Tissue image registration method and related products

Abstract

The present application discloses a tissue image registration method and related products. The method includes: acquiring a first image and a second image, both of which include target tissue; obtaining a first segmentation result by segmenting the target tissue in the first image; obtaining a second segmentation result by segmenting the target tissue in the second image; inputting the first segmentation result and the second segmentation result into a registration model to obtain first registration parameters, the first registration parameters being used to register the target tissue in the first image and the target tissue in the second image, the registration model being capable of determining the registration parameters of the target tissues in the two images. Corresponding products are also disclosed. The tissue image registration method can reduce the interference of image content other than the target tissue in the first image and image content other than the target tissue in the second image on the determination of the first registration parameters, thereby improving the accuracy of the first registration parameters.

Description

Technical Field

[0001] The present application relates to the field of medical image processing, and in particular to a tissue image registration method and related products. Background Art

[0002] In the medical field, it is often necessary to register tissues in different medical images. Specifically, when the number of medical images containing a tissue is greater than one and the tissues are located in different locations in the different medical images, registration of the target tissue in the different medical images is required.

[0003] The current approach is to register tissues in different medical images, determine registration parameters for registering the tissues in the different medical images, and then transform the tissues in the medical images based on the registration parameters to align the tissues in the different medical images.

[0004] However, the accuracy of registration parameters determined by current practices is low. Summary of the Invention

[0005] The present application provides a tissue image registration method and related products, wherein the tissue image is an image including tissue. The method can improve the accuracy of registration parameters of the tissue image.

[0006] In a first aspect, a method for registering tissue images is provided, the method comprising:

[0007] Acquiring a first image and a second image, wherein both the first image and the second image include target tissue;

[0008] Obtaining a first segmentation result by segmenting the target tissue in the first image;

[0009] obtaining a second segmentation result by segmenting the target tissue in the second image;

[0010] The first segmentation result and the second segmentation result are input into a registration model to obtain first registration parameters, wherein the first registration parameters are used to align the target tissue in the first image and the target tissue in the second image, and the registration model has the ability to determine the registration parameters of the target tissue in the two images.

[0011] In combination with any embodiment of the present application, the method further includes:

[0012] determining n first regions from the target tissue in the first segmentation result;

[0013] determining n second regions from the target tissue in the second segmentation result;

[0014] Obtaining n second registration parameters by registering the n first regions with the n second regions, wherein the second registration parameters are used to register corresponding regions among the n first regions and the n second regions;

[0015] Based on the first registration parameter and the n second registration parameters, a third registration parameter is determined, where the third registration parameter is used to register the target tissue in the first image and the target tissue in the second image.

[0016] In combination with any embodiment of the present application, determining a third registration parameter based on the first registration parameter and the n second registration parameters includes:

[0017] determining, based on the first registration parameter, m fourth registration parameters and t fifth registration parameters from the n second registration parameters, where m and t are both less than n, the sum of m and t is n, a difference between the fourth registration parameter and the first registration parameter is greater than or equal to a threshold, and a difference between the fifth registration parameter and the first registration parameter is less than the threshold;

[0018] Determining a sixth registration parameter based on the t fifth registration parameters, wherein the sixth registration parameter is used to register the areas corresponding to the t fifth registration parameters;

[0019] The third registration parameter is determined based on the first registration parameter, the m fourth registration parameters, and the sixth registration parameter.

[0020] In combination with any embodiment of the present application, determining the third registration parameter based on the first registration parameter, the m fourth registration parameters, and the sixth registration parameter includes:

[0021] determining a seventh registration parameter based on the m fourth registration parameters and the sixth registration parameter, wherein the seventh registration parameter is used to register the target tissue in the first image and the target tissue in the second image;

[0022] The third registration parameter is determined based on the first registration parameter and the seventh registration parameter.

[0023] In conjunction with any embodiment of the present application, the target tissue belongs to the target object;

[0024] The determining n first regions from the target tissue in the first segmentation result includes:

[0025] determining, based on the target tissue in the first segmentation result, n first regions, where the n first regions include a third region, and the third region is any region among the n first regions, and the farther the third region is from the lung of the target subject, the larger the area of ​​the third region;

[0026] The determining n second regions from the target tissue in the second segmentation result includes:

[0027] Based on the target tissue in the second segmentation result, n second regions are determined, where the n second regions include a fourth region, and the fourth region is any region among the n second regions. The farther the fourth region is from the lung of the target object, the larger the area of ​​the fourth region.

