System for labeling vascular ultrasound images

Abstract

The present disclosure describes a blood vessel ultrasound image annotation system, comprising: an acquisition module, a path planning module and an adjacent point connection module; the acquisition module is used to determine a plurality of discrete points of the contour of the target region in the blood vessel ultrasound image; the path planning module is used to determine the target path of the plurality of discrete points; the adjacent point connection module is used to determine the target curve segment between each pair of adjacent two discrete points in the target path, and then obtain the target curve corresponding to the contour of the target region. In this case, the target curve that fits the contour of the target region can be obtained.

Description

[0001] This application is a divisional application of the patent application filed on April 13, 2022, with application number 2022103851217, entitled "A Method, Device and Medium for Labeling Vascular Ultrasound Images". Technical Field

[0002] This disclosure relates to the field of image technology, and more specifically to a labeling system for vascular ultrasound images. Background Technology

[0003] Intravascular ultrasound (IVUS) can effectively help doctors observe the characteristics of lesions inside blood vessels. For example, in clinical surgery, IVUS can help determine the location and severity of lesions before the procedure and help analyze the nature of the lesions. In addition, IVUS can also provide important information such as the diameter of the lumen in terms of stent size selection.

[0004] Currently, the information provided by IVUS imaging can be divided into two aspects: qualitative assessment and quantitative measurement. Qualitative assessment may include determining the nature of plaques, the presence of thrombosis, anatomical features, or previous stent placement. Quantitative measurement may include measurements of the lumen (i.e., the intima), vessel (i.e., the media), plaque burden, or lesion length. Therefore, determining the target region in IVUS imaging is crucial. Existing methods for determining the target region generally employ tracing, a common method which involves first selecting several discrete control points of the target region's outline and then interpolating and connecting them.

[0005] However, different interpolation methods have a significant impact on whether the final curve closely matches the contour of the target region, thus affecting the accuracy of quantitative measurements. For example, curves drawn by common interpolation methods (such as parabolic interpolation and Bezier interpolation) do not fit the target region well (i.e., they are not smooth enough). Therefore, how to obtain curves that better match the contour of the target region remains to be studied. Summary of the Invention

[0006] This disclosure is made in view of the above-mentioned situation, and its purpose is to provide a method, device and medium for annotating vascular ultrasound images that can obtain target curves that more closely fit the contour of the target region.

[0007] To this end, the first aspect of this disclosure provides a method for annotating vascular ultrasound images, comprising: determining a plurality of discrete points of the contour of a target region in the vascular ultrasound image; determining a target path of the plurality of discrete points, wherein the target path is closed; and determining a target curve segment between each pair of adjacent discrete points in the target path, thereby obtaining a target curve corresponding to the contour, wherein each pair of adjacent discrete points is considered an adjacent point pair, and determining the target curve segment of the adjacent point pair includes: obtaining two discrete points in the target path other than the discrete points of the adjacent point pair as two auxiliary points, wherein each auxiliary point... A point is adjacent to a discrete point in the adjacent point pair, thus defining two discrete point sets. Each discrete point set includes an auxiliary point and the adjacent point pair. The auxiliary points in the two discrete point sets are different auxiliary points among the two auxiliary points. Based on the target circle and concave direction corresponding to each discrete point set, a matching curve segment corresponding to each discrete point set is determined. The matching curve segment is between the discrete points of the adjacent point pair and belongs to the target circle. The concave direction is determined by the position and preset order of the discrete points in each discrete point set. The target curve segment is obtained by matching the two matching curve segments corresponding to the two discrete point sets. In this case, by matching the arcs of adjacent discrete points, a target curve segment passing through two adjacent discrete points can be obtained, thus obtaining a target curve that fits the contour of the target region. In addition, it can support obtaining the target curve corresponding to the contour of the target region using fewer discrete points with a smaller computational load.

[0008] Furthermore, in the annotation method according to the first aspect of this disclosure, optionally, determining the target path of the plurality of discrete points further includes: establishing a polar coordinate system; converting the coordinates of each discrete point among the plurality of discrete points into coordinates based on the polar coordinate system; and sorting the plurality of discrete points according to the angles of the coordinates in the polar coordinate system to determine the target path. In this case, by sorting the plurality of discrete points on the plane around the polar coordinate system, the target path obtained by the sorting can better conform to the contour features of the target region. Thus, the connection relationship between discrete points can be easily determined.

[0009] Furthermore, in the annotation method involved in the first aspect of this disclosure, optionally, establishing the polar coordinate system further includes: using the coordinates of each discrete point as coordinates in a rectangular coordinate system, obtaining the coordinates of the geometric center of the plurality of discrete points, using the coordinates of the geometric center as the pole, and a preset direction as the positive direction of the polar axis, to establish the polar coordinate system. Thus, a polar coordinate system can be established.

[0010] Furthermore, in the annotation method according to the first aspect of this disclosure, optionally, determining the matching curve segment corresponding to each discrete point set based on the target circle and the concave direction corresponding to each discrete point set further includes: in the target circle corresponding to each discrete point set, taking the arc from one discrete point to another of the adjacent point pair in the predetermined order, and in the concave direction, as the matching curve segment corresponding to each discrete point set. Thus, the matching curve segment can be determined based on the concave direction.

[0011] Furthermore, in the annotation method involved in the first aspect of this disclosure, optionally, the concavity direction is determined by the position of discrete points in each discrete point set and the preset order, further including: letting the first vector be the vector corresponding to the first discrete point to the second discrete point in each discrete point set in the preset order, and letting the second vector be the vector corresponding to the first discrete point to the third discrete point in each discrete point set in the preset order; and obtaining the cross product result of the first vector and the second vector as the third vector. If the direction of the third vector is upward, it indicates that the concavity direction is counterclockwise; if the direction of the third vector is downward, it indicates that the concavity direction is clockwise; or if the angle between the first vector and the second vector rotated clockwise is less than 180°, it indicates that the concavity direction is clockwise; if the angle between the first vector and the second vector rotated counterclockwise is less than 180°, it indicates that the concavity direction is counterclockwise. Thus, the concavity direction can be conveniently determined by the cross product. In addition, the concavity direction can also be conveniently determined by the angle between the vectors.

[0012] Furthermore, in the annotation method according to the first aspect of this disclosure, optionally, the target region includes a region of at least one target, namely the media and intima of the blood vessel. Thus, it is possible to subsequently determine the target curves corresponding to the contours of the regions of the media and intima of the blood vessel.

