Blood vessel skeleton extraction method and device, storage medium and electronic device

Abstract

The present application relates to the field of data processing technology, and specifically to a vascular skeleton extraction method and a vascular skeleton extraction device, as well as a computer-readable storage medium and an electronic device, which solve the problem that false connections are prone to occur in the vascular skeleton extraction method. The vascular skeleton extraction method provided in the embodiment of the present application performs distance transformation on the vascular image sequence to be processed, generates a thermal map sequence corresponding to the vascular image sequence to be processed, and then performs skeleton extraction based on the thermal map sequence corresponding to the vascular image sequence to be processed to obtain the skeleton extraction result corresponding to the vascular image sequence to be processed, so that the skeleton extraction can be performed with reference to the M distance value sets in the thermal map sequence, so as to realize skeleton extraction layer by layer, so as to reduce the false connections generated by the skeleton extraction of each layer, thereby reducing the false connections in the skeleton extraction result corresponding to the vascular image sequence to be processed.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a method and device for extracting a vascular skeleton, as well as a computer-readable storage medium and an electronic device. Background Art

[0002] The skeleton is an abstract description of the connected shapes of the region. Skeleton extraction can effectively reduce redundant information and retain the geometric shape and topological structure of the spatial shape. Since there are many types of blood vessels and their shapes are diverse, skeleton extraction of blood vessels is performed to reduce redundant information and retain the topological structure of blood vessels, which can provide an accurate data basis for subsequent blood vessel segmentation. Existing vascular skeleton extraction methods all use thinning schemes. The process of the thinning scheme is: (1) obtain edge points; (2) determine whether the edge points can be eliminated, and repeat steps (1) and (2) until no edge points can be eliminated. The output result of the thinning scheme needs to meet the principle of thickness 1. However, the thinning scheme is prone to the problem of false connection. Summary of the invention

[0003] In view of this, the embodiments of the present application provide a vascular skeleton extraction method and a vascular skeleton extraction device, as well as a computer-readable storage medium and an electronic device, which solve the problem that the vascular skeleton extraction method is prone to false connections.

[0004] In a first aspect, an embodiment of the present application provides a method for extracting a blood vessel skeleton, comprising: performing a distance transform on a blood vessel image sequence to be processed to generate a heat map sequence corresponding to the blood vessel image sequence to be processed, the heat map sequence comprising a blood vessel image and M distance value sets annotated on the blood vessel image, each distance value set comprising N distance values ​​of equal size, and M and N are both positive integers; performing skeleton extraction based on the heat map sequence corresponding to the blood vessel image sequence to be processed to obtain a skeleton extraction result corresponding to the blood vessel image sequence to be processed.

[0005] In combination with the first aspect of the present application, in some embodiments, in a heat map sequence, each pixel forming a vascular image corresponds to a distance value, and skeleton extraction is performed based on the heat map sequence corresponding to the vascular image sequence to be processed to obtain a skeleton extraction result corresponding to the vascular image sequence to be processed, including: for P pixels corresponding to each distance value set in M ​​distance value sets, based on a skeletonization algorithm, determining Q skeleton pixels among the P pixels, where P is a positive integer and Q is a positive integer less than or equal to P; determining false connection pixels among the Q skeleton pixels, where the false connection pixels are pixels corresponding to vascular adhesion areas in the vascular image; and removing the false connection pixels among the Q skeleton pixels to obtain a skeleton extraction result corresponding to the vascular image sequence to be processed.

[0006] In combination with the first aspect of the present application, in some embodiments, determining a falsely connected pixel among Q skeleton pixels includes: determining D endpoint pixels among the Q skeleton pixels, D being a positive integer less than or equal to Q; for each skeleton pixel among the Q skeleton pixels, determining whether the current skeleton pixel is connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels, where Q1 is a positive integer less than Q; if the current skeleton pixel is not connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels, determining the current skeleton pixel as a falsely connected pixel.

[0007] In combination with the first aspect of the present application, in some embodiments, determining D endpoint pixels among Q skeleton pixels includes: determining a first judgment pixel set based on the Q skeleton pixels and pixels corresponding to all distance value sets except the current distance value set; determining D endpoint pixels based on the first judgment pixel set and preset endpoint judgment conditions.

[0008] In combination with the first aspect of the present application, in some embodiments, determining whether a current skeleton pixel is connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels includes: determining a second judgment pixel set based on the Q skeleton pixels; and determining whether the current skeleton pixel is connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels based on the second judgment pixel set and a preset connectivity judgment condition.

[0009] In combination with the first aspect of the present application, in some embodiments, a distance transform is performed on the vascular image sequence to be processed to generate a thermal map sequence corresponding to the vascular image sequence to be processed, including: using a classification model to perform a distance transform on the vascular image sequence to be processed to generate a thermal map sequence corresponding to the vascular image sequence to be processed.