[0028] In conjunction with any embodiment of the present application, the first image and the second image are both acquired during the breathing process of the target object, and the acquisition time of the first image is different from the acquisition time of the second image;

[0029] After determining the third registration parameter, the method further includes:

[0030] transforming the target tissue in the first image based on the third registration parameter to obtain a third image;

[0031] A lesion in the target tissue is determined based on the second image and the third image.

[0032] In a second aspect, a tissue image registration device is provided, the tissue image registration device comprising:

[0033] an acquisition unit, configured to acquire a first image and a second image, wherein both the first image and the second image include target tissue;

[0034] a segmentation unit, configured to obtain a first segmentation result by segmenting the target tissue in the first image;

[0035] The segmentation unit is further configured to obtain a second segmentation result by segmenting the target tissue in the second image;

[0036] A processing unit is used to input the first segmentation result and the second segmentation result into a registration model to obtain a first registration parameter, wherein the first registration parameter is used to align the target tissue in the first image and the target tissue in the second image, and the registration model has the ability to determine the registration parameters of the target tissue in the two images.

[0037] In combination with any embodiment of the present application, the tissue image registration device further includes:

[0038] a determining unit, configured to determine n first regions from the target tissue in the first segmentation result;

[0039] The determining unit is further configured to determine n second regions from the target tissue in the second segmentation result;

[0040] The processing unit is further configured to obtain n second registration parameters by registering the n first regions with the n second regions, wherein the second registration parameters are used to register corresponding regions among the n first regions and the n second regions;

[0041] The determining unit is further configured to determine a third registration parameter based on the first registration parameter and the n second registration parameters, wherein the third registration parameter is used to register the target tissue in the first image and the target tissue in the second image.

[0042] In combination with any embodiment of the present application, the determining unit is further configured to:

[0043] determining, based on the first registration parameter, m fourth registration parameters and t fifth registration parameters from the n second registration parameters, where m and t are both less than n, the sum of m and t is n, a difference between the fourth registration parameter and the first registration parameter is greater than or equal to a threshold, and a difference between the fifth registration parameter and the first registration parameter is less than the threshold;

[0044] Determining a sixth registration parameter based on the t fifth registration parameters, wherein the sixth registration parameter is used to register the areas corresponding to the t fifth registration parameters;

[0045] The third registration parameter is determined based on the first registration parameter, the m fourth registration parameters, and the sixth registration parameter.

[0046] In combination with any embodiment of the present application, the determining unit is further configured to:

[0047] determining a seventh registration parameter based on the m fourth registration parameters and the sixth registration parameter, wherein the seventh registration parameter is used to register the target tissue in the first image and the target tissue in the second image;

[0048] The third registration parameter is determined based on the first registration parameter and the seventh registration parameter.

[0049] In conjunction with any embodiment of the present application, the target tissue belongs to the target object;

[0050] The determining unit is further configured to:

[0051] determining, based on the target tissue in the first segmentation result, n first regions, where the n first regions include a third region, and the third region is any region among the n first regions, and the farther the third region is from the lung of the target subject, the larger the area of ​​the third region;

[0052] Based on the target tissue in the second segmentation result, n second regions are determined, where the n second regions include a fourth region, and the fourth region is any region among the n second regions. The farther the fourth region is from the lung of the target object, the larger the area of ​​the fourth region.

[0053] In conjunction with any embodiment of the present application, the first image and the second image are both acquired during the breathing process of the target object, and the acquisition time of the first image is different from the acquisition time of the second image;

[0054] The processing unit is further configured to transform the target tissue in the first image based on the third registration parameter to obtain a third image;

[0055] The tissue image registration device further includes: a determination unit configured to determine a lesion in the target tissue based on the second image and the third image.

[0056] In a third aspect, an electronic device is provided, comprising: a processor and a memory, wherein the memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes the method as described in the first aspect above and any possible implementation method thereof.

