A medical image segmentation method, device and apparatus

Abstract

The application relates to the technical field of medical imaging, in particular to a medical image segmentation method, device and apparatus, which provide a scheme for quickly segmenting the edge of a target organ in a medical image. Embodiments of the application determine layer interval information between a medical image to be segmented and a reference medical image; determine target segmentation points in the medical image to be segmented corresponding to the respective reference segmentation points in the reference medical image according to the reference segmentation points selected in the reference medical image and the layer interval information between the medical image to be segmented and the reference medical image; determine the edge of a target organ between any two adjacent target segmentation points in the medical image to be segmented according to the pixel values of the two adjacent target segmentation points and a predefined cost function; and segment the target organ from the medical image to be segmented according to the determined edge of the target organ between each two adjacent target segmentation points. The application can greatly improve the segmentation efficiency of medical images.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of medical imaging technology, in particular to a medical image segmentation method, device and apparatus. BACKGROUND

[0002] With the successful application of medical image segmentation technology in clinical medicine, image segmentation technology plays an increasingly important role in medical image processing and analysis, and the segmented images are widely used in various occasions.

[0003] The current common method for organ segmentation of medical images is that, for an obtained image sequence containing multiple medical images, a user manually outlines the edges of organs in each medical image, and then performs segmentation according to the organ edges outlined by the user. However, the current medical image segmentation method is low in efficiency. SUMMARY

[0004] The purpose of the present application is to provide a medical image segmentation method, device and apparatus, to provide a scheme for quickly segmenting the edges of target organs in medical images.

[0005] In a first aspect, the present application provides a medical image segmentation method, which comprises:

[0006] determining layer interval information between a to-be-segmented medical image and a reference medical image; wherein the to-be-segmented medical image and the reference medical image are images of different layers in a medical image sequence obtained by scanning a target organ;

[0007] determining target segmentation points in the to-be-segmented medical image corresponding to each reference segmentation point according to a plurality of reference segmentation points selected in the reference medical image and the layer interval information between the to-be-segmented medical image and the reference medical image;

[0008] for any two adjacent target segmentation points in the to-be-segmented medical image, determining the edge of the target organ between the two adjacent target segmentation points according to the pixel values of the two adjacent target segmentation points and a predefined cost function;

[0009] segmenting the target organ from the to-be-segmented image according to the determined edge of the target organ between each two adjacent target segmentation points.

[0010] In a second aspect, the present application provides a medical image segmentation device, which comprises at least one processor and at least one memory;

[0011] wherein the memory stores program code which, when executed by the processor, implements the medical image segmentation method of the first aspect described above.

[0012] In a third aspect, the present application provides a medical image segmentation device, comprising:

[0013] An acquisition module is configured to determine layer interval information between a to-be-segmented medical image and a reference medical image, wherein the to-be-segmented medical image and the reference medical image are images of different layers in a medical image sequence obtained by scanning a target organ;

[0014] A determination module is configured to determine, according to a plurality of reference segmentation points selected in the reference medical image and the layer interval information between the to-be-segmented medical image and the reference medical image, target segmentation points in the to-be-segmented medical image corresponding to the reference segmentation points respectively;

[0015] An edge detection module is configured to, for any two adjacent target segmentation points in the to-be-segmented medical image, determine an edge of the target organ between the two adjacent target segmentation points according to pixel values of the two adjacent target segmentation points and a predefined cost function;

[0016] A segmentation module is configured to segment the target organ from the to-be-segmented medical image according to the edge of the target organ between each two adjacent target segmentation points.

[0017] In a fourth aspect, the present application provides a computer readable storage medium, when instructions in the computer readable storage medium are executed by an electronic device, the electronic device can execute the medical image segmentation method in the first aspect.

[0018] In a fourth aspect, the present application provides a computer program product, comprising a computer program:

[0019] When the computer program is executed by a processor, the medical image segmentation method in the first aspect is implemented.

[0020] The technical scheme provided by the embodiments of the present application at least brings the following beneficial effects:

[0021] Since the embodiments of the present application select a reference medical image from the medical image sequence, and select a plurality of reference segmentation points from the reference medical image; for other medical images in the medical image sequence except the reference medical image, the target segmentation points in the medical image to be segmented corresponding to each reference segmentation point can be automatically determined according to the plurality of reference segmentation points selected from the reference medical image and the interlayer spacing information between the medical image to be segmented and the reference medical image, without the need for the user to manually select the segmentation points for each medical image, which can greatly improve the efficiency of determining the segmentation points in the medical image; and in the embodiments of the present application, for each medical image, after the target segmentation points in the medical image are determined, the edge of the target organ can be automatically generated according to the pixel value of the target segmentation point and the cost function. Compared with the scheme in the related art in which the user manually outlines the organ edge in the medical image, the scheme of automatically generating the edge of the target organ according to a plurality of target segmentation points can more quickly segment the target organ from the medical image, thereby improving the efficiency of medical image segmentation. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly introduced as follows. Obviously, the drawings introduced below are only some embodiments of the present application, and other drawings can also be obtained according to these drawings without creative labor for those skilled in the art.

[0023] Figure 1 An application scenario diagram of an optional image segmentation method of the embodiments of the present application;

[0024] Figure 2 An application scenario diagram of an optional image segmentation method of the embodiments of the present application;

[0025] Figure 3 A flowchart of an image segmentation method of the embodiments of the present application;

[0026] Figure 4 A display interface diagram of determining reference segmentation points of the embodiments of the present application;

[0027] Figure 5 A flowchart of determining the edge of the target organ between two reference segmentation points of the embodiments of the present application;

[0028] Figure 6 A schematic diagram of generating a directed weighted graph of the embodiments of the present application;

[0029] Figure 7 A schematic diagram of a candidate path between two adjacent reference segmentation points of the embodiments of the present application;

[0030] Figure 8A schematic diagram of determining the edge of a target organ between two adjacent reference segmentation points according to an embodiment of the present application;

[0031] Figure 9 A flowchart of a process of determining a target segmentation point in a medical image to be segmented according to an embodiment of the present application;

[0032] Figure 10 A flowchart of determining a gradient direction corresponding to a reference segmentation point according to an embodiment of the present application;

[0033] Figure 11 A schematic diagram of a gradient direction corresponding to a reference segmentation point according to an embodiment of the present application;

[0034] Figure 12A A schematic diagram of a pixel in any medical image according to an embodiment of the present application;

[0035] Figure 12B A schematic diagram of the interval between any two adjacent medical images according to an embodiment of the present application;

[0036] Figure 13 A schematic diagram of determining a horizontal coordinate offset value and a vertical coordinate offset value according to an embodiment of the present application;

[0037] Figure 14 A schematic diagram of the position of a reference segmentation point according to an embodiment of the present application;

[0038] Figure 15 A three-dimensional schematic diagram of determining a target segmentation point according to a reference segmentation point according to an embodiment of the present application;

[0039] Figure 16 A schematic diagram of determining a plane offset value and a depth offset value according to an embodiment of the present application;

[0040] Figure 17 A flowchart of verifying the accuracy of a target segmentation point according to an embodiment of the present application;

[0041] Figure 18 An effect diagram of the edge segmentation of a target organ in a medical image according to an embodiment of the present application;

[0042] Figure 19 A structural schematic diagram of a medical image segmentation device according to an embodiment of the present application;

[0043] Figure 20 A structural schematic diagram of a medical image segmentation device according to an embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. The described embodiments are some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort are within the protection scope of the present application.

