements can be associated in sequence in response to the first vascular diameter and the second vascular diameter being equal or differing within an error range.

[0113] In some examples, the first set of blood vessel diameters may satisfy the following formula:

[0114] U1={(a1,d1),(a2,d2),...(a i ,d i )} Formula 1

[0115] Among them, U1 can represent the first blood vessel diameter set, a i It can represent the i-th target vessel position between the starting position A and the ending position B of target vessel 1, d i can represent the first blood vessel diameter corresponding to the i-th target blood vessel position, (a i ,d i ) can represent the first blood vessel diameter element.

[0116] In some examples, the second set of blood vessel diameters may satisfy the following formula:

[0117] U2={(t1,r1),(t2,r2),...,(t n ,r n)} Formula 2

[0118] Among them, U2 can represent the second blood vessel diameter set, t n It can represent the frame number of the nth cross-sectional image in the first cross-sectional image sequence frame, r n It can represent the second blood vessel diameter of the cross-sectional image of the nth frame number, (t n ,r n ) can represent the second blood vessel diameter element.

[0119] When the first blood vessel diameter set and the second blood vessel diameter set are intersected, in response to the first blood vessel diameter being equal to the second blood vessel diameter, the first blood vessel diameter and the second blood vessel diameter are traversed, and the first blood vessel diameter elements and the second blood vessel diameter elements are sequentially associated to obtain a registration mapping set, that is, the registration mapping set can satisfy the following formula:

[0120] U3=U1∩U2={(a1,t1,d1),(a2,t2,d2),...,(a i ,t n ,d i )} Formula 3

[0121] or

[0122] U3=U1∩U2={(a1,t1,r1),(a2,t2,r2),...,(a i ,t n ,r n )} Formula 4

[0123] Among them, d i =r n or i =kr n , the first blood vessel diameter and the corresponding second blood vessel diameter can be equal, K indicates that the first blood vessel diameter and the corresponding second blood vessel diameter can also be proportional, d i It can represent the first blood vessel diameter corresponding to the i-th target blood vessel position, r n may represent the second blood vessel diameter of the cross-sectional image of the nth frame number, U3 may represent the registration mapping set, U1 may represent the first blood vessel diameter set, U2 may represent the second blood vessel diameter set, and a i It can represent the i-th target vessel position from the starting position A to the ending position B of target vessel 1, t n It can represent the frame number of the nth cross-sectional image in the first cross-sectional image sequence frames.

[0124] It should be noted that when determining whether the first blood vessel diameter is equal to the second blood vessel diameter, due to the calculation accuracy of the device and the curvature of the target blood vessel 1, even if they are located at the same position of the target blood vessel 1, the first blood vessel diameter obtained from the angiography image and the second blood vessel diameter obtained from the cross-sectional image may not be completely consistent. The intersection involved in the present disclosure may allow a preset error range. For example, as long as the difference between the first blood vessel diameter and the second blood vessel diameter is within the error range, it can be considered that they are located at the same position of the target blood vessel 1.

[0125] In the registration step involved in the present disclosure, since the first blood vessel diameter is a blood vessel parameter that is relatively easy to obtain from the angiography image and the second blood vessel diameter is a blood vessel parameter that is relatively easy to obtain from the cross-sectional image, and is also a relatively obvious feature shared by the angiography image and the cross-sectional image, the first blood vessel diameter and the second blood vessel diameter can be used to quickly and accurately link each position of the target blood vessel 1 in the angiography image with each cross-sectional image frame, that is, the angiography image and the cross-sectional image can be associated with the blood vessel diameter as a link; at the same time, through the intersection method, each position of the target blood vessel 1 in the angiography image can be linked with each cross-sectional image frame (that is, the mapping relationship between each position of the target blood vessel 1 in the angiography image and each cross-sectional image frame is obtained), which can simplify the calculation process, reduce the difficulty of registration, and thus improve the efficiency of registration. As a result, the registration mapping set can be obtained relatively accurately and quickly.