[0013] Furthermore, in the annotation method according to the first aspect of this disclosure, optionally, the method of adjusting two adjustment curve segments corresponding to the two discrete point sets to obtain the target curve segment further includes: taking any discrete point in the adjacent point pair as a starting point; starting from the starting point, determining a predetermined number of first adjustment points evenly distributed along the first curve segment, and a predetermined number of second adjustment points evenly distributed along the second curve segment, wherein the two adjustment curve segments include the first curve segment and the second curve segment; and determining interpolation points between the discrete points of the adjacent point pair based on the first adjustment points and the second adjustment points, and obtaining the target curve segment based on the interpolation points. In this case, adjustment points are obtained based on the two adjustment curve segments determined based on the four discrete points corresponding to the adjacent point pair, and interpolation points are obtained based on the adjustment points, thereby obtaining the target curve segment. This allows the target curve segment to better fit the contour of the target region.

[0014] Additionally, in the annotation method involved in the first aspect of this disclosure, optionally, the first curve segment is a relocation curve segment corresponding to the set of discrete points adjacent to the starting point and the auxiliary points in the two discrete point sets, and the second curve segment is a relocation curve segment corresponding to the set of discrete points not adjacent to the starting point and the auxiliary points in the two discrete point sets, wherein the coordinates M of the i-th interpolation point between the discrete points of the adjacent point pair are... i Satisfying the formula: Where N represents the preset quantity, U i V represents the coordinates of the i-th first allocation point. i This represents the coordinates of the i-th second allocation point. In this case, the allocation curve segment corresponding to the discrete point set that is closer to the corresponding discrete point in the adjacent point pair has a higher weight, which makes the transition between the obtained target curve segment and the adjacent curve segment smoother.

[0015] A second aspect of this disclosure provides an electronic device including at least one processing circuit configured to perform the steps of the annotation method described in the first aspect of this disclosure.

[0016] A third aspect of this disclosure provides a computer-readable storage medium storing at least one instruction that, when executed by a processor, implements the steps of the annotation method described in the first aspect of this disclosure.

[0017] According to this disclosure, a method, apparatus, and medium for annotating vascular ultrasound images that can obtain target curves that better fit the contour of the target region are provided. Attached Figure Description

[0018] This disclosure will now be explained in further detail by way of example only with reference to the accompanying drawings, in which:

[0019] Figure 1 This is a schematic diagram illustrating the application of the intravascular ultrasound system involved in the examples of this disclosure.

[0020] Figure 2 This is a schematic diagram showing multiple discrete points of the contour of a target region in a vascular ultrasound image involved in an example of this disclosure.

[0021] Figure 3 This is a flowchart illustrating an example of determining a target curve segment for adjacent point pairs as described in this disclosure.

[0022] Figure 4A This is a schematic diagram showing the target circle and the concave direction corresponding to the first discrete point set involved in the example of this disclosure.

[0023] Figure 4BThis is a schematic diagram showing the allocation curve segment corresponding to the first discrete point set involved in the example of this disclosure.

[0024] Figure 5A This is a schematic diagram showing the target circle and the concave direction corresponding to the second discrete point set involved in the example of this disclosure.

[0025] Figure 5B This is a schematic diagram showing the allocation curve segment corresponding to the second discrete point set involved in the example of this disclosure.

[0026] Figure 6 This is a schematic diagram illustrating the first vector and the second vector corresponding to the first discrete point set involved in the example of this disclosure.

[0027] Figure 7 This is a flowchart illustrating an example of blending based on two blending curve segments as described in this disclosure.

[0028] Figure 8 This is a schematic diagram illustrating the mixing based on two mixing curve segments as described in this disclosure example.

[0029] Figure 9 This is a flowchart illustrating an example of a method for annotating vascular ultrasound images as described in this disclosure.

[0030] Figure 10 It shows Figure 2 A schematic diagram of the target curve corresponding to multiple discrete points.

[0031] Figure 11 This is a block diagram illustrating an example of a labeling system for vascular ultrasound images as described in this disclosure.

[0032] Figure 12A This is a schematic diagram showing multiple discrete points of the contour of the tunica media on a vascular ultrasound image involved in the example of this disclosure.

[0033] Figure 12B This illustrates the results obtained using a parabolic interpolation scheme. Figure 12A A schematic diagram of the middle membrane profile corresponding to multiple discrete points.

[0034] Figure 12C This illustrates the solution obtained using the present disclosure. Figure 12A A schematic diagram of the middle membrane profile corresponding to multiple discrete points. Detailed Implementation

[0035] The preferred embodiments of this disclosure are described in detail below with reference to the accompanying drawings. In the following description, the same reference numerals are used for the same components, and repeated descriptions are omitted. Furthermore, the drawings are merely schematic diagrams, and the proportions of the components or the shapes of the components may differ from actual figures. It should be noted that the terms "comprising" and "having," and any variations thereof, in this disclosure, do not necessarily limit the process, method, system, product, or apparatus to the explicitly listed steps or units, but may include or have other steps or units not explicitly listed or inherent to these processes, methods, products, or apparatuses. All methods described in this disclosure may be performed in any suitable order unless otherwise indicated herein or clearly contradicted by the context.

[0036] The term "polar coordinate system" in this disclosure can refer to a coordinate system in a plane consisting of a pole, a polar axis, and a polar radius. Specifically, a point can be chosen in the plane, called the pole, and a ray can be drawn from the pole, called the polar axis. In this case, the coordinates of any point in the plane can be represented by the distance between that point and the pole and the angle between that point and the polar axis.

[0037] The scheme disclosed herein involves determining a target path for multiple discrete points. For each pair of adjacent discrete points in the target path, two additional discrete points are obtained as auxiliary points. Based on the two auxiliary points and the two adjacent discrete points forming two sets of discrete points (i.e., the subsequent two sets of discrete points) and the concave direction corresponding to each set of discrete points, a matching curve segment for the adjacent discrete points is obtained. The two matching curve segments are then matched to obtain the target curve segment for the adjacent discrete points, thereby determining the target curve for multiple discrete points. In some examples, the concave direction corresponding to the three discrete points in each set of discrete points can be determined based on their order and using a cross product. This allows for convenient and rapid determination of the concave direction.

[0038] Furthermore, the two auxiliary points mentioned above can be any discrete point in the target path other than each pair of adjacent discrete points. In some examples, the two auxiliary points can be discrete points located on either side of each pair of adjacent discrete points in the target path. Preferably, each of the two auxiliary points can be adjacent to one of the discrete points in each pair of adjacent discrete points. That is, the two auxiliary points can be discrete points located on either side of each pair of adjacent discrete points in the target path. In this case, the target curve segment can be determined based on the discrete points adjacent to each pair of adjacent discrete points.