[0045] In addition, in the description of the embodiments of the present application, unless otherwise specified, " / " represents the meaning of or, for example, A / B can represent A or B; "and / or" in the text only describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent: A exists alone, A and B exist together, and B exists alone, in addition, in the description of the embodiments of the present application, "multiple" means two or more than two.

[0046] Hereinafter, the terms "first" and "second" are only used for description purposes, and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more features, in the description of the embodiments of the present application, unless otherwise specified, the meaning of "multiple" is two or more than two.

[0047] In order to facilitate understanding of the interactive medical sequence image segmentation method and device provided by the embodiments of the present application, the following explains and describes some terms in the embodiments of the present application, so as to facilitate the understanding of the person skilled in the art.

[0048] (1) Computed Tomography (CT): CT is to use X-ray beam to scan the certain thickness of the layer of human body, the X-ray through the layer is received by the detector, and then is converted into visible light, and then is converted into electric signal by photoelectric conversion, and then is converted into digital by analog / digital converter, and then is input into computer for processing. The processing of image formation is to divide the selected layer into several cuboids with the same volume, which is called voxel. The information obtained by scanning is calculated to obtain the X-ray attenuation coefficient or absorption coefficient of each voxel, and then is arranged into matrix, i.e. digital matrix, which can be stored in disk or optical disk. Each number in the digital matrix is converted into a small block with different gray scales from black to white, i.e. pixel, by digital / analog converter, and then is arranged in matrix, i.e. CT image. Therefore, the CT image is a reconstructed image. The X-ray absorption coefficient of each voxel can be calculated by different mathematical methods. The working procedure of CT is as follows: according to the different absorption and transmission rates of different tissues of human body to X-ray, the instrument with extremely high sensitivity is used to measure the human body, and then the data obtained by measurement is input into the computer, and then the computer processes the data to obtain the sectional or three-dimensional image of the part of human body to be examined, and to find the small lesions in any part of the body.

[0049] (2) Pixel: Pixel refers to small squares composed of an image, and each small block has a definite position and assigned color value. The color and position of the small square determine the appearance of the image. Pixel can be regarded as an indivisible unit or element in the entire image. Indivisible means that it cannot be cut into smaller units or elements, and it exists in the form of a small square with a single color.

[0050] (3) Cost function / loss function: Cost function refers to a function that maps the values of random events or related random variables to non-negative real numbers to represent the "risk" or "loss" of the random event. The smaller the cost function value is, the smaller the error is, and the closer the model and the parameters are to the actual situation.

[0051] The application scenario of the optional image segmentation method provided by the embodiments of the application will be introduced below with reference to the drawings. As shown in FIG. 1, a medical image device 10 includes an image acquisition unit 101 and an image processing unit 102; wherein: Figure 1

[0052] The image acquisition unit 101 is configured to acquire a medical image sequence; the medical image sequence includes a plurality of medical images; ​

[0053] The image processing unit 102 is configured to determine layer interval information between a to-be-segmented medical image and a reference medical image, wherein the to-be-segmented medical image and the reference medical image are images of different layers in a medical image sequence obtained by scanning a target organ; determine target segmentation points in the to-be-segmented medical image corresponding to a plurality of reference segmentation points selected in the reference medical image according to the plurality of reference segmentation points and the layer interval information between the to-be-segmented medical image and the reference medical image; for any two adjacent target segmentation points in the to-be-segmented medical image, determine an edge of the target organ between the two adjacent target segmentation points according to pixel values of the two adjacent target segmentation points and a predefined cost function; and segment the target organ from the to-be-segmented image according to the determined edge of the target organ between each two adjacent target segmentation points.

[0054] In addition, the embodiments of the present application also provide an optional application scenario of the image segmentation method, as shown in Figure 2 The application scenario includes a medical image acquisition device 20 and a server 11.

[0055] The medical image acquisition device 20 is configured to acquire a medical image sequence including a plurality of medical images, and send the acquired medical image sequence to the server 11.

[0056] The server 11 is configured to receive the medical image sequence sent by the medical image acquisition device 20, determine layer interval information between a to-be-segmented medical image and a reference medical image, wherein the to-be-segmented medical image and the reference medical image are images of different layers in a medical image sequence obtained by scanning a target organ, determine target segmentation points in the to-be-segmented medical image corresponding to a plurality of reference segmentation points selected in the reference medical image according to the plurality of reference segmentation points and the layer interval information between the to-be-segmented medical image and the reference medical image, for any two adjacent target segmentation points in the to-be-segmented medical image, determine an edge of the target organ between the two adjacent target segmentation points according to pixel values of the two adjacent target segmentation points and a predefined cost function, and segment the target organ from the to-be-segmented image according to the determined edge of the target organ between each two adjacent target segmentation points.

[0057] Of course, the method provided by the embodiments of the present application is not limited to the application scenario as shown in Figure 1 or Figure 2 The method provided by the embodiments of the present application can also be used in other possible application scenarios, which are not limited herein.

[0058] As shown in Figure 3 The image segmentation method provided by the embodiments of the present application can specifically include the following steps:

[0059] In step S301, layer interval information between the to-be-segmented medical image and the reference medical image is determined, wherein the to-be-segmented medical image and the reference medical image are images of different layers in a medical image sequence obtained by scanning a target organ;

[0060] In step S302, target segmentation points corresponding to the reference segmentation points respectively in the to-be-segmented medical image are determined according to the reference segmentation points selected in the reference medical image and the layer interval information between the to-be-segmented medical image and the reference medical image.

[0061] In step S303, edges of the target organ between any two adjacent target segmentation points in the to-be-segmented medical image are determined according to pixel values of the two adjacent target segmentation points and a predefined cost function.

[0062] In step S304, the target organ is segmented from the to-be-segmented medical image according to the edges of the target organ between each two adjacent target segmentation points.

[0063] In the embodiment, one reference medical image is selected from the medical image sequence, and a plurality of reference segmentation points are selected from the reference medical image. For other medical images in the medical image sequence except the reference medical image, target segmentation points corresponding to the reference segmentation points respectively in the to-be-segmented medical image can be determined automatically according to the reference segmentation points selected in the reference medical image and the layer interval information between the to-be-segmented medical image and the reference medical image, without the need for the user to manually select segmentation points for each medical image, which can greatly improve the efficiency of determining segmentation points in the medical image. In the embodiment, for each medical image, after the target segmentation points in the medical image are determined, the edges of the target organ can be automatically generated according to the pixel values of the target segmentation points and the cost function. Compared with the scheme in the related art in which the user manually outlines the edges of the organ in the medical image, the scheme of automatically generating the edges of the target organ according to a plurality of target segmentation points can more quickly segment the target organ from the medical image, thereby improving the efficiency of segmenting the medical image.