[0126] In some examples, the process of intersecting the first vessel branch diameter set and the second vessel branch diameter set may be similar to the process of intersecting the first vessel diameter set and the second vessel diameter set. Specifically, the first vessel branch diameter set may include multiple first vessel branch diameter elements, and the second vessel branch diameter set may include multiple second vessel branch diameter elements. When the first vessel branch diameter set and the second vessel branch diameter set are intersected, in response to the first vessel branch diameter and the second vessel branch diameter being equal or differing within an error range, the first vessel diameter elements and the second vessel diameter elements may be sequentially associated to obtain a registration mapping set.

[0127] In some examples, the process of intersecting the first blood vessel branch position set and the second blood vessel branch position set may be similar to the process of intersecting the first blood vessel branch diameter set and the second blood vessel branch diameter set, which will not be described in detail herein.

[0128] In the present disclosure, since the blood vessel branch 12 has relatively obvious image features in both the angiography image and the cross-sectional image, the use of the blood vessel branch 12 can improve the accuracy of the registration of the angiography image and the first cross-sectional image set at the blood vessel branch 12, thereby improving the registration accuracy between the angiography image and the first cross-sectional image set.

[0129] Return to reference Figure 1 In some examples, in step S300, a target position set can be obtained based on the target cross-sectional image set and the registration mapping set. Specifically, the registration mapping set obtained in step S270 can be used as a registration function that reflects the mapping relationship between multiple cross-sectional images and various positions of the angiography image. After selecting a target cross-sectional image set that meets the preset conditions, the position information corresponding to the target cross-sectional image set can be locked in the angiography image through the registration function. In this case, the target position set can be obtained more conveniently and accurately.

[0130] Fig.10 is a schematic diagram showing the outline and position of marking a target blood vessel on a target blood vessel 1 according to an example of the present disclosure.

[0131] For some examples, see Figure 1 In step S400, the contour and position of the target blood vessel 10 may be acquired based on the target position set and the angiography image. Fig.10 , the angiography image can be marked based on the target position set to obtain the contour and position of the target blood vessel 10 in the target blood vessel 1, that is, according to the target position set obtained in step S300, the specific position (also called the marked position) corresponding to the target cross-sectional image set in the target blood vessel 1 of the angiography image can be determined, and the angiography image is marked at the marked position, and the contour and position of the target blood vessel 10 can be displayed in the target blood vessel 1 of the angiography image. In this case, the target blood vessel 10 can be displayed intuitively and clearly in the angiography image, so that it is convenient for operators to observe and analyze.

[0132] In some examples, marking the angiography image may mean drawing the outline of the target blood vessel 10 on the angiography image according to the target position set using a graphic drawing method.

[0133] The present disclosure also provides a method for manually correcting the contour and position of the target blood vessel 10. Specifically, after obtaining the contour of the target blood vessel 10, a secondary adjustment can be performed by manual verification. Thus, the accuracy of obtaining the contour and position of the target blood vessel 10 can be improved.

[0134] Fig.11 is a flow chart showing the manual verification involved in the examples of the present disclosure. Fig.12 is a schematic diagram showing a longitudinal cross-sectional image of a target involved in the example of the present disclosure.

[0135] In some examples, the manual verification method may include adjusting the contour of the target blood vessel 10 based on the target longitudinal section image. The target longitudinal section image may be an image of a length in the longitudinal section image of the target blood vessel 1. In this case, since the target longitudinal section image can more intuitively display the axial characteristics of the target blood vessel 10, it is convenient for the operator to compare and analyze the target longitudinal section image with the contour of the target blood vessel 10, thereby improving the convenience of manual verification.

[0136] For some examples, see Fig.11 The manual verification method may include: acquiring a target longitudinal section image based on a target cross-sectional image set (step S510), obtaining a target length based on the contour of the target blood vessel 10 (step S530), and comparing the target length with the length of the target longitudinal section image to evaluate the accuracy of the contour of the abnormal blood vessel (step S550).