[0039] The following description uses the annotation of vascular ultrasound images as an example to illustrate this disclosure, and such description does not limit the scope of this disclosure. Those skilled in the art can use other medical images without limitation. Furthermore, the solutions described in the examples of this disclosure can also be used in other fields besides medicine, such as path planning, to determine a smooth trajectory for multiple discrete points.

[0040] The examples of this disclosure relate to a method for annotating vascular ultrasound images (sometimes also referred to as an annotation method, target contour planning method, or target contour drawing method, hereinafter simply referred to as an annotation method). The annotation method described in this disclosure can obtain target curves that more closely fit the contour of the target region. Furthermore, it supports obtaining the target curve corresponding to the contour of the target region using fewer discrete points and with lower computational cost.

[0041] The vascular ultrasound images (also referred to as intravascular ultrasound images) described in this disclosure can be medical images of blood vessels obtained using IVUS technology. For example, vascular ultrasound images can be acquired using the intravascular ultrasound system described in this disclosure.

[0042] Figure 1 This is a schematic diagram illustrating the application of the intravascular ultrasound system 100 as described in this disclosure.

[0043] In some examples, such as Figure 1 As shown, the intravascular ultrasound system 100 may include an imaging device 110, a drive shaft 120, and a motor 130. The imaging device 110 may be placed inside the blood vessel and, while rotating, emit imaging signals toward the vessel wall to acquire intravascular ultrasound images. The drive shaft 120 may be connected to the imaging device 110 and extend outside the body along the blood vessel. The motor 130 may be located outside the body and may be connected to the drive shaft 120, actuating the imaging device 110 via the drive shaft 120.

[0044] In some examples, the imaging device 110 can be inserted into a blood vessel and can emit imaging signals toward the vessel wall. In some examples, the imaging device 110 can be an ultrasound transducer, and the imaging signal can be an ultrasound signal (e.g., an ultrasound pulse). In some examples, the ultrasound transducer can emit ultrasound pulses toward the vessel wall radially along the vessel and receive ultrasound echoes carrying vascular information. In other embodiments, the imaging signal can also be an optical signal, an electromagnetic wave signal, etc.

[0045] In some examples, motor 130 can actuate imaging device 110 via drive shaft 120. In some examples, motor 130 can actuate imaging device 110 to rotate via drive shaft 120. In some examples, motor 130 can actuate imaging device 110 to move along the extension direction of the blood vessel via drive shaft 120.

[0046] Additionally, the intravascular ultrasound system 100 may also include an outer tube (not shown). The outer tube may extend along the blood vessel to the outside of the body and has an inner cavity for accommodating the imaging instrument 110 and the drive shaft 120. The imaging instrument 110 can rotate within the inner cavity of the outer tube.

[0047] Figure 2 A schematic diagram of multiple discrete points representing the contour of a target region in a vascular ultrasound image involved in this disclosure is shown.

[0048] In some examples, the intravascular ultrasound system 100 may also include a processor 140 (see...) Figure 1 The processor 140 can be used to process vascular ultrasound images. In some examples, the intravascular ultrasound system 100 may also include a display 150 (see...). Figure 1 The display 150 can be used to display vascular ultrasound images to facilitate user operation of the images. For example, the user can annotate the displayed vascular ultrasound image to determine multiple discrete points defining the contour of a target region. As an example, Figure 2 This diagram illustrates the contour of the tunica media or intima of a blood vessel in a vascular ultrasound image using seven discrete points. These seven discrete points may include discrete points D1, D2, D3, D4, D5, D6, and D7. The following example will use these seven discrete points as an example.

[0049] The annotation method disclosed herein can obtain the target curve corresponding to the contour of the target region by determining multiple discrete points of the contour of the target region in a vascular ultrasound image, determining the target path of the multiple discrete points, and determining the target curve segment between each pair of adjacent discrete points in the target path.

[0050] Furthermore, the target path can be used to represent the adjacency relationships between discrete points. That is, the target path can be used to determine the discrete points adjacent to each other. In some examples, the target path can be closed. That is, the starting and ending discrete points in the target path can be the same. For example, for multiple discrete points of the contour of a target region in a vascular ultrasound image, since the target region is a cross-sectional area related to the blood vessel (e.g., the media and intima), the target path can be closed, and the resulting target curve can be a circular closed curve. As an example, Figure 2 A schematic diagram of the target path L1 for the aforementioned seven discrete points is shown. It should be noted that the diagram does not represent connecting line segments within the target path L1. Discrete points D1 and D2, D2 and D3, ..., D7 and D1 are two adjacent discrete points within the target path L1.

[0051] However, the examples disclosed herein are not limited to this; in other examples, the target path may not be closed. For example, a target curve segment may be determined between two discrete points that are partially (e.g., at least one pair) adjacent to each other in the target path, thereby obtaining the target curve corresponding to the contour of the target region.

[0052] Furthermore, in the target path, any three adjacent discrete points can be non-collinear. Therefore, the target circle can be determined based on these three discrete points.

[0053] As described above, in the annotation method of this disclosure, the target curve segment between each pair of adjacent discrete points can be determined. The method for determining the target curve segment between two adjacent discrete points according to the example of this disclosure (hereinafter referred to as the adjacent point connection method) will be described in detail below with reference to the accompanying drawings. For ease of description, each pair of adjacent discrete points will be referred to as an adjacent point pair. Figure 3 This is a flowchart illustrating an example of determining a target curve segment for adjacent point pairs as described in this disclosure.

[0054] The adjacent point connection method disclosed herein can determine a target curve segment between two intermediate discrete points (i.e., two adjacent discrete points) based on four discrete points. Preferably, the four discrete points can be adjacent discrete points in the target path. In this case, determining the target curve segment between the two adjacent discrete points by combining the discrete points before and after the two adjacent discrete points enables a smoother transition between the target curve segment and adjacent target curve segments, and obtains a target curve segment that better fits the contour of the target region.

[0055] like Figure 3 As shown, the method for connecting adjacent points may include step S102. In step S102, four discrete points, including pairs of adjacent points, can be obtained.

[0056] In some examples, the four discrete points may include adjacent point pairs and two auxiliary points. In some examples, two discrete points in the target path other than the discrete points of the adjacent point pairs can be taken as two auxiliary points to determine the four discrete points. In some examples, two discrete points on either side (e.g., before and after) of adjacent point pairs in the target path can be taken as two auxiliary points to determine the four discrete points.