[0010] In combination with the first aspect of the present application, in some embodiments, a classification model is used to perform a distance transformation on a vascular image sequence to be processed to generate a thermal map sequence corresponding to the vascular image sequence to be processed, including: slicing the vascular image sequence to be processed to obtain a plurality of image blocks to be processed, wherein the resolution of the image blocks to be processed is the same as the physical resolution corresponding to the vascular image sequence to be processed; using a classification model, a distance transformation is performed on the plurality of image blocks to be processed to generate a thermal map sequence corresponding to each of the plurality of image blocks to be processed; based on the thermal map sequences corresponding to each of the plurality of image blocks to be processed, a thermal map sequence corresponding to the vascular image sequence to be processed is generated.

[0011] In a second aspect, an embodiment of the present application provides a vascular skeleton extraction device, comprising: a heat map generation module, configured to perform a distance transformation on a vascular image sequence to be processed, and generate a heat map sequence corresponding to the vascular image sequence to be processed, the heat map sequence comprising a vascular image and M distance value sets annotated on the vascular image, each distance value set comprising N distance values ​​of equal size, and M and N are both positive integers; a skeleton extraction module, configured to perform skeleton extraction based on the heat map sequence corresponding to the vascular image sequence to be processed, and obtain a skeleton extraction result corresponding to the vascular image sequence to be processed.

[0012] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, which stores instructions. When the instructions are executed by a processor of an electronic device, the electronic device can execute the blood vessel skeleton extraction method mentioned in the first aspect.

[0013] In a fourth aspect, an embodiment of the present application provides an electronic device, the electronic device comprising: a processor; a memory for storing computer executable instructions; a processor, for executing computer executable instructions to implement the vascular skeleton extraction method mentioned in the first aspect above.

[0014] The blood vessel skeleton extraction method provided in the embodiment of the present application performs distance transformation on the blood vessel image sequence to be processed, generates a heat map sequence corresponding to the blood vessel image sequence to be processed, and then performs skeleton extraction based on the heat map sequence corresponding to the blood vessel image sequence to be processed to obtain a skeleton extraction result corresponding to the blood vessel image sequence to be processed, so that the skeleton extraction can be performed with reference to the M distance value sets in the heat map sequence to realize skeleton extraction layer by layer, so as to reduce the false connections generated by the skeleton extraction of each layer, thereby reducing the false connections in the skeleton extraction result corresponding to the blood vessel image sequence to be processed. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The figure is a schematic diagram of an application scenario of a blood vessel skeleton extraction method provided in an embodiment of the present application.

[0016] Figure 2 FIG. 1 is a flow chart of a blood vessel skeleton extraction method provided in an embodiment of the present application.

[0017] Figure 3 Shown is a sequence of blood vessel images to be processed provided by an embodiment of the present application.

[0018] Figure 4 Shown Figure 3 A frame of image in the thermal map sequence corresponding to the blood vessel image sequence to be processed is shown.

[0019] Figure 5 Shown are the skeleton extraction results obtained using the thinning scheme.

[0020] Figure 6 Shown is the skeleton extraction result provided by an embodiment of the present application.

[0021] Figure 7 Shown is a schematic flow chart of a blood vessel skeleton extraction method provided in another embodiment of the present application.

[0022] Figure 8 Shown is a frame of image in a thermal map sequence provided by an embodiment of the present application.

[0023] Fig. 9 The skeletonization algorithm is shown in Figure 8 The image shown is the skeletonized image of the black area.

[0024] Fig.10 Shown is removal Fig. 9 The resulting image after pseudo-connecting pixels in the illustrated embodiment.

[0025] Fig.11 Shown is a skeleton extraction result corresponding to a blood vessel image sequence to be processed provided by an embodiment of the present application.

[0026] Fig.12 Shown is another skeleton extraction result obtained using the thinning scheme.

[0027] Fig.13 Shown is a schematic flow chart of a blood vessel skeleton extraction method provided in another embodiment of the present application.

[0028] Fig.14 Endpoint pixels are shown.

[0029] Fig.15 Shown are line segment point pixels.

[0030] Fig.16 Shown are node pixels.

[0031] Fig.17 Shown is a schematic flow chart of a blood vessel skeleton extraction method provided in another embodiment of the present application.

[0032] Fig.18 Shown is a schematic flow chart of a blood vessel skeleton extraction method provided in another embodiment of the present application.

[0033] Fig.19 Shown is a schematic flow chart of a blood vessel skeleton extraction method provided in another embodiment of the present application.

[0034] Fig. 20 Shown is a schematic flow chart of a blood vessel skeleton extraction method provided in another embodiment of the present application.

[0035] Fig.21Shown is a schematic structural diagram of a blood vessel skeleton extraction device provided in one embodiment of the present application.

[0036] Fig. 22 Shown is a schematic structural diagram of a blood vessel skeleton extraction device provided in another embodiment of the present application.

[0037] Fig.23 Shown is a schematic structural diagram of a blood vessel skeleton extraction device provided in another embodiment of the present application.