[0137] In some examples, in step S510 , a target longitudinal section image may be acquired based on the target cross-sectional image set, wherein the target longitudinal section image may represent a longitudinal section image of the target blood vessel 10 .

[0138] In some examples, the longitudinal section image of the target blood vessel 1 can be acquired by sequentially superimposing the cross-sectional images of each frame number in the first cross-sectional image set in the acquisition order (see Fig.12 ). In some examples, the longitudinal cross-sectional image of the target blood vessel 1 may reflect the actual length of the target blood vessel 1 , and the length of the target blood vessel 1 may be equal to the distance that the ultrasound probe 2 is withdrawn in the target blood vessel 1 .

[0139] In some examples, the target longitudinal section image can be acquired by marking the target vessel 10 on the longitudinal section image of the target vessel 1 according to the target cross-sectional image set acquired in step S150 and combining the length of the longitudinal section image of the target vessel 1 .

[0140] In some examples, in step S530, the target length may be obtained based on the contour of the target blood vessel 10. Specifically, the target length may be obtained by calculating the length of the target blood vessel 10 in the angiography image according to the contour of the target blood vessel 10 acquired in step S400. In this case, the target length corresponding to the length of the target blood vessel 10 in the angiography image can be obtained relatively quickly.

[0141] In some examples, in step S550, the target length and the length of the target longitudinal section image can be compared to evaluate the accuracy of the contour of the abnormal blood vessel. Specifically, the comparison result can be obtained by comparing the length of the target blood vessel 10 in the angiography image and the length of the target longitudinal section image, and the accuracy of the contour of the abnormal blood vessel can be evaluated according to the comparison result. In this case, the target length can be compared with the length of the target longitudinal section image more conveniently, and the accuracy of the comparison result can be improved.

[0142] In some examples, in response to the comparison result being within a preset range, it can be confirmed that the accuracy of the contour of the target blood vessel 10 acquired in step S400 is high, and no manual secondary adjustment is required.

[0143] In some examples, in response to the comparison result not being within the preset range, it can be confirmed that the accuracy of the contour of the target blood vessel 10 obtained in step S400 is low, and the contour of the target blood vessel 10 in the angiography image can be manually adjusted for a second time so that the comparison result is within the preset range. In this case, the accuracy of marking the contour of the target blood vessel 10 can be improved by manually checking the contour of the target blood vessel 10 for a second time, thereby effectively reducing the possibility of inaccurate marking of the contour of the target blood vessel 10 due to errors in the registration between the angiography image and the first cross-sectional image set.

[0144] Fig.13 is a block diagram showing the structure of an apparatus for marking a target blood vessel 10 according to an example of the present disclosure.

[0145] As described above, the present disclosure also relates to a device for marking a target blood vessel 10 based on an intracavitary image and an angiographic image (also referred to as a marking device 100). Fig.13 The marking device 100 may include an image acquisition module 110 , a registration module 120 , a recognition module 130 , and a marking module 140 .

[0146] In some examples, the image acquisition module 110 may include a first acquisition unit and a second acquisition unit. The first acquisition unit may be used to acquire angiographic images of the target blood vessel 1 (for example, the first acquisition unit may be a device that acquires angiographic images using DSA technology), and the second acquisition unit may be used to acquire multiple cross-sectional images of the target blood vessel 1 (for example, the second acquisition unit may be an IVUS imaging system). In some examples, the multiple cross-sectional images acquired by the second acquisition unit may constitute a first cross-sectional image set. For details, please refer to the relevant description in step S100.

[0147] In some examples, the registration module 120 may be used to implement the registration method involved in the present disclosure to obtain a set of registration maps.

[0148] In some examples, the registration module 120 may be configured to register multiple cross-sectional images based on the angiography image to obtain a registration mapping set. Specifically, the registration module 120 may use the registration method of the present disclosure to register the first cross-sectional image set based on the angiography image of the target blood vessel 1 to obtain a registration mapping set. For details, please refer to the relevant description in step S200.