[0057] In a fourth aspect, another electronic device is provided, comprising: a processor, a sending device, an input device, an output device and a memory, wherein the memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes the method as described in the first aspect above and any possible implementation method thereof.

[0058] In a fifth aspect, a computer-readable storage medium is provided, in which a computer program is stored. The computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to execute the method as described in the first aspect above and any possible implementation method thereof.

[0059] In a sixth aspect, a computer program product is provided, which includes a computer program or instructions, and when the computer program or instructions are run on a computer, the computer is enabled to execute the method of the above-mentioned first aspect and any possible implementation thereof.

[0060] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application.

[0061] In an embodiment of the present application, both the first image and the second image include target tissue. After acquiring the first image and the second image, the registration device segments the target tissue in the first image to obtain a first segmentation result, and segments the target tissue in the second image to obtain a second segmentation result. Because the registration model is capable of determining registration parameters for the target tissues in the two images, the registration device inputs the first segmentation result and the second segmentation result into the registration model to obtain first registration parameters, wherein the first registration parameters are used to register the target tissue in the first image and the target tissue in the second image. Because the target tissue in the first image can be determined based on the first segmentation result, and the target tissue in the second image can be determined based on the second segmentation result, after the first segmentation result and the second segmentation result are input into the registration model, the registration model can determine the target tissue in the first image based on the first segmentation result, and determine the target tissue in the second image based on the second segmentation result, and then determine the first registration parameter based on the target tissue in the first image and the target tissue in the second image. This can reduce the interference of the image content other than the target tissue in the first image and the image content other than the target tissue in the second image on the determination of the first registration parameter, thereby improving the accuracy of the first registration parameter. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the background technology, the drawings required for use in the embodiments of the present application or the background technology will be described below.

[0063] The drawings herein are incorporated into and constitute a part of the specification. These drawings illustrate embodiments consistent with the present application and, together with the specification, are used to illustrate the technical solutions of the present application.

[0064] Figure 1 A schematic diagram of a flow chart of a tissue image registration method provided in an embodiment of the present application;

[0065] Figure 2 A schematic diagram of a first segmentation result provided in an embodiment of the present application;

[0066] Figure 3 A schematic diagram of a second segmentation result provided in an embodiment of the present application;

[0067] Figure 4 A schematic diagram of obtaining a first registration parameter based on a registration model provided in an embodiment of the present application;

[0068] Figure 5 A schematic diagram of a transformed segmentation result provided in an embodiment of the present application;

[0069] Figure 6a A schematic diagram of a first image provided in an embodiment of the present application;

[0070] Figure 6b A schematic diagram of a second image provided in an embodiment of the present application;

[0071] Figure 6c A schematic diagram of a third image provided in an embodiment of the present application;

[0072] Figure 7 A schematic structural diagram of a tissue image registration device provided in an embodiment of the present application;

[0073] Figure 8 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0074] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0075] The terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish between different objects, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0076] References to "embodiments" herein mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, that the embodiments described herein may be combined with other embodiments. It should be understood that, in this application, "at least one (item)" means one or more, "a plurality" means two or more, and "at least two (items)" means two or three or more.

[0077] The embodiments of this application are implemented by a tissue image registration device (hereinafter referred to as the registration device). The registration device can be any electronic device capable of executing the technical solutions disclosed in the method embodiments of this application. Optionally, the registration device can be any of the following: a mobile phone, a computer, a tablet computer, or a wearable smart device.

[0078] It should be understood that the method embodiment of the present application can also be implemented by a processor executing computer program code. The following describes the embodiment of the present application in conjunction with the drawings in the embodiment of the present application. Figure 1 , Figure 1 A flowchart of a tissue image registration method provided in an embodiment of the present application.

[0079] 101. Acquire a first image and a second image, wherein both the first image and the second image include target tissue.

[0080] In the embodiment of the present application, the first image and the second image are both medical images, and the type of the first image and the type of the second image can be one of the following: an ultrasound image, a computed tomography (CT) image, or a radiograph.

[0081] In the embodiment of the present application, the target tissue can be any tissue in the target subject's body, for example, the target tissue is a blood vessel, another example, the target tissue is an organ, and another example, the target tissue is cartilage. Optionally, the target tissue is a kidney.