[0038] Fig.24 Shown is a schematic structural diagram of a blood vessel skeleton extraction device provided in another embodiment of the present application.

[0039] Fig.25 Shown is a schematic structural diagram of a blood vessel skeleton extraction device provided in another embodiment of the present application.

[0040] Fig.26 Shown is a schematic structural diagram of a blood vessel skeleton extraction device provided in another embodiment of the present application.

[0041] Fig. 27 Shown is a schematic structural diagram of a blood vessel skeleton extraction device provided in another embodiment of the present application.

[0042] Fig.28 Shown is a schematic structural diagram of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0044] Example scenarios

[0045] Figure 1 The figure is a schematic diagram of an application scenario of a blood vessel skeleton extraction method provided in an embodiment of the present application. Figure 1The scene shown includes a server 110 and an image acquisition device 120 that is communicatively connected to the server 110. Specifically, the server 110 is used to perform distance transformation on the vascular image sequence to be processed, generate a thermal map sequence corresponding to the vascular image sequence to be processed, the thermal map sequence includes a vascular image and M distance value sets annotated on the vascular image, each distance value set includes N distance values ​​of equal size, and M and N are both positive integers; skeleton extraction is performed based on the thermal map sequence corresponding to the vascular image sequence to be processed, and a skeleton extraction result corresponding to the vascular image sequence to be processed is obtained. The image acquisition device 120 is used to acquire the vascular image sequence to be processed and send the vascular image sequence to be processed to the server 110 so that the server 110 performs the above operations.

[0046] Exemplary Methods

[0047] Figure 2 FIG. 2 is a flow chart of a blood vessel skeleton extraction method provided in an embodiment of the present application. Figure 2 As shown, the blood vessel skeleton extraction method includes the following steps.

[0048] Step 210 , performing distance transformation on the blood vessel image sequence to be processed to generate a heat map sequence corresponding to the blood vessel image sequence to be processed.

[0049] Specifically, the heat map sequence includes a blood vessel image and M distance value sets marked on the blood vessel image, each distance value set includes N distance values ​​of equal size, and M and N are both positive integers.

[0050] For example, Figure 3 Shown is a sequence of blood vessel images to be processed provided by an embodiment of the present application. Figure 4 Shown Figure 3 A frame of image in the heat map sequence corresponding to the blood vessel image sequence to be processed is shown. Each distance value set can be represented in the heat map sequence using a color. Each distance value can be the distance between the pixel forming the blood vessel image and the background image.

[0051] Step 220 , performing skeleton extraction based on the heat map sequence corresponding to the blood vessel image sequence to be processed, and obtaining a skeleton extraction result corresponding to the blood vessel image sequence to be processed.

[0052] For example, Figure 5 Shown are the skeleton extraction results obtained using the thinning scheme. Figure 6 The figure shows the skeleton extraction result provided by an embodiment of the present application. Figure 5 As shown in the figure, the area enclosed by the dotted circle has the problem of false connection. Figure 6 As shown, the area encircled by the dotted circle reduces false connections.

[0053] The blood vessel skeleton extraction method provided in the embodiment of the present application performs distance transformation on the blood vessel image sequence to be processed, generates a heat map sequence corresponding to the blood vessel image sequence to be processed, and then performs skeleton extraction based on the heat map sequence corresponding to the blood vessel image sequence to be processed to obtain a skeleton extraction result corresponding to the blood vessel image sequence to be processed, so that the skeleton extraction can be performed with reference to the M distance value sets in the heat map sequence to realize skeleton extraction layer by layer, so as to reduce the false connections generated by the skeleton extraction of each layer, thereby reducing the false connections in the skeleton extraction result corresponding to the blood vessel image sequence to be processed.

[0054] Figure 7 FIG. 2 is a flow chart of a blood vessel skeleton extraction method provided by another embodiment of the present application. Figure 2 Based on the embodiment shown, Figure 7 The embodiment shown is described below in detail. Figure 7 The embodiment shown and Figure 2 The differences and similarities of the illustrated embodiments are not described in detail.

[0055] like Figure 7 As shown, in an embodiment of the present application, the step of performing skeleton extraction based on the heat map sequence corresponding to the blood vessel image sequence to be processed to obtain the skeleton extraction result corresponding to the blood vessel image sequence to be processed includes the following steps.

[0056] Step 710 : for P pixels corresponding to each distance value set in the M distance value sets, determine Q skeleton pixels in the P pixels based on a skeletonization algorithm.

[0057] Specifically, P is a positive integer, and Q is a positive integer less than or equal to P. In the thermal map sequence, each pixel forming the blood vessel image corresponds to a distance value. The skeleton pixel is a pixel corresponding to the image region obtained by skeletonizing the image region composed of P pixels using a skeletonization algorithm.