[0057] Preferably, two discrete points adjacent to an adjacent point pair in the target path can be obtained as two auxiliary points to determine four discrete points. Specifically, two discrete points other than the discrete points of the adjacent point pair in the target path can be obtained as two auxiliary points, wherein each of the two auxiliary points can be adjacent to a discrete point of the adjacent point pair (that is, the two auxiliary points may not belong to the adjacent point pair). Thus, four discrete points including the adjacent point pair can be obtained. For example, the two auxiliary points may include a first auxiliary point and a second auxiliary point, wherein the first auxiliary point and the second auxiliary point may not belong to the adjacent point pair.

[0058] Taking an auxiliary point as an example, which is a discrete point located on both sides of an adjacent point pair and adjacent to that adjacent point pair, such as... Figure 2 As shown, for the pair of adjacent discrete points D1 and D2, the four corresponding discrete points can include discrete points D7, D1, D2 and D3, where discrete points D7 and D3 can be two auxiliary points.

[0059] In other examples, the system may accept user selection of discrete points to determine four discrete points. In still other examples, a suitable coordinate system (e.g., polar coordinates) may be selected to sort the discrete points and determine the four discrete points. In yet another example, the adjacency relationships between the discrete points may be pre-defined. That is, the four discrete points corresponding to each pair of adjacent points may be pre-defined.

[0060] Furthermore, for the case where there are only three discrete points in the target path, the two auxiliary points mentioned above can be the same discrete point. That is, the first auxiliary point and the second auxiliary point can be the same discrete point. Therefore, in subsequent steps, the two discrete point sets corresponding to each pair of adjacent points are the same, and thus the two allocation curve segments corresponding to the two discrete point sets are also the same. Finally, the target curve segment is the allocation curve segment (that is, regardless of whether the allocation is based on two allocation curve segments, the target curve segment is always the allocation curve segment).

[0061] like Figure 3 As shown, the method for connecting adjacent points may include step S104. In step S104, two discrete point sets can be determined based on the four discrete points obtained in step S102.

[0062] Furthermore, each discrete point set can include one of the two auxiliary points and a pair of adjacent points, and the auxiliary points in the two discrete point sets can be different auxiliary points. Specifically, the two discrete point sets can include a first discrete point set and a second discrete point set. The first discrete point set can include a first auxiliary point and a pair of adjacent points, and the second discrete point set can include a second auxiliary point and a pair of adjacent points. Thus, two discrete point sets can be obtained.

[0063] For example, such as Figure 2 As shown, for the adjacent pair of discrete points D1 and D2, the first set of discrete points may include discrete points D7, D1, and D2, and the second set of discrete points may include discrete points D1, D2, and D3. It should be noted that, unless otherwise specified, the examples involving the first and second sets of discrete points below assume that the first set includes discrete points D7, D1, and D2, and the second set includes discrete points D1, D2, and D3.

[0064] Furthermore, the three discrete points in the first set of discrete points are not collinear, and the three discrete points in the second set of discrete points are also not collinear. In some examples, if there are three collinear adjacent discrete points in the target path, the discrete points can be filtered to ensure that there are no collinear adjacent discrete points in the target path.

[0065] Figure 4A This is a schematic diagram showing the target circle U1 and the concave direction corresponding to the first discrete point set involved in the example of this disclosure. Figure 4B This is a schematic diagram showing the allocation curve segment Q1 corresponding to the first discrete point set involved in the example of this disclosure.

[0066] like Figure 3 As shown, the method for connecting adjacent points may include step S106. In step S106, the adjustment curve segment corresponding to each discrete point set can be determined based on the concave direction corresponding to each discrete point set.

[0067] Furthermore, the shape of the adjustment curve segment can be an arc. In some examples, the adjustment curve segment corresponding to each discrete point set can be determined based on the target circle and the concavity direction corresponding to each discrete point set. Specifically, the adjustment curve segment corresponding to each discrete point set can be an arc between discrete points of adjacent point pairs and belonging to the target circle.

[0068] Furthermore, the target circle corresponding to each discrete point set can be determined jointly by the discrete points in each set. In some examples, the center coordinates can be determined by the coordinates of three discrete points in each set, and then the target circle can be determined based on the center coordinates. As an example, Figure 4A The target circle U1, determined by discrete points D7, D1, and D2 in the first discrete point set, is shown.

[0069] Furthermore, the concavity direction corresponding to each discrete point set can be determined by the position of the discrete points in each set (i.e., the position of the discrete points in the target region) and a preset order. The concavity direction can be clockwise or counterclockwise. That is, the concavity direction can be the spatial distribution direction of the discrete points in each set under a preset order. Specifically, the concavity direction can be divided into clockwise and counterclockwise directions based on the position of the discrete points in each set and the preset order. The preset order can be any order based on the adjacency relationship between the discrete points in each set and can be selected as needed.

[0070] For example, such as Figure 4A As shown, for the first discrete point set, the preset order can be either a first order S1, where discrete point D7 → discrete point D1 → discrete point D2 is ordered sequentially, or a second order S2, where discrete point D2 → discrete point D1 → discrete point D7 is ordered sequentially. Therefore, from Figure 4A It can be seen that for the first sequence S1, the first concave direction F1 corresponding to the first discrete point set can be clockwise, and for the second sequence S2, the second concave direction F2 corresponding to the first discrete point set can be counterclockwise.

[0071] In some examples, the preset order of the discrete point sets corresponding to all adjacent point pairs can be consistent. That is, a uniform order is used. For example, the preset order can be determined by any direction along the target path. In other examples, the preset order of the discrete point sets can be inconsistent.

[0072] As described above, the allocation curve segment corresponding to each discrete point set can be determined based on the concave direction corresponding to each discrete point set. Specifically, based on the concave direction, one arc can be selected from the two arcs located between discrete points in the target circle corresponding to each discrete point set as the allocation curve segment corresponding to each discrete point set.

[0073] In some examples, within the target circle corresponding to each discrete point set, arcs arranged in a predetermined order from one discrete point to another in the concave direction of adjacent point pairs can be used as the allocation curve segments for each discrete point set. This allows the allocation curve segments to be determined based on the concave direction. For example, as... Figure 4B As shown, the arc from discrete point D1 to discrete point D2 (i.e., adjacent point pairs) in the first discrete point set in the first order S1, and in the first concave direction F1 (i.e., clockwise direction), can be used as the matching curve segment Q1 corresponding to the first discrete point set.