[0064] Before step S301, the embodiment needs to determine one medical image from the medical image sequence as the reference medical image.

[0065] In the implementation, the reference medical image can be selected from the medical image sequence in various ways.

[0066] For example, in the embodiments of this application, a specific medical image can be selected from a medical image sequence as a reference medical image according to a pre-set rule; for example, the image with the largest edge contour of the target organ in the medical image sequence can be selected as the reference medical image, or the image with the smallest edge contour of the target organ in the medical image sequence can be selected as the reference medical image.

[0067] For example, arbitrarily select a medical image from a medical image sequence as a reference medical image.

[0068] The above methods are merely examples of embodiments of this application. In practical applications, the methods for selecting reference medical images are not limited.

[0069] Based on the acquired reference medical image, in this embodiment of the application, the user can select a reference segmentation point from the reference medical image; wherein, the reference segmentation point is located at the edge of the target organ in the reference medical image.

[0070] In practice, users can select multiple reference segmentation points from a reference medical image.

[0071] like Figure 4 As shown, the medical imaging device displays a reference medical image. Users can determine the location of the target organ in the reference image based on their experience and select multiple reference segmentation points along the edge of the target organ in the reference image by clicking. For example... Figure 4 Points A and B are shown in the diagram.

[0072] In this embodiment of the application, after the user selects a reference segmentation point from a reference medical image, the edge of the target organ can be automatically generated based on the reference segmentation point selected by the user.

[0073] In this embodiment of the application, the edge of the target organ can be generated based on two adjacent reference segmentation points selected by the user during the process of the user selecting reference segmentation points;

[0074] For example, if the user selects point A as the reference segmentation point in a medical image, then point A is determined as the starting point for image segmentation. After the user determines the next reference segmentation point B, point B is determined as the ending point for image segmentation, and the edge of the target organ between point A and point B is generated. After determining the edge of the target organ between points A and B, point B is updated as the starting point for image segmentation. After the user determines the next reference segmentation point C, point C is determined as the ending point for image segmentation, and the edge of the target organ between points B and C is generated. This process continues until the target organ is segmented from the medical image.

[0075] Alternatively, in this embodiment of the application, after the user selects all the reference segmentation points from the reference medical image, the edge of the target organ between each two adjacent reference segmentation points can be generated;

[0076] For example, in embodiments of this application, the number of reference segmentation points can be preset, and after detecting that the user selects a preset number of reference segmentation points from the reference medical image, the edge of the target organ between each two adjacent reference segmentation points can be automatically generated; or, after the user selects multiple reference segmentation points, the edge of the target organ between each two adjacent reference segmentation points can be generated in response to the image segmentation operation triggered by the user.

[0077] The following details the process of determining the edge of the target organ between two reference segmentation points:

[0078] like Figure 5 As shown in the flowchart of this application embodiment for determining the edge of a target organ between two reference segmentation points, it may specifically include the following steps:

[0079] Step S501: For one of two adjacent reference segmentation points, generate a directed weighted graph starting from the reference segmentation point based on the pixel value of the reference segmentation point and a predefined cost function.

[0080] Given two reference split points, a directed weighted graph is generated, with one of the reference split points as the starting point (where the starting point can be a split point selected by the user beforehand).

[0081] In this directed weighted graph, the nodes are the pixels in the medical image to be segmented, the edges represent the adjacency relationship between pixels, and the weight of the edge in the directed weighted graph is the cost value between two adjacent pixels determined according to a predefined cost function.

[0082] In implementation, a reference segmentation point is used as the starting point. The eight neighboring pixels adjacent to the reference segmentation point are used as nodes in a directed weighted graph. The cost value between the reference segmentation point and its eight neighboring pixels is determined according to a predefined cost function, and this cost value is used as the edge weight value between the reference segmentation point and its corresponding neighboring pixels. Then, the eight neighboring pixels of the reference segmentation point are each used as a center pixel. For any center pixel, the neighboring pixels adjacent to that center pixel are used as nodes in the directed weighted graph. The cost value between the center pixel and each of its neighboring pixels is determined according to a predefined cost function, and this cost value is used as the edge weight value between the center pixel and its corresponding neighboring pixels. This process is repeated to generate a directed weighted graph starting from the reference segmentation point.

[0083] Optional, a predefined cost function is shown in the following formula:

[0084] Cost(p, q) = A*f l (p) + B*f g (p) + C*f gd (p, q) + D*||p, q||2

[0085] wherein, p is a reference split point or a center pixel point, q is a neighborhood pixel point, f l is a function based on Laplace operator, f g is a function based on Sobel gradient operator, f gd is a function based on gradient direction, ||p, q||2 is a two-norm based on point p and point q, A, B, C, D are weights corresponding to each function which are set in advance, and Cost(p, q) is a cost value between point p and point q.

[0086] An alternative way, in the process of generating a directed weighted graph with a reference split point as a starting point, after detecting the next reference split point, after determining the cost value between the next reference split point and the corresponding neighborhood pixel point, the generation process of the directed weighted graph can be stopped, thereby obtaining a directed weighted graph with the reference split point as a starting point and the next reference split point of the reference split point as a terminal point.

[0087] As Figure 6 shown, the embodiment of the present application is a schematic diagram of generating a directed weighted graph. For example, a user selects point A as a reference split point in a reference medical image, then determines that the starting point of image segmentation is point A, after the user determines the next reference split point point B, then determines that point B is the terminal point of image segmentation;

[0088] Firstly, taking point A as the starting point of the directed weighted graph, taking the neighborhood pixel points E1, E2, E3, E4, E5, E6, E7, E8 of point A as nodes, and according to the pre-defined cost function, the cost values of point A and its eight neighborhood pixel points E1, E2, E3, E4, E5, E6, E7, E8 are calculated respectively, and the determined cost values between point A and each neighborhood pixel point are taken as the weight values of the edges between point A and its eight neighborhood pixels. Then, taking E1, E2, E3, E4, E5, E6, E7, E8 as the center pixel point respectively, taking each neighborhood pixel point corresponding to the center pixel point as the node, and according to the pre-defined cost function, the cost values of the neighborhood pixel points of E1, E2, E3, E4, E5, E6, E7, E8 are calculated respectively when E1, E2, E3, E4, E5, E6, E7, E8 are taken as the center pixel point; for example, taking point E1 as the center pixel point, taking the neighborhood pixel points E2, A, E8, C6, C7 of point E1 as nodes, and calculating the cost values of E1 and E2, A, E8, C6, C7 respectively; and the like, until the cost values of the reference segmentation point B and its eight neighborhood pixel points B1, E7, E6, B4, B5, B6, B7, B8 are determined, and the weight values of the edges between the terminal point B and its eight neighborhood pixels are determined. Thus, the directed weighted graph with point A as the starting point and point B as the terminal point can be obtained as shown in Figure 6

[0089] Step S502, according to the generated directed weighted graph, determining the candidate path between two adjacent reference segmentation points;

[0090] In the implementation, after obtaining the directed weighted graph, the candidate path between a reference segmentation point and the next reference segmentation point is determined from the directed weighted graph;

[0091] Suppose the generated directed weighted graph is as shown in Figure 6 , then the candidate path between point A and point B is selected from the directed weighted graph as shown in Figure 6 , and the selected candidate path is as shown in Figure 7 , for example, A-E7-B, A-E6-B, A-E6-B4-B, A-E5-B4-B, and the like.