[0058] For example, Figure 8 The figure shows a frame of image in a thermal image sequence provided by an embodiment of the present application. Figure 8 As shown, the frame image includes three colors: black, dark gray, and light gray. Each color represents a distance value set. For example, the distance value in the distance value set represented by black can be 1mm. The distance value in the distance value set represented by dark gray can be 2mm. The distance value in the distance value set represented by light gray can be 3mm. Figure 8 It can be seen that there is adhesion between the two blood vessels.

[0059] For example, Fig. 9 The skeletonization algorithm is shown in Figure 8The image is obtained by skeletonizing the black area in the image shown. In practical applications, it can be a skeletonized three-dimensional image obtained by skeletonizing the heat map sequence. For the convenience of visualization, only Figure 8 The image frame and Fig. 9 A frame of image shown. Fig. 9 As shown in the figure, two blood vessels are connected by a skeletonized line segment. However, in fact, the two blood vessels are not connected, but are just stuck together. Figure 8 The skeletonized image of the black areas in the image shown has a false connection problem.

[0060] Step 720, determining false connected pixels among the Q skeleton pixels.

[0061] Specifically, the pseudo-connected pixels are pixels corresponding to the vascular adhesion area in the vascular image. Fig. 9 As shown, the pixels corresponding to the line segment between the two blood vessels are false connection pixels.

[0062] Step 730 , removing false connected pixels from the Q skeleton pixels, and obtaining a skeleton extraction result corresponding to the blood vessel image sequence to be processed.

[0063] For example, Fig.10 Shown is removal Fig. 9 The image obtained after the pseudo-connected pixels in the embodiment shown. Fig.10 As shown, continue to Fig.10 The dark grey area in the image shown in the figure performs steps 710 to 730 to obtain the skeleton extraction result corresponding to the blood vessel image sequence to be processed.

[0064] For example, Fig.11 The figure shows the skeleton extraction result corresponding to the blood vessel image sequence to be processed provided by an embodiment of the present application. Specifically, Fig.11 Shown is the pair Figure 8 The image shown is an image obtained by looping through steps 710 to 730. Fig.11 As shown, there is no false connection between the two blood vessel segments. Fig.12 Another skeleton extraction result obtained using the thinning scheme is shown. Specifically, Fig.12 Shown is the pair Figure 8 The image shown is obtained using the thinning scheme. Fig.12 As shown, the two blood vessels are connected by a line segment, that is, there is a false connection between the two blood vessels.

[0065] By removing false connection pixels, the false connection problem in the skeleton extraction result corresponding to the finally obtained blood vessel image sequence to be processed is reduced, and the accuracy of the blood vessel skeleton extraction method is further improved.

[0066] Fig.13 FIG. 2 is a flow chart of a blood vessel skeleton extraction method provided by another embodiment of the present application. Figure 7 Based on the embodiment shown, Fig.13 The embodiment shown is described below in detail. Fig.13 The embodiment shown and Figure 7 The differences and similarities of the illustrated embodiments are not described in detail.

[0067] like Fig.13 As shown, in the embodiment of the present application, the step of determining the pseudo-connected pixels among the Q skeleton pixels includes the following steps.

[0068] Step 1310, determining D endpoint pixels among the Q skeleton pixels.

[0069] Specifically, D is a positive integer less than or equal to Q.

[0070] Specifically, Fig.14 Endpoint pixels are shown. Fig.15 Shown are line segment point pixels. Fig.16 The figure shows node pixels. Pixels can be divided into three categories: endpoint pixels, line segment pixels, and node pixels. Fig.14 As shown in , within a 3*3*3 cubic pixel, if there are two connected pixels, the pixel at the end of the two pixels is the endpoint pixel. Fig.15 As shown in , within a 3*3*3 cubic pixel, if there are three connected pixels, the middle pixel among the three pixels is a line segment point pixel. Fig.16 As shown, within a 3*3*3 cubic pixel, if the number of connected pixels is greater than 3, the central pixel among the greater than 3 pixels is a node pixel.

[0071] Specifically, Fig. 9 As shown, there are 4 endpoint pixels, that is, D is 4.

[0072] Step 1320 , for each skeleton pixel among the Q skeleton pixels, determine whether the current skeleton pixel is connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels.

[0073] Specifically, Q1 is a positive integer less than Q.

[0074] Step 1330: If the current skeleton pixel is not connected to any of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels, the current skeleton pixel is determined to be a false connection pixel.

[0075] Specifically, if the current skeleton pixel is connected to any one or more of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels, the current skeleton pixel is not a pseudo-connected pixel.

[0076] For example, Fig. 9 As shown in FIG. 1 , the four line segments at the ends of the two blood vessels are all connected to the endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels. Therefore, the four line segments at the ends of the two blood vessels are retained and not deleted. The line segment connecting the two blood vessels cannot be connected to any of the four endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels. Therefore, the skeleton pixels forming the line segment connecting the two blood vessels are all determined as false connection pixels.

[0077] Fig.17 FIG. 2 is a flow chart of a blood vessel skeleton extraction method provided by another embodiment of the present application. Fig.13 Based on the embodiment shown, Fig.17 The embodiment shown is described below in detail. Fig.17 The embodiment shown and Fig.13 The differences and similarities of the illustrated embodiments are not described in detail.