[0074] The above description uses the first discrete point set as an example to illustrate the target circle, concavity direction, and adjustment curve segment corresponding to the discrete point set. Similarly, the target circle, concavity direction, and adjustment curve segment corresponding to the second discrete point set can be obtained. Figure 5A This is a schematic diagram showing the target circle U2 and the concavity direction corresponding to the second discrete point set involved in the example of this disclosure. Figure 5B This is a schematic diagram showing the tuning curve segment Q2 corresponding to the second discrete point set involved in the example of this disclosure.

[0075] like Figure 5A As shown, for the second discrete point set, the preset order can be a third order S3, where discrete point D1 → discrete point D2 → discrete point D3 is ordered sequentially, or a fourth order S4, where discrete point D3 → discrete point D2 → discrete point D1 is ordered sequentially. Therefore, from Figure 5A It can be seen that for the third sequence S3, the third concave direction F3 corresponding to the second discrete point set can be counterclockwise, and for the fourth sequence S4, the fourth concave direction F4 corresponding to the second discrete point set can be clockwise.

[0076] Additionally, an example of the allocation curve segment corresponding to the second discrete point set can be found in [reference needed]. Figure 5B .like Figure 5B As shown, the arc from discrete point D1 to discrete point D2 (i.e., adjacent point pairs) in the second discrete point set in the third preset order S3 and in the third concave direction F3 (i.e., counterclockwise direction) can be used as the matching curve segment Q2 corresponding to the second discrete point set.

[0077] As described above, the concavity direction corresponding to each discrete point set can be determined by the position and preset order of the discrete points in each discrete point set. Those skilled in the art can adjust the method for obtaining the concavity direction as needed without deviating from the definition of the concavity direction. Preferably, the examples of this disclosure also provide two methods for obtaining the concavity direction.

[0078] Figure 6 This illustrates the first vector corresponding to the first discrete point set involved in the example of this disclosure. Second vector A schematic diagram.

[0079] In some examples, both methods for obtaining the direction of the depression can first determine the first and second vectors corresponding to each discrete point set.

[0080] Specifically, the first vector can be defined as the vectors corresponding to the first discrete point to the second discrete point in each discrete point set, arranged in the aforementioned preset order. The second vector can also be defined as the vectors corresponding to the first discrete point to the third discrete point in each discrete point set, arranged in the aforementioned preset order. Furthermore, the first and second vectors can share a common starting point. Taking the first discrete point set as an example... Figure 6As shown, following the first order S1 (i.e., discrete point D7 → discrete point D1 → discrete point D2), the vectors corresponding to the first discrete point D7 to the second discrete point D1 in the first discrete point set can be the first vector. The vectors corresponding to the first discrete point D7 to the third discrete point D2 in the first discrete point set can be the second vector.

[0081] In some examples, in the first method of obtaining the concavity direction, after determining the first and second vectors corresponding to each discrete point set, the concavity direction can be obtained based on the cross product.

[0082] Specifically, the cross product of the first and second vectors can be used as the third vector. If the direction of the third vector is upward (or positive), it indicates that the concavity direction is counterclockwise; if the direction of the third vector is downward (or negative), it indicates that the concavity direction is clockwise. Thus, the concavity direction can be conveniently determined through the cross product.

[0083] In some examples, in the second method of obtaining the concavity direction, after determining the first and second vectors corresponding to each discrete point set, the concavity direction can be obtained based on the angle between the vectors.

[0084] Specifically, if the angle between the first vector and the second vector when rotated clockwise is less than 180°, the concavity direction is clockwise; conversely, if the angle between the first vector and the second vector when rotated counterclockwise is less than 180°, the concavity direction is counterclockwise. Therefore, the concavity direction can be determined by the angle between the vectors. Taking the first discrete point set as an example... Figure 6 As shown, the first vector Rotate clockwise to the second vector The included angle θ1 is less than 180°, therefore, the concave direction can be clockwise. That is, the arc from discrete points D1 to D2 in the clockwise direction can be taken as the matching curve segment corresponding to the first discrete point set (i.e., Figure 4B (Q1 segment of the adjustment curve in the diagram). Therefore, the direction of the depression can be easily determined by the angle between the vectors.

[0085] Return to reference Figure 3 The method for connecting adjacent points may include step S108. In step S108, two matching curve segments corresponding to two discrete point sets can be matched to obtain the target curve segment (hereinafter referred to as the target curve segment) of the adjacent point pair.

[0086] As described above, two discrete point sets can be determined. Therefore, based on the two discrete point sets, a matching curve segment can be determined for each adjacent point pair, wherein the two endpoints of each matching curve segment are discrete points in the adjacent point pair. Thus, two matching curve segments corresponding to the two discrete point sets can be obtained.

[0087] In some examples, during allocation, the target curve segment can be obtained by setting the proportions of the two allocation curve segments at different positions. In some examples, the target curve segment can be obtained based on the interpolation points between discrete points of adjacent point pairs. Specifically, during allocation, the contribution of the allocation points on the two allocation curve segments to the interpolation points forming the target curve segment can be adjusted by weighting to obtain the target curve segment.

[0088] As described above, the center coordinates of a circle can be determined from the coordinates of three discrete points in each set of discrete points, thus determining the target circle. In some examples, the arc angle of the allocation curve segment corresponding to each set of discrete points can be determined based on the center coordinates. Then, the step size is determined based on the arc angle and the number of allocation points, and allocation points are inserted evenly at this step size. That is, the allocation point corresponding to the allocation curve segment can be seen as obtaining one discrete point of an adjacent pair of points by rotating it along the allocation curve segment around the center coordinates according to the step size to another discrete point. In some examples, the step size can be the arc angle divided by the number of allocation points.

[0089] In some examples, the arc angle of the curve segment can be the absolute value of the difference between the arc angles of two discrete points in an adjacent point pair.

[0090] In some examples, if multiple discrete points are in a rectangular coordinate system, the i-th allocation point Z on each allocation curve segment... i (X i ,Y i The formula can be satisfied:

[0091] X i =(X1-X o )×cos(i×step)-(Y1-Y o )×sin(i×step)+X o ,

[0092] Y i =(X1-Xo)×sin(i×step)+(Y1-Yo)×cos(i×step)+Y o ,

[0093] Among them, (X) i ,Y i ) can represent the i-th allocation point Z on the allocation curve segment. iThe coordinates (Xo,Yo) can represent the center coordinates of the circle (that is, the center coordinates of the circle determined by the coordinates of three discrete points in the set of discrete points corresponding to the adjustment curve segment), (X1,Y1) can represent the coordinates of any discrete point in the adjacent point pair, and step can represent the step size.