[0092] Step S503, determining the cost value of each path, and taking the path with the minimum cost value between two adjacent reference segmentation points as the edge of the target organ between the two adjacent reference segmentation points.

[0093] In the implementation, after determining the candidate path between two adjacent reference segmentation points, the cost value corresponding to each candidate path is calculated according to each candidate path;

[0094] ​Specifically, the sum of the weight values of all edges on the candidate path is taken as the generation value corresponding to the candidate path.

[0095] As shown in Figure 8 , a schematic diagram of determining the edge of the target organ between two adjacent reference segmentation points according to an embodiment of the present application is shown. As shown in Figure 7 , the generation value corresponding to each candidate path is calculated. For example, the weight value of the edge between A and E7 is 2, the weight value of the edge between E7 and B is 1, the weight value of the edge between A and E6 is 3, the weight value of the edge between E6 and B is 2, the weight value of the edge between E6 and B4 is 1, the weight value of the edge between B4 and B is 2, the weight value of the edge between A and E5 is 1, and the weight value of the edge between E5 and B4 is 2. Therefore, the generation value of the path A-E7-B is 3, the generation value of the path A-E6-B is 5, the generation value of the path A-E6-B4-B is 6, and the generation value of the path A-E5-B4-B is 5. Among them, the generation value of the path A-E7-B is the smallest, and therefore the path A-E7-B is selected as the edge of the target organ between the two adjacent reference segmentation points with point A as the starting point and point B as the ending point.

[0096] Based on the above method, the edge of the target organ in the reference medical image can be obtained in the embodiment of the present application. In addition, based on the reference medical image, the edge of the target organ in the other medical images in the medical image sequence except the reference medical image can be automatically determined in the embodiment of the present application.

[0097] The image segmentation method for any one of the medical images in the medical image sequence except the reference medical image will be described in detail below:

[0098] In the implementation, first, the target segmentation points in the to-be-segmented medical image are determined according to the selected reference segmentation points in the reference medical image, and then the image segmentation is performed according to the target segmentation points in the to-be-segmented image.

[0099] The determination process of the target segmentation points in the to-be-segmented medical image according to the selected reference segmentation points in the reference medical image will be described in detail below:

[0100] As shown in Figure 9 , a flowchart for determining the target segmentation points in the to-be-segmented medical image according to an embodiment of the present application is shown. The flowchart can specifically include the following steps:

[0101] Step S901, determining the layer interval information between the to-be-segmented medical image and the reference medical image.

[0102] The to-be-segmented medical image and the reference medical image are images of different layers in the medical image sequence obtained by scanning the target organ.

[0103] Optionally, the layer interval information between the to-be-segmented medical image and the reference medical image is that the to-be-segmented medical image and the reference medical image are spaced by a number of layers in the medical image sequence.

[0104] For example, if the reference medical image is the first image in the medical image sequence and the to-be-segmented medical image is the third image in the medical image sequence, the layer interval information num between the to-be-segmented medical image and the reference medical image is 2.

[0105] In step S902, the gradient direction corresponding to the reference segmentation point is determined according to the plurality of reference segmentation points selected in the reference medical image.

[0106] Optionally, as shown in Figure 10 The gradient direction corresponding to the reference segmentation point is determined by the following manner in the embodiment of the application:

[0107] In step S1001, the gradient value between the reference segmentation point and each neighborhood pixel point is determined according to the pixel value of the reference segmentation point and the pixel value of the neighborhood pixel point adjacent to the reference segmentation point.

[0108] In step S1002, the target neighborhood pixel point with the maximum gradient value between the reference segmentation point is determined from the neighborhood pixel points of the reference segmentation point.

[0109] In step S1003, the direction between the reference segmentation point and the target neighborhood pixel point is taken as the gradient direction corresponding to the reference segmentation point.

[0110] As shown in Figure 11 The gradient direction corresponding to the reference segmentation point is determined by the following manner in the embodiment of the application: a reference segmentation point in the reference medical image is point E, the neighborhood pixel points of point E are points E1, E2, E3, E4, E5, E6, E7, and E8, the gradient value between the reference segmentation point and points E1, E2, E3, E4, E5, E6, E7, and E8 is calculated, the neighborhood pixel point with the maximum gradient value between the reference segmentation point and its neighborhood pixel points is obtained, and the gradient direction corresponding to the reference segmentation point is determined. For example, if the gradient value between the reference segmentation point point E and its neighborhood pixel point point E5 is the maximum, the direction from the reference segmentation point point E to the neighborhood pixel point point E5 is taken as the gradient direction corresponding to the reference segmentation point point E.

[0111] In step S903, the offset information corresponding to the reference segmentation point is determined according to the layer interval information between the to-be-segmented medical image and the reference medical image.

[0112] In the embodiment of the application, the offset information corresponding to the reference segmentation point includes a plane offset value and a depth offset value.

[0113] Optionally, the plane offset value corresponding to the reference segmentation point can be determined by the following manner in the embodiment of the application:

[0114] According to the pixel spacing in the medical image, a pixel offset value between two adjacent medical images in the medical image sequence is determined; according to the layer interval information between the to-be-segmented medical image and the reference medical image and the pixel offset value, a plane offset value corresponding to the reference segmentation point is determined.

[0115] It should be noted that the present embodiment defines the pixel spacing between each pixel point and its left and right adjacent pixel points in the medical image as spacingX, and the pixel spacing between each pixel point and its upper and lower adjacent pixel points in the medical image as spacingY, and the layer spacing between two adjacent medical images as spacingZ.

[0116] As shown in FIG. 1, points A, B and C are pixel points in any one medical image in the medical image sequence, the pixel spacing between the pixel point A and the pixel point B is spacingX, and the pixel spacing between the pixel point A and the pixel point C is spacingY. Figure 12A As shown in FIG. 2, the layer spacing between any two adjacent medical images is spacingZ. Figure 12B

[0117] Optionally, the present embodiment determines the pixel offset value corresponding to the reference segmentation point by the following formula:

[0118]

[0119] wherein spacingX and spacingY are the pixel spacing between the pixel points in the medical image, num is the layer interval information between the to-be-segmented medical image and the reference medical image, and d is the pixel offset value corresponding to the reference segmentation point.

[0120] The present embodiment can determine the plane offset value corresponding to the reference segmentation point according to the following formula:

[0121]

[0122] wherein num is the layer interval information between the to-be-segmented medical image and the reference medical image, and D is the plane offset value corresponding to the reference segmentation point.

[0123] Optionally, the present embodiment can determine the depth offset value corresponding to the reference segmentation point according to the following manner:

[0124] According to the layer spacing between two adjacent medical images in the medical image sequence and the layer interval information between the to-be-segmented medical image and the reference medical image, the depth offset value corresponding to the reference segmentation point is determined.