[0078] like Fig.17 As shown, in the embodiment of the present application, the step of determining D endpoint pixels among Q skeleton pixels includes the following steps.

[0079] Step 1710: Determine a first determination pixel set based on the Q skeleton pixels and pixels corresponding to all distance value sets except the current distance value set.

[0080] Specifically, the first judgment pixel set includes Q skeleton pixels and pixels corresponding to all distance value sets except the current distance value set.

[0081] For example, Fig. 9 As shown, Q skeleton pixels are pixels forming 5 black line segments. Pixels corresponding to all distance value sets except the current distance value set are pixels forming dark gray areas and light gray areas. The first judgment pixel set includes Q skeleton pixels and pixels forming dark gray areas and light gray areas.

[0082] Step 1720: Determine D endpoint pixels based on the first judgment pixel set and the preset endpoint judgment condition.

[0083] Specifically, the preset endpoint judgment condition may be whether the Q skeleton pixels are endpoint pixels within the area formed by the first judgment pixel set.

[0084] For example, Fig. 9As shown in FIG. 1 , the area formed by the first judgment pixel set is 5 black line segments, a dark gray area, and a light gray area. Fig. 9 In the example, based on the first judgment pixel set and the preset endpoint judgment condition, the D endpoint pixels determined are the four endpoint pixels at the ends of the two blood vessels.

[0085] Fig.18 FIG. 2 is a flow chart of a blood vessel skeleton extraction method provided by another embodiment of the present application. Fig.13 Based on the embodiment shown, Fig.18 The embodiment shown is described below in detail. Fig.18 The embodiment shown and Fig.13 The differences and similarities of the illustrated embodiments are not described in detail.

[0086] like Fig.18 As shown, in the embodiment of the present application, the step of determining whether the current skeleton pixel is connected to any one of the D endpoint pixels through the Q1 skeleton pixels among the Q skeleton pixels includes the following steps.

[0087] Step 1810: determine a second judgment pixel set based on Q skeleton pixels.

[0088] Specifically, the second judgment pixel set includes Q skeleton pixels.

[0089] For example, Fig. 9 As shown, the Q skeleton pixels are pixels forming five black line segments. The second determination pixel set includes pixels forming five black line segments.

[0090] Step 1820 , based on the second judgment pixel set and the preset connectivity judgment condition, determine whether the current skeleton pixel is connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels.

[0091] Specifically, the preset connection judgment condition may be that the current skeleton pixel is connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels.

[0092] For example, Fig. 9 As shown in FIG. 1 , the area formed by the second judgment pixel set is 5 black line segments. Fig. 9 In the figure, the D endpoint pixels are the four endpoint pixels at the ends of the two blood vessels. The points on the four line segments at the ends of the two blood vessels are all connected to the endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels. The pixel points forming the line segment connecting the two blood vessels cannot be connected to the endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels.

[0093] Fig.19FIG. 2 is a flow chart of a blood vessel skeleton extraction method provided by another embodiment of the present application. Figure 2 Based on the embodiment shown, Fig.19 The embodiment shown is described below in detail. Fig.19 The embodiment shown and Figure 2 The differences and similarities of the illustrated embodiments are not described in detail.

[0094] like Fig.19 As shown, in the embodiment of the present application, the step of performing distance transformation on the blood vessel image sequence to be processed to generate a thermal map sequence corresponding to the blood vessel image sequence to be processed includes the following steps.

[0095] Step 1910 , using the classification model, performing distance transformation on the blood vessel image sequence to be processed, and generating a heat map sequence corresponding to the blood vessel image sequence to be processed.

[0096] The classification model is used to perform distance transformation on the vascular image sequence to be processed, and a heat map sequence corresponding to the vascular image sequence to be processed is generated, which is convenient, fast and efficient.

[0097] Fig. 20 FIG. 2 is a flow chart of a blood vessel skeleton extraction method provided by another embodiment of the present application. Fig.19 Based on the embodiment shown, Fig. 20 The embodiment shown is described below in detail. Fig. 20 The embodiment shown and Fig.19 The differences and similarities of the illustrated embodiments are not described in detail.

[0098] like Fig. 20 As shown, in the embodiment of the present application, the step of using the classification model to perform distance transformation on the blood vessel image sequence to be processed and generating a thermal map sequence corresponding to the blood vessel image sequence to be processed includes the following steps.

[0099] Step 2010 , the blood vessel image sequence to be processed is cut into blocks to obtain a plurality of image blocks to be processed, wherein the resolution of the image blocks to be processed is the same as the physical resolution corresponding to the blood vessel image sequence to be processed.

[0100] In practical applications, the resolution of the image block to be processed is the same as the physical resolution corresponding to the blood vessel image sequence to be processed, so that the skeleton extraction results obtained by performing skeleton extraction on the heat map sequence corresponding to the image block to be processed can be merged together.