[0149] In some examples, the recognition module 130 may be used to implement the feature recognition steps involved in the present disclosure to obtain a target cross-sectional image set.

[0150] In some examples, the recognition module 130 may be configured to select a target cross-sectional image that meets a preset condition from a plurality of cross-sectional images and form a target cross-sectional image set. That is, the recognition module 130 may select a target cross-sectional image that meets a preset condition from a first cross-sectional image set and form a target cross-sectional image set. For details, see the relevant description in step S100.

[0151] In some examples, the marking module 140 can be used to draw the target blood vessel 10 in the angiography image.

[0152] In some examples, the marking module 140 may be configured to obtain a target position set based on the target cross-sectional image set and the registration mapping set, and mark the angiography image based on the target position set to obtain the contour and position of the target blood vessel 10. For details, see the relevant descriptions in step S300 and step S400.

[0153] In the marking device 100 involved in the present disclosure, by registering the angiography image and the first cross-sectional image set, each position in the angiography image can form a one-to-one mapping relationship with each frame of the cross-sectional image in the first cross-sectional image set, so that after selecting the target cross-sectional image set that meets the preset conditions, the corresponding position information can be locked in the angiography image more accurately, and the contour and position of the target blood vessel 10 can be marked more accurately in the angiography image using the obtained position information. As a result, the target blood vessel 10 can be displayed intuitively and clearly in the angiography image, so that it is convenient for operators to observe and analyze. In addition, when registering the angiography image and the first cross-sectional image set, since the blood vessel diameter is a blood vessel parameter that is relatively easy to obtain from the angiography image and the cross-sectional image, and is also a relatively obvious feature shared by the angiography image and the cross-sectional image, the blood vessel diameter can be used to quickly and accurately link each position in the angiography image with each frame of the cross-sectional image, that is, the blood vessel diameter is used as a link to associate the angiography image and the cross-sectional image, which can simplify the calculation process, reduce the difficulty of registration, and thus improve the efficiency of registration. In this way, a set of registration maps can be obtained more quickly.

[0154] For some examples, see Fig.13 The marking device 100 involved in the present disclosure may also include a human-computer interaction module, wherein the human-computer interaction module may include a display unit and an input module 150.

[0155] In some examples, the display unit can be used to display angiographic images, cross-sectional images, and longitudinal sectional images. In this case, it is convenient for operators to observe and analyze the angiographic images, cross-sectional images, and longitudinal sectional images of the target blood vessel 1 more intuitively.

[0156] In some examples, the input module 150 can be used to adjust the contour and position of the target blood vessel 10. In this case, the input module 150 can provide a human-computer interaction function, thereby improving the convenience of adjusting the contour and position of the target blood vessel 10 by manual verification. As a result, the accuracy of obtaining the contour and position of the target blood vessel 10 can be improved.

[0157] Although the present disclosure is specifically described above in conjunction with the accompanying drawings and examples, it is to be understood that the above description does not limit the present disclosure in any form. Those skilled in the art may modify and change the present disclosure as needed without departing from the essential spirit and scope of the present disclosure, and these modifications and changes all fall within the scope of the present disclosure.

Claims

1. A method for marking a target blood vessel based on an intracavitary image and angiography image, which is a method for marking the contour and position of a target blood vessel in angiography image, characterized in that: include: Acquire angiographic images containing the target vessel, acquiring an intraluminal image of the target blood vessel, and acquiring a plurality of cross-sectional images of the target blood vessel based on the intraluminal image to acquire a first cross-sectional image set, Selecting target cross-sectional images that meet preset conditions from the first cross-sectional image set and forming a target cross-sectional image set, registering the first set of cross-sectional images based on the angiography image to obtain a set of registration maps, acquiring a set of target positions based on the set of target cross-sectional images and the set of registration maps, and The angiography image is marked based on the target position set to obtain the contour and position of the target blood vessel in the angiography image.