[0125] In the implementation, the present embodiment can determine the depth offset value corresponding to the reference segmentation point according to the following formula: ​

[0126] dz = num * spacingZ

[0127] Where num represents the layer spacing information between the medical image to be segmented and the reference medical image, spacingZ represents the layer spacing, and dz represents the depth offset value corresponding to the reference segmentation point.

[0128] Step S904: Determine the target segmentation point in the medical image to be segmented based on the gradient direction and offset information corresponding to the determined multiple reference segmentation points.

[0129] In this embodiment of the application, after determining the planar offset value and depth offset value corresponding to the reference segmentation point, the target segmentation point in the medical image to be segmented is determined based on the position information of the reference segmentation point in the reference medical image, the planar offset value and depth offset value corresponding to the reference segmentation point, and the gradient direction.

[0130] Optionally, based on the planar offset value and gradient direction corresponding to the reference segmentation point, the horizontal and vertical coordinate offset values ​​corresponding to the reference segmentation point are determined respectively; and based on the horizontal coordinate value and horizontal coordinate offset value of the reference segmentation point, the horizontal coordinate value of the target segmentation point in the medical image to be segmented corresponding to the reference segmentation point is determined; and based on the vertical coordinate value and vertical coordinate offset value of the reference segmentation point, the vertical coordinate value of the target segmentation point in the medical image to be segmented corresponding to the reference segmentation point is determined; and

[0131] Based on the depth coordinates and depth offset of the reference segmentation point, the depth coordinates of the target segmentation point in the medical image to be segmented, corresponding to the reference segmentation point, are determined.

[0132] like Figure 13 The diagram shown illustrates how the horizontal and vertical coordinate offset values ​​are determined based on the planar offset values ​​in this embodiment of the application.

[0133] In the figure, point E is the reference segmentation point in the reference medical image, the gradient direction makes an angle α with the X-axis, the plane offset value is D, then the horizontal coordinate offset value is D*cosα, and the vertical coordinate offset value is D*sina.

[0134] Since the location information of the reference segmentation point in the reference medical image includes the x-coordinate, y-coordinate, and depth coordinates, determining the location of the target segmentation point in the medical image to be segmented requires determining the x-coordinate, y-coordinate, and depth coordinates of the target segmentation point. These are described below:

[0135] 1. The x-coordinate value of the target segmentation point in the medical image to be segmented;

[0136] In implementation, the horizontal coordinate value of the target segmentation point corresponding to the reference segmentation point in the medical image to be segmented is determined according to the horizontal coordinate value of the reference segmentation point and the horizontal coordinate offset value.

[0137] Specifically, the sum of the horizontal coordinate value of the reference segmentation point and the horizontal coordinate offset value is taken as the horizontal coordinate value of the target segmentation point.

[0138] 2, the vertical coordinate value of the target segmentation point in the medical image to be segmented;

[0139] In implementation, the vertical coordinate value of the target segmentation point corresponding to the reference segmentation point in the medical image to be segmented is determined according to the vertical coordinate value of the reference segmentation point and the vertical coordinate offset value.

[0140] Specifically, the sum of the vertical coordinate value of the reference segmentation point and the vertical coordinate offset value is taken as the vertical coordinate value of the target segmentation point.

[0141] 3, the depth coordinate value of the target segmentation point in the medical image to be segmented;

[0142] In implementation, the depth coordinate value of the target segmentation point corresponding to the reference segmentation point in the medical image to be segmented is determined according to the depth coordinate value of the reference segmentation point and the depth offset value.

[0143] Specifically, the sum of the depth coordinate value of the reference segmentation point and the depth coordinate offset value is taken as the depth coordinate value of the target segmentation point.

[0144] In order to facilitate understanding, the position of the reference segmentation point in the reference medical image will be illustrated first. As shown in the figure, Figure 14 A three-dimensional coordinate diagram of the position of a reference segmentation point according to an embodiment of the present application is shown. Wherein, point E is a reference segmentation point in a reference medical image, for example, the coordinates of point E are (x, y, z), then the position of the reference segmentation point E corresponding to the X axis is x, the position corresponding to the Y axis is y, and the depth corresponding to the Z axis is z.

[0145] In the embodiment of the present application, the position of the target segmentation point in the medical image to be segmented is determined according to the position information of the reference segmentation point, and the horizontal coordinate offset value, the vertical coordinate offset value and the depth offset value corresponding to the reference segmentation point.

[0146] As shown in the figure, Figure 15 A three-dimensional diagram for determining a target segmentation point according to a reference segmentation point according to an embodiment of the present application is shown. Figure 15 In the figure, point E is a reference segmentation point in a reference medical image, and the coordinates of point E are E(x, y, z). According to the horizontal coordinate offset value dx, the vertical coordinate offset value dy and the depth offset value dz corresponding to the reference segmentation point, the coordinates of the target segmentation point in the medical image to be segmented can be obtained as Figure 15A target segmentation point in a to-be-segmented medical image, the position of the target segmentation point in the to-be-segmented medical image being F(x+dx, y+dy, z+dz).

[0147] As shown in FIG. 1, point E is a reference segmentation point in a reference medical image, and according to the determined plane offset value and depth offset value, a corresponding target segmentation point F in a to-be-segmented medical image can be obtained. Figure 16

[0148] After determining the target segmentation points in the to-be-segmented medical image corresponding to each reference segmentation point, the accuracy of the target segmentation points is verified. If the accuracy verification of the target segmentation points passes, image segmentation is performed according to the determined target segmentation points. If the accuracy verification of the target segmentation points fails, the target segmentation points are modified.

[0149] The process of verifying the accuracy of the target segmentation points is described in detail as follows.

[0150] As shown in FIG. 1, the flowchart of verifying the accuracy of the target segmentation points according to an embodiment of the present application can include the following steps. Figure 17 In step S1701, for any target segmentation point, the HU value of the target segmentation point and the HU value of the reference segmentation point corresponding to the target segmentation point are obtained.

[0151] The HU value is the pixel value of each pixel point in the medical image collected by the medical device. If the medical image is a CT image, the HU value is also a CT value.

[0152] It should be noted that in the present embodiment, when verifying the target segmentation points in the to-be-segmented medical image, the determined target segmentation point can be verified after being determined according to one reference segmentation point. Alternatively, all the determined target segmentation points can be verified after being determined.

[0153] In step S1702, the difference between the HU value of the target segmentation point and the HU value of the corresponding reference segmentation point is determined.

[0154] In step S1703, it is determined whether the difference between the HU value of the target segmentation point and the HU value of the corresponding reference segmentation point is within a set range. If yes, step S1704 is performed, and if no, step S1705 is performed.

[0155] The set range in the present embodiment can be a pre-set fixed range, or the set range can be determined according to the HU value of the corresponding reference segmentation point. For example, the set range is

[0156] ​​wherein a is a HU value of a reference segmentation point.

[0157] In step S1704, it is determined that the target segmentation point passes the check.

[0158] In step S1705, gradient values between the reference segmentation point and each of the neighboring pixel points are determined, and position information of a neighboring pixel point with the largest gradient value is taken as position information of a new target segmentation point.