[0101] Step 2020: Use the classification model to perform distance transformation on the multiple image blocks to be processed, and generate heat map sequences corresponding to the multiple image blocks to be processed.

[0102] The classification model is used to perform distance transformation on multiple image blocks to be processed, and a heat map sequence corresponding to each of the multiple image blocks to be processed is generated, thereby reducing the requirement for video memory.

[0103] Step 2030 : generating a heat map sequence corresponding to the blood vessel image sequence to be processed based on the heat map sequences corresponding to the plurality of image blocks to be processed.

[0104] Specifically, the heat map sequences corresponding to the plurality of image blocks to be processed are merged together to generate a heat map sequence corresponding to the blood vessel image sequence to be processed.

[0105] Combination of the above Figures 1 to 20 , describes the method embodiment of the present application in detail, and the following is combined with Figure 21 to Figure 27 , describes the device embodiment of the present application in detail. It should be understood that the description of the method embodiment corresponds to the description of the device embodiment, so the parts not described in detail can refer to the previous method embodiment.

[0106] Exemplary Devices

[0107] Fig.21 FIG. 1 is a schematic diagram of the structure of a blood vessel skeleton extraction device provided in an embodiment of the present application. Fig.21 As shown, the blood vessel skeleton extraction device 2100 of the embodiment of the present application includes a heat map generation module 2110 and a skeleton extraction module 2120 .

[0108] Specifically, the heat map generation module 2110 is configured to perform distance transformation on the vascular image sequence to be processed, and generate a heat map sequence corresponding to the vascular image sequence to be processed, wherein the heat map sequence includes a vascular image and M distance value sets annotated on the vascular image, each distance value set includes N distance values ​​of equal size, and M and N are both positive integers. The skeleton extraction module 2120 is configured to perform skeleton extraction based on the heat map sequence corresponding to the vascular image sequence to be processed, and obtain a skeleton extraction result corresponding to the vascular image sequence to be processed.

[0109] Fig. 22 FIG. 1 is a schematic diagram of the structure of a blood vessel skeleton extraction device provided by another embodiment of the present application. Fig.21 Based on the embodiment shown, Fig. 22 The embodiment shown is described below in detail. Fig. 22 The embodiment shown and Fig.21 The differences and similarities of the illustrated embodiments are not described in detail.

[0110] like Fig. 22 As shown, the skeleton extraction module 2120 of the embodiment of the present application includes: a hierarchical skeletonization unit 2121, a falsely connected pixel determination unit 2122 and a falsely connected pixel removal unit 2123.

[0111] Specifically, in the heat map sequence, each pixel forming the vascular image corresponds to a distance value. The hierarchical skeletonization unit 2121 is configured to determine Q skeleton pixels among the P pixels corresponding to each distance value set in the M distance value sets based on the skeletonization algorithm, where P is a positive integer and Q is a positive integer less than or equal to P. The false connection pixel determination unit 2122 is configured to determine the false connection pixels among the Q skeleton pixels, where the false connection pixels are pixels corresponding to the vascular adhesion area in the vascular image. The false connection pixel removal unit 2123 is configured to remove the false connection pixels among the Q skeleton pixels to obtain the skeleton extraction result corresponding to the vascular image sequence to be processed.

[0112] Fig.23 FIG. 1 is a schematic diagram of the structure of a blood vessel skeleton extraction device provided by another embodiment of the present application. Fig. 22 Based on the embodiment shown, Fig.23 The embodiment shown is described below in detail. Fig.23 The embodiment shown and Fig. 22 The differences and similarities of the illustrated embodiments are not described in detail.

[0113] like Fig.23 As shown, the false connection pixel determination unit 2122 of the embodiment of the present application includes: an endpoint pixel determination subunit 2310, a connectivity judgment subunit 2320 and a false connection pixel determination subunit 2330.

[0114] Specifically, the endpoint pixel determination subunit 2310 is configured to determine D endpoint pixels among the Q skeleton pixels, where D is a positive integer less than or equal to Q. The connectivity determination subunit 2320 is configured to determine, for each skeleton pixel among the Q skeleton pixels, whether the current skeleton pixel is connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels, where Q1 is a positive integer less than Q. The pseudo-connected pixel determination subunit 2330 is configured to determine the current skeleton pixel as a pseudo-connected pixel if the current skeleton pixel is not connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels.

[0115] Fig.24 FIG. 1 is a schematic diagram of the structure of a blood vessel skeleton extraction device provided by another embodiment of the present application. Fig.23 Based on the embodiment shown, Fig.24 The embodiment shown is described below in detail. Fig.24 The embodiment shown and Fig.23 The differences and similarities of the illustrated embodiments are not described in detail.

[0116] like Fig.24As shown, the endpoint pixel determination subunit 2310 of the embodiment of the present application includes: a first judgment pixel set determination subunit 2311 and D endpoint pixel determination subunits 2312.