[0094] The following combination Figure 7 An exemplary method for mixing based on two mixing curve segments is described. Figure 7 This is a flowchart illustrating an example of blending based on two blending curve segments as described in this disclosure. Figure 8 This is a schematic diagram illustrating the mixing based on two mixing curve segments as described in this disclosure example.

[0095] like Figure 7 As shown, the allocation based on two allocation curve segments may include step S202. In step S202, any discrete point in an adjacent point pair can be used as the starting point.

[0096] Continue to refer to Figure 7 The allocation based on two allocation curve segments may include step S204. In step S204, starting from the aforementioned starting point, a predetermined number of first allocation points distributed along the first curve segment and a predetermined number of second allocation points distributed along the second curve segment are determined. The two allocation curve segments may include a first curve segment and a second curve segment. That is, the first curve segment can be one of the two allocation curve segments, and the second curve segment can be the other allocation curve segment.

[0097] In some examples, the first and second allocation points can be evenly distributed on the first and second curve segments, respectively. Specifically, starting from the aforementioned starting point, a predetermined number (e.g., 200) of the first allocation points can be evenly inserted on the first curve segment, and a predetermined number of the second allocation points can be evenly inserted on the second curve segment. That is, starting from the same starting point, the same number of allocation points can be evenly inserted on the first and second curve segments.

[0098] Continue to refer to Figure 7 The process of adjusting curve segments based on two adjustment curve segments may include step S206. In step S206, interpolation points between discrete points of adjacent point pairs can be determined based on the first and second adjustment points, and target curve segments of adjacent point pairs can be obtained based on the interpolation points. In this case, adjustment points are obtained based on the two adjustment curve segments determined by the four discrete points corresponding to the adjacent point pairs, and interpolation points are obtained based on the adjustment points, thereby obtaining the target curve segments. This allows the target curve segments to better fit the contour of the target region. In some examples, the interpolation points can be connected sequentially (e.g., by fitting) to obtain the target curve segments of adjacent point pairs.

[0099] In some examples, the corresponding first and second allocation points (i.e., those at the same position relative to the starting point or with the same index relative to the starting point) can be weighted (i.e., multiplied by a weight) to obtain the corresponding interpolation points. In some examples, the weights of the first and second allocation points can be negatively correlated. That is, the higher the weight of the first allocation point, the lower the weight of the second allocation point, or vice versa. In some examples, the sum of the weights of the first and second allocation points can equal 1.

[0100] In some examples, the first curve segment can be a relocation curve segment corresponding to a set of discrete points whose auxiliary points are adjacent to the aforementioned starting point, and the second curve segment can be a relocation curve segment corresponding to a set of discrete points whose auxiliary points are not adjacent to the aforementioned starting point. For example, as... Figure 8 As shown, if discrete point D1 is taken as the starting point, the allocation curve segment Q1 corresponding to the set of discrete points adjacent to discrete point D1 (i.e., the first set of discrete points) can be the first curve segment, and the allocation curve segment Q2 corresponding to the set of discrete points not adjacent to discrete point D1 (i.e., the second set of discrete points) can be the second curve segment. Similarly, if discrete point D2 is taken as the starting point, then allocation curve segment Q2 can be the first curve segment, and allocation curve segment Q1 can be the second curve segment.

[0101] In this scenario, starting from the aforementioned starting point, the weight of the first interpolation point can decrease linearly, and the weight of the second interpolation point can increase linearly. Taking a target region in a vascular ultrasound image as an example, in some examples, the weight of the first interpolation point can be (Ni) / N, and the weight of the second interpolation point can be i / N, where N represents the aforementioned preset number, and i represents the index of the interpolation point. Specifically, the coordinates M of the i-th interpolation point between discrete points of adjacent point pairs... i The formula can be satisfied:

[0102]

[0103] Where N can represent the aforementioned preset quantity, U i V can represent the coordinates of the i-th first allocation point (i.e., the i-th allocation point on the first curve segment). iThis can represent the coordinates of the i-th second allocation point (i.e., the i-th allocation point on the second curve segment). In this case, the weight of the allocation curve segment corresponding to the discrete point set closer to the corresponding discrete point in the adjacent point pair is higher, thus making the transition between the obtained target curve segment and the adjacent curve segment smoother. In addition, the method of obtaining the coordinates of the interpolation points involved in the target region in the above-mentioned vascular ultrasound image is also applicable to other images of roughly circular target regions, or roughly circular target regions deformed by the pushing and pulling of several points.

[0104] As an example, Figure 8 The target curve segment Q3 is shown as obtained by adjusting the curve segment Q1 corresponding to the first discrete point set and the curve segment Q2 corresponding to the second discrete point set.

[0105] It should be noted that when applied to target areas outside of vascular ultrasound images, the weights of the first and second allocation points can be adjusted according to the shape of the target, which does not imply any limitation on this disclosure.

[0106] Figure 9 This is a flowchart illustrating an example of a method for annotating vascular ultrasound images according to an example of this disclosure. The annotation method according to an example of this disclosure can determine the target curve segment of two adjacent discrete points (also referred to as adjacent point pairs) using the adjacent point connection method described above.

[0107] like Figure 9 As shown, in this embodiment, the annotation method may include step S302. In step S302, multiple discrete points of the contour of the target region in the vascular ultrasound image can be determined.

[0108] In some examples, the user can annotate the contour of a target region in a vascular ultrasound image to determine multiple discrete points of the target region's contour. Specifically, the display 150 can display a vascular ultrasound image, and the user can annotate the displayed vascular ultrasound image to acquire multiple discrete points of the target region's contour. However, the examples of this disclosure are not limited to this; in other examples, the multiple discrete points of the target region's contour can also be discrete points on the contour obtained through a target segmentation algorithm. In still other examples, the multiple discrete points of the target region's contour can also be discrete points pre-stored in a storage medium.

[0109] In some examples, the number of discrete points can be greater than or equal to 3. For example, the number of discrete points can be 3, 4, 5, 6, 7, 8, 9, 10, 12, 14, 16, 18, or 20, etc. Furthermore, for a total of 3 discrete points, as mentioned above, regardless of whether the adjustment is based on two adjustment curve segments, the target curve segment is the adjustment curve segment. Therefore, the final target curve determined by the subsequent 3 discrete points can be the circle corresponding to those 3 discrete points.