[0159] After the target segmentation point in the medical image to be segmented is obtained by using the above-mentioned manner, the embodiment of the present application determines the edge of the target organ in the medical image to be segmented according to the determined target segmentation point.

[0160] Optionally, for one of the two adjacent target segmentation points, a directed weighted graph is generated with the one target segmentation point as a starting point according to a pixel value of the one target segmentation point and a predefined cost function; wherein a node in the directed weighted graph is a pixel point in the medical image to be segmented, an edge in the directed weighted graph represents an adjacent relationship between pixels, and a weight of an edge in the directed weighted graph is a cost value between two adjacent pixel points corresponding to the edge determined according to the predefined cost function; a cost value of each path between the two adjacent target segmentation points is determined according to the generated directed weighted graph; and a path with the minimum cost value between the two adjacent target segmentation points is taken as the edge of the target organ between the two adjacent target segmentation points.

[0161] It should be noted that the detailed process of determining the edge of the target organ in the medical image to be segmented according to the target segmentation point in the medical image to be segmented can refer to the process of determining the edge of the target organ in the medical image according to the reference segmentation point in the reference medical image, which will not be described in detail here.

[0162] As shown in FIG. 19, an effect diagram of the edge segmentation of the target organ in the medical image according to the embodiment of the present application. Figure 18 Based on the same inventive concept, the embodiment of the present application provides a medical image segmentation device; as shown in FIG. 19, the medical image segmentation device 1900 includes at least one processor 1901 and at least one memory 1902.

[0163] Figure 19 The memory 1902 stores program code, and when the program code is executed by the processor 1901, the processor performs the following processes:

[0164] The memory 1902 stores program code, and when the program code is executed by the processor 1901, the processor performs the following processes:

[0165] ​determining layer interval information between the to-be-segmented medical image and the reference medical image, wherein the to-be-segmented medical image and the reference medical image are images of different layers in a medical image sequence obtained by scanning a target organ;

[0166] determining, according to a plurality of reference segmentation points selected in the reference medical image and the layer interval information between the to-be-segmented medical image and the reference medical image, target segmentation points in the to-be-segmented medical image respectively corresponding to the reference segmentation points;

[0167] determining, for any two adjacent target segmentation points in the to-be-segmented medical image, an edge of the target organ between the two adjacent target segmentation points according to pixel values of the two adjacent target segmentation points and a predefined cost function;

[0168] segmenting the target organ from the to-be-segmented medical image according to the edges of the target organ between each two adjacent target segmentation points.

[0169] Optionally, the processor 1901 is specifically configured to:

[0170] performing the following operations respectively for any one reference segmentation point in the reference medical image:

[0171] determining a gradient direction corresponding to the reference segmentation point according to a pixel value of the reference segmentation point and pixel values of adjacent pixel points adjacent to the reference segmentation point; and

[0172] determining offset information corresponding to the reference segmentation point according to the layer interval information between the to-be-segmented medical image and the reference medical image;

[0173] Optionally, the processor 1901 is specifically configured to:

[0174] determining a gradient value between the reference segmentation point and each adjacent pixel point according to a pixel value of the reference segmentation point and pixel values of adjacent pixel points adjacent to the reference segmentation point;

[0175] determining a target adjacent pixel point with a maximum gradient value between the reference segmentation point and the target adjacent pixel point among the adjacent pixel points of the reference segmentation point;

[0176] taking a direction between the reference segmentation point and the target adjacent pixel point as the gradient direction corresponding to the reference segmentation point.

[0177] Optionally, the offset information corresponding to the reference segmentation point includes a plane offset value and a depth offset value.

[0178] The processor 1901 is specifically configured to:

[0179] determine a pixel offset value between two adjacent medical images in the medical image sequence according to a pixel interval in the medical images, determine a plane offset value corresponding to the reference segmentation point according to the layer interval information between the medical image to be segmented and the reference medical image and the pixel offset value, and

[0180] determine a depth offset value corresponding to the reference segmentation point according to the layer interval information between the medical image to be segmented and the reference medical image and a layer interval between two adjacent medical images in the medical image sequence.

[0181] Optionally, the position information of the reference segmentation point in the reference medical image includes a horizontal coordinate value, a vertical coordinate value and a depth coordinate value.

[0182] The processor 1901 is specifically configured to:

[0183] determine a horizontal coordinate offset value and a vertical coordinate offset value corresponding to the reference segmentation point respectively according to the plane offset value corresponding to the reference segmentation point and the gradient direction, determine a horizontal coordinate value of a target segmentation point corresponding to the reference segmentation point in the medical image to be segmented according to the horizontal coordinate value of the reference segmentation point and the horizontal coordinate offset value, and determine a vertical coordinate value of the target segmentation point corresponding to the reference segmentation point in the medical image to be segmented according to the vertical coordinate value of the reference segmentation point and the vertical coordinate offset value; and

[0184] determine a depth coordinate value of a target segmentation point corresponding to the reference segmentation point in the medical image to be segmented according to the depth coordinate value of the reference segmentation point and the depth offset value.

[0185] Optionally, the processor 1901 is further configured to:

[0186] for any target segmentation point, obtain a HU value of the target segmentation point and a HU value of a reference segmentation point corresponding to the target segmentation point;

[0187] determine that a difference between the HU value of the target segmentation point and the HU value of the corresponding reference segmentation point is within a set range.

[0188] Optionally, the processor 1901 is further configured to:

[0189] if the difference between the HU value of the target segmentation point and the HU value of the corresponding reference segmentation point is not within the set range, determine gradient values between the reference segmentation point and each adjacent pixel point, and take position information of an adjacent pixel point with the largest gradient value as new position information of the target segmentation point.

[0190] Optionally, the processor 1901 is specifically configured to:

[0191] For one of the two adjacent target segmentation points, a directed weighted graph starting from the one target segmentation point is generated according to a pixel value of the one target segmentation point and a predefined cost function; wherein a node in the directed weighted graph is a pixel point in the medical image to be segmented, an edge in the directed weighted graph represents a neighboring relationship between pixels, and a weight of an edge in the directed weighted graph is a cost value between two adjacent pixel points corresponding to the edge determined according to the predefined cost function.

[0192] According to the generated directed weighted graph, a cost value of each path between the two adjacent target segmentation points is determined; and a path with a minimum cost value between the two adjacent target segmentation points is taken as an edge of the target organ between the two adjacent target segmentation points.

[0193] As shown in Figure 20 The embodiment of the present application provides a medical image segmentation device 2000, which can be applied to the medical image device in the above embodiment, and comprises:

[0194] An acquisition module 2001 is configured to determine layer interval information between a medical image to be segmented and a reference medical image; wherein the medical image to be segmented and the reference medical image are images of different layers in a medical image sequence obtained by scanning a target organ;

[0195] A determination module 2002 is configured to determine, according to a plurality of reference segmentation points selected in the reference medical image and the layer interval information between the medical image to be segmented and the reference medical image, target segmentation points in the medical image to be segmented corresponding to the reference segmentation points respectively;

[0196] An edge detection module 2003 is configured to, for any two adjacent target segmentation points in the medical image to be segmented, determine an edge of the target organ between the two adjacent target segmentation points according to pixel values of the two adjacent target segmentation points and a predefined cost function;

[0197] A segmentation module 2004 is configured to segment the target organ from the medical image to be segmented according to the determined edge of the target organ between each two adjacent target segmentation points.