[0117] Specifically, the first judgment pixel set determination subunit 2311 is configured to determine the first judgment pixel set based on the Q skeleton pixels and the pixels corresponding to all distance value sets except the current distance value set. The D endpoint pixel determination subunit 2312 is configured to determine the D endpoint pixels based on the first judgment pixel set and the preset endpoint judgment condition.

[0118] Fig.25 FIG. 1 is a schematic diagram of the structure of a blood vessel skeleton extraction device provided by another embodiment of the present application. Fig.23 Based on the embodiment shown, Fig.25 The embodiment shown is described below in detail. Fig.25 The embodiment shown and Fig.23 The differences and similarities of the illustrated embodiments are not described in detail.

[0119] like Fig.25 As shown, the connectivity judgment subunit 2320 of the embodiment of the present application includes: a second judgment pixel set determination subunit 2321 and a judgment subunit 2322 .

[0120] Specifically, the second judgment pixel set determination subunit 2321 is configured to determine the second judgment pixel set based on the Q skeleton pixels. The judgment subunit 2322 is configured to judge whether the current skeleton pixel is connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels based on the second judgment pixel set and the preset connectivity judgment condition.

[0121] Fig.26 FIG. 1 is a schematic diagram of the structure of a blood vessel skeleton extraction device provided by another embodiment of the present application. Fig.21 Based on the embodiment shown, Fig.26 The embodiment shown is described below in detail. Fig.26 The embodiment shown and Fig.21 The differences and similarities of the illustrated embodiments are not described in detail.

[0122] like Fig.26 As shown, the heat map generation module 2110 of the embodiment of the present application includes: a distance transformation unit 2111.

[0123] Specifically, the distance transformation unit 2111 is configured to perform distance transformation on the blood vessel image sequence to be processed using the classification model, and generate a heat map sequence corresponding to the blood vessel image sequence to be processed.

[0124] Fig. 27FIG. 1 is a schematic diagram of the structure of a blood vessel skeleton extraction device provided by another embodiment of the present application. Fig.26 Based on the embodiment shown, Fig. 27 The embodiment shown is described below in detail. Fig. 27 The embodiment shown and Fig.26 The differences and similarities of the illustrated embodiments are not described in detail.

[0125] like Fig. 27 As shown, the distance transformation unit 2111 of the embodiment of the present application includes: a slicing subunit 2710 , a distance transformation subunit 2720 and a heat map sequence determination subunit 2730 .

[0126] Specifically, the slicing subunit 2710 is configured to perform slicing processing on the vascular image sequence to be processed to obtain multiple image blocks to be processed, and the resolution of the image blocks to be processed is the same as the physical resolution corresponding to the vascular image sequence to be processed. The distance transformation subunit 2720 is configured to perform distance transformation on the multiple image blocks to be processed using a classification model to generate a thermal map sequence corresponding to each of the multiple image blocks to be processed. The thermal map sequence determination subunit 2730 is configured to generate a thermal map sequence corresponding to the vascular image sequence to be processed based on the thermal map sequences corresponding to each of the multiple image blocks to be processed.

[0127] Exemplary Electronic Devices

[0128] Fig.28 FIG. 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Fig.28 As shown, the electronic device 2800 includes: one or more processors 2801 and a memory 2802; and computer program instructions stored in the memory 2802, and when the computer program instructions are executed by the processor 2801, the processor 2801 executes a blood vessel skeleton extraction method as in any of the above embodiments.

[0129] The processor 2801 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0130] The memory 2802 may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 2801 may execute the program instructions to implement the steps in the vascular skeleton extraction method of each embodiment of the present application described above and / or other desired functions.

[0131] In one example, the electronic device 2800 may further include: an input device 2803 and an output device 2804, which are connected via a bus system and / or other forms of connection mechanisms ( Fig.28 ) interconnected.

[0132] In addition, the input device 2803 may also include, for example, a keyboard, a mouse, a microphone, etc.

[0133] The output device 2804 can output various information to the outside, and can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto.

[0134] Of course, to simplify, Fig.28 Only some of the components related to the present application in the electronic device 2800 are shown, and components such as a bus, an input device / output interface, etc. are omitted. In addition, according to specific application conditions, the electronic device 2800 may also include any other appropriate components.

[0135] Exemplary computer-readable storage media

[0136] In addition to the above methods and devices, the embodiments of the present application may also be a computer program product, including computer program instructions, which, when executed by a processor, enable the processor to execute the steps in the blood vessel skeleton extraction method of any of the above embodiments.

[0137] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0138] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of the vascular skeleton extraction method according to various embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0139] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0140] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.