[0110] Furthermore, for vascular ultrasound images, the target region can include at least one target region within the tunica media and tunica intima (also known as the lumen) of the blood vessel. This allows for the subsequent determination of the target curves corresponding to the contours of the regions within the tunica media and tunica intima of the blood vessel.

[0111] In some examples, preprocessing can be performed on multiple discrete points. In some examples, preprocessing may include removing outliers from multiple discrete points. Outliers can be discrete points that do not conform to the contour features of the target region. In this case, the accuracy of the target curve can be improved subsequently.

[0112] like Figure 9 As shown, in this embodiment, the annotation method may include step S304. In step S304, the target path for multiple discrete points can be determined.

[0113] As mentioned above, a target path can be used to represent the adjacency relationships between discrete points. In some examples, the target path can be closed. Therefore, multiple discrete points can be represented as a circular array, where if k = 0 represents the first discrete point, then k-1 represents the last discrete point. In other examples, the target path may not be closed.

[0114] Furthermore, target paths for multiple discrete points can be determined in any way. In some examples, the target paths for multiple discrete points can be pre-set. For instance, user settings for the adjacency relationships of multiple discrete points can be received to determine the target paths. In some examples, multiple discrete points can be sorted to determine the target paths.

[0115] Preferably, the target path for multiple discrete points can be determined using a polar coordinate system. Specifically, a polar coordinate system can be established; the coordinates of each discrete point can be converted to polar coordinates; and the multiple discrete points can be sorted according to the angles of their corresponding polar coordinates to determine the target path. In this case, by sorting the multiple discrete points on the plane using a polar coordinate system, the resulting target path can better conform to the contour features of the target region. This allows for convenient determination of the connection relationships (or connection order) between the discrete points.

[0116] In some examples, the coordinates of each discrete point can be in a Cartesian coordinate system. In other examples, for multiple discrete points with Cartesian coordinates, in establishing a polar coordinate system, the coordinates of the geometric center of these discrete points can be obtained. Using the coordinates of the geometric center as the pole and a preset direction as the positive direction of the polar axis, a polar coordinate system can be established. Thus, a polar coordinate system can be established.

[0117] Furthermore, the preset direction can be any direction. Preferably, the preset direction can be horizontal to the right. Additionally, for multiple discrete points in a Cartesian coordinate system, the preset direction can also be the positive direction of the X-axis. It should be noted that although the target path is determined by converting to a polar coordinate system, subsequent steps (e.g., step S306) can still use the original coordinate system (e.g., Cartesian coordinate system) as needed.

[0118] In some examples, after determining the target path for multiple discrete points, if there are three collinear discrete points in the target path, the discrete points can be filtered to ensure that there are no three collinear discrete points in the target path, and the target curve can be obtained based on the filtered discrete points.

[0119] Figure 10 It shows Figure 2 A schematic diagram of the target curve Q4 corresponding to multiple discrete points.

[0120] like Figure 9 As shown, in this embodiment, the annotation method may include step S306. In step S306, the target curve segment between each pair of adjacent discrete points can be determined in the target path, thereby obtaining the target curve corresponding to the contour of the target region. As an example, Figure 10 It shows Figure 2 The target curve Q4 corresponds to multiple discrete points in the curve.

[0121] In some examples, starting from any discrete point among multiple discrete points along any direction of the target path, the target curve segment between each pair of adjacent discrete points can be determined sequentially, thereby obtaining the target curve corresponding to the contour of the target region. That is, the step of determining the target curve segment between different adjacent discrete points can be repeated until all adjacent discrete points in the target path have been traversed. In some examples, the target curve segment between adjacent discrete points can be determined using the adjacent point connection method described above.

[0122] In some examples, the annotation method may also include receiving operations to delete or add discrete points to update the target curve (not illustrated). Specifically, it may receive user operations to delete or add discrete points, and update the target curve segment associated with the deleted or added discrete points (e.g., deleting the target curve segment, or redefining the target curve segment between two adjacent discrete points). This allows for convenient updating of poorly annotated discrete points, complex contours, or irregular contours to obtain a more accurate target curve.

[0123] The following combination Figure 11This disclosure describes a labeling system 200 for vascular ultrasound images (sometimes referred to as a labeling system 200, target contour planning system, or target contour drawing system, hereinafter simply referred to as labeling system 200). Labeling system 200 can be used to implement the labeling methods described above. It should be noted that, unless otherwise specified, the descriptions of the labeling methods described above also apply to labeling system 200. Figure 11 This is a block diagram illustrating an example of a vascular ultrasound image annotation system 200 as described in this disclosure.

[0124] In some examples, such as Figure 11 As shown, the annotation system 200 may include an acquisition module 202, a path planning module 204, and an adjacent point connection module 206.

[0125] In this embodiment, the acquisition module 202 can be used to determine multiple discrete points of the contour of the target region in the vascular ultrasound image. For details, please refer to the relevant description in step S302, which will not be repeated here.

[0126] In this embodiment, the path planning module 204 can be used to determine the target path for multiple discrete points. For details, please refer to the relevant description in step S304, which will not be repeated here.

[0127] In this embodiment, the adjacent point connection module 206 can be used to determine the target curve segment between each pair of adjacent discrete points in the target path, thereby obtaining the target curve corresponding to the contour of the target region. For details, please refer to the relevant description in step S306, which will not be repeated here.

[0128] This disclosure also relates to an electronic device (not shown) that may include at least one processing circuit. The at least one processing circuit is configured to perform one or more steps of the above-described annotation method.

[0129] This disclosure also relates to a computer-readable storage medium that may store at least one instruction that, when executed by a processor, implements one or more steps of the annotation method described above.

[0130] The annotation method, device, and medium for vascular ultrasound images disclosed herein involve determining multiple discrete points of the contour of a target region in a vascular ultrasound image and the target path of these discrete points. It then determines the target curve segment between each pair of adjacent discrete points along the target path, thereby obtaining the target curve corresponding to the contour of the target region. Furthermore, within the target curve segment of the adjacent discrete points, it obtains four adjacent discrete points corresponding to the two adjacent discrete points. Based on the first three discrete points and their corresponding concavity directions, and the last three discrete points and their corresponding concavity directions, it obtains adjustment curve segments for each of the adjacent discrete points. These two adjustment curve segments are then adjusted to obtain the target curve segment for the adjacent discrete points. In this case, by adjusting the arcs of adjacent discrete points to obtain the target curve segment passing through the adjacent discrete points, a target curve that fits the contour of the target region can be obtained. Additionally, it supports obtaining the target curve corresponding to the contour of the target region using fewer discrete points with less computation. Furthermore, the concavity direction is determined using a cross product. Therefore, the concavity direction can be determined conveniently and quickly.