[0198] Optionally, the determination module 2002 is specifically configured to:

[0199] For any one reference segmentation point in the reference medical image, the following operations are performed respectively:

[0200] According to a pixel value of the reference segmentation point and pixel values of neighboring pixel points adjacent to the reference segmentation point, a gradient direction corresponding to the reference segmentation point is determined; and

[0201] determine offset information corresponding to the reference segmentation point according to interlayer interval information between the medical image to be segmented and the reference medical image;

[0202] determine a target segmentation point in the medical image to be segmented corresponding to the reference segmentation point according to position information of the reference segmentation point in the reference medical image, offset information corresponding to the reference segmentation point, and a gradient direction.

[0203] Optionally, the determining module 2002 is specifically configured to:

[0204] determine a gradient value between the reference segmentation point and each adjacent pixel point according to a pixel value of the reference segmentation point and a pixel value of the adjacent pixel point adjacent to the reference segmentation point;

[0205] determine a target adjacent pixel point with the largest gradient value between the reference segmentation point and the target adjacent pixel point among the adjacent pixel points of the reference segmentation point;

[0206] take a direction between the reference segmentation point and the target adjacent pixel point as the gradient direction corresponding to the reference segmentation point.

[0207] Optionally, the offset information corresponding to the reference segmentation point includes a plane offset value and a depth offset value.

[0208] The determining module 2002 is specifically configured to:

[0209] determine a pixel offset value between two adjacent medical images in the medical image sequence according to a pixel interval in the medical image; determine a plane offset value corresponding to the reference segmentation point according to interlayer interval information between the medical image to be segmented and the reference medical image and the pixel offset value; and

[0210] determine a depth offset value corresponding to the reference segmentation point according to an interlayer interval between two adjacent medical images in the medical image sequence and the interlayer interval information between the medical image to be segmented and the reference medical image.

[0211] Optionally, the position information of the reference segmentation point in the reference medical image includes an abscissa value, an ordinate value and a depth coordinate value.

[0212] The determining module 2002 is specifically configured to:

[0213] determine a horizontal coordinate offset value and a vertical coordinate offset value of the reference segmentation point according to the plane offset value corresponding to the reference segmentation point and the gradient direction; determine a horizontal coordinate value of a target segmentation point corresponding to the reference segmentation point in the medical image to be segmented according to a horizontal coordinate value of the reference segmentation point and the horizontal coordinate offset value; and determine a vertical coordinate value of the target segmentation point corresponding to the reference segmentation point in the medical image to be segmented according to a vertical coordinate value of the reference segmentation point and the vertical coordinate offset value; and

[0214] determine a depth coordinate value of a target segmentation point corresponding to the reference segmentation point in the medical image to be segmented according to a depth coordinate value of the reference segmentation point and the depth offset value.

[0215] Optionally, the determining module 2002 is further configured to:

[0216] for any target segmentation point, obtain a HU value of the target segmentation point and a HU value of a reference segmentation point corresponding to the target segmentation point;

[0217] determine that a difference between the HU value of the target segmentation point and the HU value of the corresponding reference segmentation point is within a set range.

[0218] Optionally, the determining module 2002 is further configured to:

[0219] if the difference between the HU value of the target segmentation point and the HU value of the corresponding reference segmentation point is not within the set range, determine gradient values between the reference segmentation point and each adjacent pixel point, and take position information of an adjacent pixel point with the largest gradient value as position information of a new target segmentation point.

[0220] Optionally, the edge detection module 2003 is specifically configured to:

[0221] for one of the two adjacent target segmentation points, generate a directed weighted graph with the one target segmentation point as a starting point according to a pixel value of the one target segmentation point and a predefined cost function; wherein a node in the directed weighted graph is a pixel point in the medical image to be segmented, an edge in the directed weighted graph represents an adjacent relationship between pixels, and a weight of an edge in the directed weighted graph is a cost value between two adjacent pixel points determined according to the predefined cost function;

[0222] determine a cost value of each path between the two adjacent target segmentation points according to the generated directed weighted graph; and determine a cost value between the two adjacent target segmentation points.

[0223] In an example embodiment, a computer readable storage medium including instructions, such as a memory including instructions, is also provided. The instructions can be executable by a processor to perform the above-described method for segmenting a medical image. Optionally, the storage medium can be a non-transitory computer readable storage medium, such as a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disc, and an optical data storage device, etc.

[0224] In an example embodiment, a computer program product including a computer program is also provided. The computer program, when executed by a processor, implements any of the above-described methods for segmenting a medical image.

[0225] In an example embodiment, each of the above-described methods for segmenting a medical image can also be implemented as a program product, which includes a program code. When the program product is run on a computer device, the program code causes the computer device to perform the steps of the above-described methods for segmenting a medical image according to various example embodiments of the present disclosure.

[0226] The program product can take any combination of one or more readable media. The readable media can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disc, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0227] The program product for the method for segmenting a medical image according to the embodiments of the present disclosure can take a portable compact disc read-only memory (CD-ROM) and include a program code, and can be run on an electronic device. However, the program product of the present disclosure is not limited thereto, and in this document, the readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0228] Those skilled in the art will appreciate that embodiments of the present application can be devised for a variety of other systems which are currently developed or later developed. Therefore, the present application is intended to cover all such modifications and variations of this application that are within the scope of the appended claims and their equivalents. It is intended that each element of claim 1 and 2 is independent of one another. No element of claim 1 and 2, or any other claim, is implied to depend on any other element or limitation of claim 1 and 2 or any other claim except where expressly recited in that claim.

[0229] The present application is described in reference to the flowchart and / or block diagrams of the method, apparatus (system) and computer program product according to this application. It will be understood that each block of the flowchart and / or block diagrams, and combinations of blocks in the flowchart and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0230] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0231] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0232] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method for segmenting medical images, characterized in that, The method includes: A medical image is selected from a medical image sequence as a reference medical image. The reference medical image is selected from the medical image sequence according to a pre-set rule, or the reference medical image is any medical image in the medical image sequence. Determine the layer spacing information between the medical image to be segmented and the reference medical image; wherein the medical image to be segmented and the reference medical image are images of different layers in a medical image sequence obtained by scanning a target organ; For any reference segmentation point in the reference medical image, perform the following operations: determine the gradient direction corresponding to the reference segmentation point based on the pixel value corresponding to the reference segmentation point and the pixel values ​​of the neighboring pixels adjacent to the reference segmentation point; determine the offset information corresponding to the reference segmentation point based on the layer spacing information between the medical image to be segmented and the reference medical image; and determine the target segmentation point in the medical image to be segmented corresponding to the reference segmentation point based on the position information of the reference segmentation point in the reference medical image, the offset information corresponding to the reference segmentation point, and the gradient direction. For any two adjacent target segmentation points in the medical image to be segmented, the edge of the target organ between the two adjacent target segmentation points is determined based on the pixel values ​​of the two adjacent target segmentation points and a predefined cost function. The target organ is segmented from the medical image to be segmented based on the edge of the target organ between every two adjacent target segmentation points.