[0082] Both the first image and the second image include the target tissue, that is, both the first image and the second image include image content corresponding to the target tissue. The position of the target tissue in the first image differs from the position of the target tissue in the second image. In one possible implementation, the first image and the second image are acquired at different times. For example, at a first moment, the target tissue of the target object is scanned using an ultrasound probe to acquire the first image, and at a second moment, the target tissue of the target object is scanned using an ultrasound probe to acquire the second image, wherein the first moment and the second moment are different times.

[0083] In another possible implementation, the first image and the second image are acquired at the same time by different image acquisition devices. For example, the first image is acquired by scanning the target tissue of the target object at a first time by a first ultrasound probe, and the second image is acquired by scanning the target tissue of the target object at a first time by a second ultrasound probe.

[0084] 102. Obtain a first segmentation result by segmenting the target tissue in the first image.

[0085] The registration device segments the target tissue in the first image to determine the target tissue in the first image, thereby obtaining a first segmentation result. Based on the first segmentation result, the target tissue in the first image can be determined.

[0086] In one possible implementation, the registration device segments the target tissue in the first image and determines the position of the target tissue's outline in the first image, thereby obtaining a first segmentation result. Based on the first segmentation result, the target tissue's outline in the first image can be determined, and further, the area enclosed by the target tissue's outline can be determined as the target tissue.

[0087] In another possible implementation, the registration device segments the target tissue in the first image to determine pixels in the first image that are semantically the target tissue. A first segmentation result is obtained based on the pixels that are semantically the target tissue. Based on the first segmentation result, the pixels in the first image that are semantically the target tissue can be determined, and further, the target tissue in the first image can be determined based on the pixels that are semantically the target tissue. For example, a region consisting of pixels that are semantically the target tissue can be determined as the target tissue.

[0088] Optionally, the first segmentation result is an image, and the size of the first segmentation result is the same as the size of the first image. The first segmentation result includes the target tissue, and the target tissue in the first image can be determined based on the target tissue in the first segmentation result. Optionally, the first segmentation result includes the outline of the target tissue, for example, pixel s1 in the first segmentation result is the outline of the target tissue. If the position of pixel s2 in the first image is the same as the position of pixel s1 in the first segmentation result, then based on pixel s1, pixel s2 can be determined to be the outline of the target tissue. Optionally, the first segmentation result includes pixels that are semantically target tissue, for example, the semantics of pixel s1 in the first segmentation result is target tissue. If the position of pixel s2 in the first image is the same as the position of pixel s1 in the first segmentation result, then based on pixel s1, the semantics of pixel s2 can be determined to be target tissue.

[0089] See also Figure 2 , Figure 2A schematic diagram of a first segmentation result provided in an embodiment of the present application. The first segmentation result includes the target tissue and the area other than the target tissue, wherein Figure 2 In the first segmentation result shown, the white area is the target tissue, and the black area is the area other than the target tissue.

[0090] 103. Obtain a second segmentation result by segmenting the target tissue in the second image.

[0091] The registration device segments the target tissue in the second image to determine the target tissue in the second image, thereby obtaining a second segmentation result. Based on the second segmentation result, the target tissue in the second image can be determined.

[0092] In one possible implementation, the registration device segments the target tissue in the second image and determines the position of the target tissue's outline in the second image, thereby obtaining a second segmentation result. Based on the second segmentation result, the target tissue's outline in the second image can be determined, and further, the area enclosed by the target tissue's outline can be determined as the target tissue.

[0093] In another possible implementation, the registration device segments the target tissue in the second image to determine pixels in the second image that are semantically the target tissue. A second segmentation result is obtained based on the pixels that are semantically the target tissue. Based on the second segmentation result, the pixels in the second image that are semantically the target tissue can be determined, and further, the target tissue in the second image can be determined based on the pixels that are semantically the target tissue. For example, a region consisting of pixels that are semantically the target tissue can be determined as the target tissue.