2. The method according to claim 1, characterized in that The step of registering the first cross-sectional image set based on the angiography image to obtain the registration mapping set comprises: marking the starting position and the ending position of the target blood vessel in the angiography image and the intraluminal image, Acquire a first blood vessel image feature set for the angiography image between the starting position and the ending position; Acquire a second blood vessel image feature set for the first cross-sectional image set from the starting position to the ending position; The registration mapping set is obtained based on an intersection of the first blood vessel image feature set and the second blood vessel image feature set.

3. The method according to claim 2, characterized in that The first blood vessel image feature set includes a first blood vessel diameter set in the angiography image, the second blood vessel image feature set includes a second blood vessel diameter set in the first cross-sectional image set, and the registration mapping set is acquired based on an intersection of the first blood vessel diameter set and the second blood vessel diameter set; and / or the first blood vessel image feature set includes a first blood vessel branch diameter set in the angiography image, the second blood vessel image feature set includes a second blood vessel branch diameter set in the first cross-sectional image set, and the registration mapping set is acquired based on an intersection of the first blood vessel branch diameter set and the second blood vessel branch diameter set; And / or the first vascular image feature set includes a first vascular branch position set in the angiography image, the second vascular image feature set includes a second vascular branch position set in the first cross-sectional image set, and the registration mapping set is obtained based on the intersection of the first vascular branch position set and the second vascular branch position set.

4. The method according to claim 1, characterized in that: The method for acquiring the intraluminal image of the target blood vessel includes: acquiring the intraluminal image of the target blood vessel by intravascular ultrasound imaging or acquiring the intraluminal image of the target blood vessel by intravascular optical coherence tomography.

5. The method according to claim 2, characterized in that: The target blood vessel includes blood vessel walls on two opposite sides in the angiography image, the starting position is represented by the intersection of the blood vessel cross section at the starting position and the blood vessel walls on the opposite sides, and the ending position is represented by the intersection of the blood vessel cross section at the ending position and the blood vessel walls on the opposite sides.

6. The method according to claim 1, characterized in that The target cross-sectional image that meets the preset condition is selected through feature recognition, wherein the feature recognition step comprises: obtaining the plaque load of each cross-sectional image in the first cross-sectional image set, The preset condition is that the plaque load is not less than a preset threshold, In response to the plaque load being not less than the preset threshold, a corresponding cross-sectional image is acquired as the target cross-sectional image.

7. The method according to claim 6, characterized in that In response to the multiple target cross-sectional images being adjacent cross-sectional images, the target cross-sectional image set is obtained based on the multiple target cross-sectional images.

8. The method according to claim 1, characterized in that After obtaining the contour of the target blood vessel, a secondary adjustment is performed by manual verification. The manual verification method includes adjusting the contour of the target blood vessel based on a target longitudinal section image, and the target longitudinal section image is obtained based on the target cross-sectional image set.

9. The method according to claim 8, characterized in that A target length is obtained based on the contour of the target blood vessel, and the target length is compared with the length of the target longitudinal cross-sectional image to evaluate the accuracy of the contour of the target blood vessel.

10. A device for marking a target blood vessel based on an intracavity image and an angiography image, characterized in that: It includes image acquisition module, registration module, recognition module, marking module and human-computer interaction module; The image acquisition module includes a first acquisition unit for acquiring an angiography image of a target blood vessel and a second acquisition unit for acquiring a plurality of cross-sectional images of the target blood vessel. The registration module is configured to register the plurality of cross-sectional images based on the angiography image to obtain a registration mapping set, The recognition module is configured to select a target cross-sectional image that meets a preset condition from the plurality of cross-sectional images and form a target cross-sectional image set. The marking module is configured to obtain a target position set based on the target cross-sectional image set and the registration mapping set, and mark the angiography image based on the target position set to obtain the contour and position of the target blood vessel, The human-computer interaction module includes a display unit for displaying the angiography image, the cross-sectional image, and the longitudinal section image of the target blood vessel, and an input unit for adjusting the contour and position of the target blood vessel.

Images