[0131] Figure 12A This is a schematic diagram showing multiple discrete points of the contour of the tunica media on a vascular ultrasound image involved in the example of this disclosure. Figure 12B This illustrates the results obtained using a parabolic interpolation scheme. Figure 12A A schematic diagram of the middle membrane profile Q5 corresponding to multiple discrete points. Figure 12C This illustrates the solution obtained using the present disclosure. Figure 12A A schematic diagram of the middle membrane profile Q6 corresponding to multiple discrete points.

[0132] In addition, to verify the effectiveness of the annotation method involved in the examples of this disclosure, a comparison is made with the parabolic interpolation scheme. Figure 12A The diagram shows six discrete points selected for the contour of the middle membrane. These six discrete points may include discrete points D8, D9, D10, D11, D12, and D13. Figure 12B The mid-film profile Q5 is drawn for the parabolic interpolation scheme of the prior art. Figure 12C The membrane profile Q6 is plotted for the scheme of this disclosure. It is evident that the membrane profile plotted by the scheme of this disclosure is more rounded and accurate than that plotted by parabolic interpolation, thus providing accurate data for quantitative measurements (e.g., accurate calculation of plaque load).

[0133] While the present disclosure has been specifically described above in conjunction with the accompanying drawings and examples, it is to be understood that the foregoing description does not limit the present disclosure in any way. Those skilled in the art can make modifications and variations to the present disclosure as needed without departing from its essential spirit and scope, and all such modifications and variations shall fall within the scope of the present disclosure.

Claims

1. A labeling system for vascular ultrasound images, characterized in that, The annotation system includes: an acquisition module, a path planning module, and an adjacent point connection module; the acquisition module is used to determine multiple discrete points of the contour of the target region in the vascular ultrasound image; the path planning module is used to determine the target path of the multiple discrete points, the target path being used to represent the adjacency relationship between the discrete points; the adjacent point connection module is used to determine the target curve segment between each pair of adjacent discrete points in the target path, thereby obtaining the target curve corresponding to the contour of the target region, wherein each pair of adjacent discrete points is called an adjacent point pair, and two discrete points other than each adjacent point pair are acquired as two auxiliary points, each auxiliary point being adjacent to one discrete point of the adjacent point pair, and the two auxiliary points forming two discrete point sets with the adjacent point pair respectively, based on each The concave direction corresponding to each discrete point set is used to obtain the adjustment curve segment for the adjacent point pair. The adjustment curve segment is between the discrete points of the adjacent point pair and belongs to the target circle. The center coordinates of the target circle are determined by the coordinates of three discrete points in each discrete point set. The concave direction is determined based on the position and preset order of the three discrete points. Based on the concave direction, one arc is selected from the two arcs located between the discrete points of the adjacent point pair in the target circle corresponding to each discrete point set as the adjustment curve segment corresponding to each discrete point set. The three discrete points in each discrete point set are located on the corresponding target circle. The preset order is based on the adjacency relationship between the discrete points in each discrete point set. The two adjustment curve segments corresponding to the two discrete point sets are adjusted to obtain the target curve segment.

2. The annotation system for vascular ultrasound images according to claim 1, characterized in that, The target path is closed. The adjacent point connection module is used to determine the target curve segment of each pair of adjacent points in any direction along the target path, starting from any discrete point in the target path, and then obtain the target curve.

3. The annotation system for vascular ultrasound images according to claim 1, characterized in that, The adjacent point connection module is also used to receive operations to delete discrete points or add discrete points to update the target curve.

4. The annotation system for vascular ultrasound images according to claim 1, characterized in that, The target region includes at least one target region of the tunica media and tunica intima of the blood vessel.

5. The annotation system for vascular ultrasound images according to claim 1, characterized in that, The adjacent point connection module is used to, in the target circle corresponding to each discrete point set, take the arc from one discrete point of the adjacent point pair to another discrete point in the preset order, and in the concave direction, as the adjustment curve segment corresponding to each discrete point set.

6. The annotation system for vascular ultrasound images according to claim 1, characterized in that, The target curve segment is obtained based on the interpolation points between discrete points of the adjacent point pairs, wherein the contribution of the adjustment points on the two adjustment curve segments to the interpolation points forming the target curve segment is adjusted by weights to obtain the target curve segment.

7. The annotation system for vascular ultrasound images according to claim 1 or 6, characterized in that, The two allocation curve segments include a first curve segment and a second curve segment. The first curve segment is the allocation curve segment corresponding to the set of discrete points adjacent to the starting point and the auxiliary points in the two discrete point sets. The second curve segment is the allocation curve segment corresponding to the set of discrete points not adjacent to the starting point and the auxiliary points in the two discrete point sets. The weight of the first allocation point in the first curve segment is... The weight of the second adjustment point in the second curve segment is ,in, Indicates the number of interpolation points. The index represents the interpolation point, where the starting point is any discrete point in the adjacent point pair.

8. The annotation system for vascular ultrasound images according to claim 1, characterized in that, The direction of the depression is determined based on the position of each discrete point in the set of discrete points and the preset order, including: Let the first vector be the vector corresponding to the first discrete point to the second discrete point in each set of discrete points, arranged in the preset order; and let the second vector be the vector corresponding to the first discrete point to the third discrete point in each set of discrete points, arranged in the preset order. The cross product of the first and second vectors is taken as the third vector. If the direction of the third vector is upward, it indicates that the concave direction is counterclockwise; if the direction of the third vector is downward, it indicates that the concave direction is clockwise. If the angle between the first vector and the second vector when the first vector rotates clockwise is less than 180°, it indicates that the direction of the concavity is clockwise. If the angle between the first vector and the second vector when the first vector rotates counterclockwise is less than 180°, it indicates that the direction of the concavity is counterclockwise.

9. The annotation system for vascular ultrasound images according to claim 1, characterized in that, Determining the target path of the plurality of discrete points using a polar coordinate system includes: establishing the polar coordinate system, converting the coordinates of each discrete point into coordinates based on the polar coordinate system, and sorting the plurality of discrete points according to the coordinate angles of the polar coordinate system corresponding to each discrete point to determine the target path.

Images