[0094] Optionally, the second segmentation result is an image, and the size of the second segmentation result is the same as the size of the second image. The second segmentation result includes the target tissue, and the target tissue in the second image can be determined based on the target tissue in the second segmentation result. Optionally, the second segmentation result includes the outline of the target tissue, for example, pixel s3 in the second segmentation result is the outline of the target tissue. If the position of pixel s4 in the second image is the same as the position of pixel s3 in the second segmentation result, then based on pixel s3, pixel s4 can be determined to be the outline of the target tissue. Optionally, the second segmentation result includes pixels that are semantically target tissue, for example, the semantics of pixel s3 in the second segmentation result is target tissue. If the position of pixel s4 in the second image is the same as the position of pixel s3 in the second segmentation result, then based on pixel s3, the semantics of pixel s4 can be determined to be target tissue.

[0095] See also Figure 3 , Figure 3 A schematic diagram of a second segmentation result provided in an embodiment of the present application. The second segmentation result includes the target tissue and the area other than the target tissue, wherein Figure 3 In the second segmentation result shown in FIG, the white area is the target tissue, and the black area is the area other than the target tissue.

[0096] 104. Input the first segmentation result and the second segmentation result into a registration model to obtain a first registration parameter, wherein the first registration parameter is used to register the target tissue in the first image and the target tissue in the second image, and the registration model has the ability to determine the registration parameters of the target tissue in the two images.

[0097] In the embodiments of the present application, the registration model can have any structure. The registration model can be any type of model. Optionally, the registration model is one of the following: a machine learning model, a deep learning model, for example, a neural network in a deep learning model, such as a U-Net.

[0098] Optionally, the hardware resources required to run the registration model are relatively small, for example, the registration model is obtained by model compression. This reduces the amount of data processing required to obtain the first registration parameter using the registration model, thereby increasing the speed at which the registration device obtains the first registration parameter using the registration model.

[0099] In an embodiment of the present application, a registration parameter is a parameter used to register target tissues in two images, or a registration parameter is a parameter used to align target tissues in two images. Specifically, if there is a difference between the position of the target tissue in the first image and the position of the target tissue in the second image, registering the target tissue in the first image with the target tissue in the second image based on the first registration parameter can reduce the difference between the position of the target tissue in the first image and the position of the target tissue in the second image, thereby aligning the target tissue in the first image with the target tissue in the second image.

[0100] In an embodiment of the present application, the registration parameter is one of the following: a rigid transformation, a similarity transformation, an affine transformation, a projective transformation, and a nonlinear transformation. Optionally, the registration parameter includes one or more of the following: a translation parameter and a rotation parameter, wherein the translation parameter is used to translate the target tissue in the image, and the rotation parameter is used to rotate the target tissue in the image. Optionally, the translation parameter is used to translate the target tissue in the image in the pixel coordinate system of the image, and the rotation parameter is used to rotate the target tissue in the image in the pixel coordinate system of the image. When both the first image and the second image are two-dimensional images, the pixel coordinate system of the image is a two-dimensional coordinate system, and accordingly, the translation parameter is a two-dimensional translation parameter, for example, the translation parameter includes x and y, wherein x represents the amount of translation along the horizontal axis of the pixel coordinate system of the image, and y represents the amount of translation along the vertical axis of the pixel coordinate system of the image.

[0101] Optionally, the registration parameter is a vector comprising a translation parameter and a rotation parameter. For example, if both the first image and the second image are two-dimensional images, the registration parameter is the following three-dimensional vector: [x, y, r], where x and y are translation parameters and r is a rotation parameter. Specifically, x represents the amount of translation along the horizontal axis of the image's pixel coordinate system, y represents the amount of translation along the vertical axis of the image's pixel coordinate system, and r represents the rotation angle of the target tissue in the image's pixel coordinate system.

[0102] In an embodiment of the present application, the registration model has the ability to determine the registration parameters of the target tissues in the two images, that is, after the two images including the target tissues are input into the registration model, the registration model can determine the registration parameters between the target tissues in the two images by processing the two input images, and the registration parameters are used to align the target tissues in the two images.