[0141] The block diagrams of the devices, apparatuses, equipment, and systems involved in this application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0142] It should also be noted that in the apparatus, device and method of the present application, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0143] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0144] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

[0145] The above are only preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for extracting a blood vessel skeleton, characterized in that: include: Performing distance transformation on the vascular image sequence to be processed to generate a thermal map sequence corresponding to the vascular image sequence to be processed, wherein the thermal map sequence includes a vascular image and M distance value sets annotated on the vascular image, each of the distance value sets includes N distance values ​​of equal size, and M and N are both positive integers; Perform skeleton extraction based on the heat map sequence corresponding to the blood vessel image sequence to be processed, and obtain the skeleton extraction result corresponding to the blood vessel image sequence to be processed, Wherein, in the heat map sequence, each pixel forming a blood vessel image corresponds to a distance value, and the skeleton extraction is performed based on the heat map sequence corresponding to the blood vessel image sequence to be processed to obtain the skeleton extraction result corresponding to the blood vessel image sequence to be processed, including: For P pixels corresponding to each distance value set in the M distance value sets, Based on a skeletonization algorithm, determine Q skeleton pixels among the P pixels, where P is a positive integer and Q is a positive integer less than or equal to P; Determine a false connection pixel among the Q skeleton pixels, where the false connection pixel is a pixel corresponding to a vascular adhesion region in the vascular image; Falsely connected pixels among the Q skeleton pixels are removed to obtain a skeleton extraction result corresponding to the blood vessel image sequence to be processed.

2. The blood vessel skeleton extraction method according to claim 1, characterized in that: The determining of the pseudo-connected pixels among the Q skeleton pixels comprises: Determine D endpoint pixels among the Q skeleton pixels, where D is a positive integer less than or equal to Q; For each skeleton pixel in the Q skeleton pixels, Determine whether the current skeleton pixel is connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels, where Q1 is a positive integer less than Q; If the current skeleton pixel is not connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels, the current skeleton pixel is determined as the pseudo-connected pixel.

3. The blood vessel skeleton extraction method according to claim 2, characterized in that: The determining of D endpoint pixels among the Q skeleton pixels comprises: Determine a first judgment pixel set based on the Q skeleton pixels and pixels corresponding to all distance value sets except the current distance value set; The D endpoint pixels are determined based on the first judgment pixel set and a preset endpoint judgment condition.

4. The blood vessel skeleton extraction method according to claim 2, characterized in that: The determining whether the current skeleton pixel is connected to any one of the D endpoint pixels through Q1 skeleton pixels among the Q skeleton pixels comprises: Based on the Q skeleton pixels, determining a second judgment pixel set; Based on the second judgment pixel set and the preset connectivity judgment condition, it is judged whether the current skeleton pixel is connected with any one of the D endpoint pixels through Q1 skeleton pixels of the Q skeleton pixels.

5. The blood vessel skeleton extraction method according to claim 1, characterized in that: Performing distance transformation on the to-be-processed blood vessel image sequence to generate a thermal map sequence corresponding to the to-be-processed blood vessel image sequence includes: Performing block processing on the to-be-processed blood vessel image sequence to obtain a plurality of to-be-processed image blocks, wherein the resolution of the to-be-processed image blocks is the same as the physical resolution corresponding to the to-be-processed blood vessel image sequence; Performing distance transformation on the multiple image blocks to be processed to generate heat map sequences corresponding to the multiple image blocks to be processed; Based on the heat map sequences corresponding to the multiple image blocks to be processed, a heat map sequence corresponding to the blood vessel image sequence to be processed is generated.

6. A blood vessel skeleton extraction device, characterized in that: include: A heat map generation module is configured to perform distance transformation on a to-be-processed vascular image sequence to generate a heat map sequence corresponding to the to-be-processed vascular image sequence, wherein the heat map sequence includes a vascular image and M distance value sets annotated on the vascular image, each of the distance value sets includes N distance values ​​of equal size, and both M and N are positive integers; a skeleton extraction module configured to perform skeleton extraction based on the thermal map sequence corresponding to the to-be-processed blood vessel image sequence, and obtain a skeleton extraction result corresponding to the to-be-processed blood vessel image sequence; in, In the heat map sequence, each pixel forming the blood vessel image corresponds to a distance value, and the skeleton extraction module is further configured as follows: For P pixels corresponding to each distance value set in the M distance value sets, Based on a skeletonization algorithm, determine Q skeleton pixels among the P pixels, where P is a positive integer and Q is a positive integer less than or equal to P; Determine a false connection pixel among the Q skeleton pixels, where the false connection pixel is a pixel corresponding to a vascular adhesion region in the vascular image; Falsely connected pixels among the Q skeleton pixels are removed to obtain a skeleton extraction result corresponding to the blood vessel image sequence to be processed.

7. A computer-readable storage medium, characterized in that: The storage medium stores instructions, and when the instructions are executed by a processor of an electronic device, the electronic device is enabled to execute the blood vessel skeleton extraction method according to any one of claims 1 to 5.

8. An electronic device, characterized in that: The electronic device comprises: processor; memory for storing computer executable instructions; The processor is used to execute the computer executable instructions to implement the blood vessel skeleton extraction method according to any one of claims 1 to 5.

Images