2. The method as described in claim 1, characterized in that, Determining the gradient direction corresponding to the reference segmentation point based on the pixel value corresponding to the reference segmentation point and the pixel values ​​of the neighboring pixels adjacent to the reference segmentation point includes: Based on the pixel value corresponding to the reference segmentation point and the pixel values ​​of the neighboring pixels adjacent to the reference segmentation point, the gradient value between the reference segmentation point and each neighboring pixel is determined. Determine the target neighbor pixel with the largest gradient value among the neighbor pixels of the reference segmentation point; The direction between the reference segmentation point and the target neighboring pixel is taken as the gradient direction corresponding to the reference segmentation point.

3. The method as described in claim 1, characterized in that, The offset information corresponding to the reference segmentation point includes planar offset value and depth offset value; The step of determining the offset information corresponding to the reference segmentation point based on the layer spacing information between the medical image to be segmented and the reference medical image includes: Based on the pixel spacing in the medical images, determine the pixel offset value between two adjacent medical images in the medical image sequence; based on the layer spacing information between the medical image to be segmented and the reference medical image and the pixel offset value, determine the planar offset value corresponding to the reference segmentation point; and The depth offset value corresponding to the reference segmentation point is determined based on the layer spacing between two adjacent medical images in the medical image sequence and the layer spacing information between the medical image to be segmented and the reference medical image.

4. The method as described in claim 3, characterized in that, The location information of the reference segmentation point in the reference medical image includes the horizontal coordinate value, the vertical coordinate value, and the depth coordinate value; The step of determining the target segmentation point in the medical image to be segmented, corresponding to the reference segmentation point, based on the position information of the reference segmentation point in the reference medical image, the offset information corresponding to the reference segmentation point, and the gradient direction, includes: Based on the planar offset value corresponding to the reference segmentation point and the gradient direction, the horizontal and vertical coordinate offset values ​​corresponding to the reference segmentation point are determined respectively; and based on the horizontal coordinate value and the horizontal coordinate offset value of the reference segmentation point, the horizontal coordinate value of the target segmentation point in the medical image to be segmented corresponding to the reference segmentation point is determined; and based on the vertical coordinate value and the vertical coordinate offset value of the reference segmentation point, the vertical coordinate value of the target segmentation point in the medical image to be segmented corresponding to the reference segmentation point is determined; and Based on the depth coordinates of the reference segmentation point and the depth offset value, the depth coordinates of the target segmentation point in the medical image to be segmented, corresponding to the reference segmentation point, are determined.

5. The method according to any one of claims 1 to 4, characterized in that, After determining the target segmentation points in the medical image to be segmented that correspond to each reference segmentation point, and before determining the edge of the target organ between two adjacent target segmentation points based on the pixel values ​​of the two adjacent target segmentation points and a predefined cost function, the method further includes: For any target segmentation point, obtain the HU value of the target segmentation point and the HU value of the reference segmentation point corresponding to the target segmentation point; The difference between the HU value of the target segmentation point and the HU value of the corresponding reference segmentation point is determined to be within a set range.

6. The method as described in claim 5, characterized in that, The method further includes: If the difference between the HU value of the target segmentation point and the HU value of the corresponding reference segmentation point is not within the set range, then the gradient value between the reference segmentation point and each neighboring pixel is determined, and the position information of the neighboring pixel with the largest gradient value is used as the position information of the new target segmentation point.

7. The method according to any one of claims 1 to 4, characterized in that, The step of determining the edge of the target organ between two adjacent target segmentation points based on the pixel values ​​of the two adjacent target segmentation points and a predefined cost function includes: For one of the two adjacent target segmentation points, a directed weighted graph is generated starting from the target segmentation point based on the pixel value of the target segmentation point and a predefined cost function; wherein the nodes in the directed weighted graph are the pixels in the medical image to be segmented, the edges in the directed weighted graph represent the adjacency relationship between pixels, and the weight of the edge in the directed weighted graph is the cost value between two adjacent pixels determined according to the predefined cost function; Based on the generated directed weighted graph, the cost value of each path between the two adjacent target segmentation points is determined; and the path with the minimum cost value between the two adjacent target segmentation points is taken as the edge of the target organ between the two adjacent target segmentation points.

8. A medical image segmentation device, characterized in that, The device includes at least one processor and at least one memory; The memory stores program code, which, when executed by the processor, causes the processor to perform the following processes: A medical image is selected from a medical image sequence as a reference medical image. The reference medical image is selected from the medical image sequence according to a pre-set rule, or the reference medical image is any medical image in the medical image sequence. Determine the layer spacing information between the medical image to be segmented and the reference medical image; wherein the medical image to be segmented and the reference medical image are images of different layers in a medical image sequence obtained by scanning a target organ; For any reference segmentation point in the reference medical image, perform the following operations: determine the gradient direction corresponding to the reference segmentation point based on the pixel value corresponding to the reference segmentation point and the pixel values ​​of the neighboring pixels adjacent to the reference segmentation point; determine the offset information corresponding to the reference segmentation point based on the layer spacing information between the medical image to be segmented and the reference medical image; and determine the target segmentation point in the medical image to be segmented corresponding to the reference segmentation point based on the position information of the reference segmentation point in the reference medical image, the offset information corresponding to the reference segmentation point, and the gradient direction. For any two adjacent target segmentation points in the medical image to be segmented, the edge of the target organ between the two adjacent target segmentation points is determined based on the pixel values ​​of the two adjacent target segmentation points and a predefined cost function. The target organ is segmented from the medical image to be segmented based on the edge of the target organ between every two adjacent target segmentation points.

9. A medical image segmentation device, characterized in that, The device includes: The acquisition module is used to determine a medical image as a reference medical image from a medical image sequence. The reference medical image is selected from the medical image sequence according to a pre-set rule, or the reference medical image is any medical image in the medical image sequence. The module also determines the layer spacing information between the medical image to be segmented and the reference medical image. The medical image to be segmented and the reference medical image are images from different layers in the medical image sequence obtained by scanning a target organ. The determination module performs the following operations for any reference segmentation point in the reference medical image: determining the gradient direction corresponding to the reference segmentation point based on the pixel value corresponding to the reference segmentation point and the pixel values ​​of the neighboring pixels adjacent to the reference segmentation point; determining the offset information corresponding to the reference segmentation point based on the layer spacing information between the medical image to be segmented and the reference medical image; and determining the target segmentation point in the medical image to be segmented corresponding to the reference segmentation point based on the position information of the reference segmentation point in the reference medical image, the offset information corresponding to the reference segmentation point, and the gradient direction. The edge detection module determines the edge of the target organ between any two adjacent target segmentation points in the medical image to be segmented, based on the pixel values ​​of the two adjacent target segmentation points and a predefined cost function. The segmentation module segments the target organ from the medical image to be segmented based on the edge of the target organ between every two adjacent target segmentation points.

Images