[0103] Optionally, the registration model acquires the ability to determine registration parameters of target tissues in two images through training. The training process includes: obtaining a first training image and a second training image, wherein both the first training image and the second training image include target tissues. Segmenting the target tissue in the first training image to obtain a first training segmentation result. Segmenting the target tissue in the second training image to obtain a second training segmentation result. Inputting the first training segmentation result and the second training segmentation result into the model to be trained, obtaining training registration parameters for the target tissues in the first training image and the target tissues in the second training image output by the model to be trained, wherein the training registration parameters are used to align the target tissues in the first training image and the target tissues in the second training image. Transforming the target tissue in the first training image based on the training registration parameters to obtain a third training image. Based on the difference between the target tissue in the second training image and the target tissue in the third training image, obtaining a loss for the model to be trained, wherein the loss is negatively correlated with the difference. Based on the loss of the model to be trained, updating the parameters of the model to be trained until the loss of the model to be trained converges, thereby obtaining a registration model.

[0104] Optionally, the second training image is obtained by transforming the target tissue in the first training image. For example, after acquiring the first training image, the target tissue in the first training image is transformed based on training transformation parameters to obtain the second training image, where the training transformation parameters include one or more of translation parameters and rotation parameters. Optionally, the training transformation parameters are randomly generated.

[0105] Optionally, obtaining the loss of the model to be trained based on the difference between the target tissue in the second training image and the target tissue in the third training image includes: determining the loss of the model to be trained based on the target tissue in the second training image, the target tissue in the third training image and the loss function.

[0106] Optionally, the loss function is one of the following: mean squared error (MSE) or Frobenius norm.

[0107] In one possible implementation, the registration device splices the first segmentation result and the second segmentation result to obtain a spliced ​​segmentation result. The spliced ​​segmentation result is then input into the registration model to obtain the first registration parameter. Figure 4 , Figure 4 This is a schematic diagram of obtaining a first registration parameter based on a registration model provided in an embodiment of the present application. Figure 4 As shown, the input of the registration model includes the segmentation results after splicing, wherein the segmentation results after splicing are obtained by Figure 2 The first segmentation result and Figure 3 After the spliced ​​segmentation result is input into the registration model, the first registration parameter can be obtained.

[0108] Optionally, after obtaining the first registration parameter, the registration device may transform the target tissue in the first segmentation result based on the first registration parameter to obtain a transformed segmentation result.

[0109] Optionally, when the first registration parameter is a vector including a translation parameter and a rotation parameter, the registration device first determines a two-dimensional rigid transformation matrix based on the first registration parameter, and transforms the target tissue in the first segmentation result based on the two-dimensional rigid transformation matrix to obtain a transformed segmentation result.

[0110] See also Figure 5 , Figure 5 The transformed segmentation result is a schematic diagram of a transformed segmentation result provided in an embodiment of the present application. The transformed segmentation result includes the target tissue and the area other than the target tissue, wherein Figure 5 In the transformed segmentation result, the white area is the target tissue, and the black area is the area other than the target tissue. Figure 5 The transformed segmentation result shown is based on Figure 4 The first registration parameter pair in Figure 2 By comparing Figure 2 、 Figure 3 、 Figure 5 It can be seen that the target organization is Figure 2 The location and target organization in Figure 3 The difference in position in the target tissue Figure 5 The location and target organization in Figure 3 That is, the target tissue in the first segmentation result is transformed based on the first registration parameter, so that the target tissue in the first segmentation result can be aligned with the target tissue in the second segmentation result.

[0111] In an embodiment of the present application, both the first image and the second image include target tissue. After acquiring the first image and the second image, the registration device segments the target tissue in the first image to obtain a first segmentation result, and segments the target tissue in the second image to obtain a second segmentation result. Because the registration model is capable of determining registration parameters for the target tissues in the two images, the registration device inputs the first segmentation result and the second segmentation result into the registration model to obtain first registration parameters, wherein the first registration parameters are used to register the target tissue in the first image and the target tissue in the second image. Because the target tissue in the first image can be determined based on the first segmentation result, and the target tissue in the second image can be determined based on the second segmentation result, after the first segmentation result and the second segmentation result are input into the registration model, the registration model can determine the target tissue in the first image based on the first segmentation result, and determine the target tissue in the second image based on the second segmentation result, and then determine the first registration parameter based on the target tissue in the first image and the target tissue in the second image. This can reduce the interference of the image content other than the target tissue in the first image and the image content other than the target tissue in the second image on the determination of the first registration parameter, thereby improving the accuracy of the first registration parameter.

[0112] As an optional implementation, the registration device further performs the following steps:

[0113] 201. Determine n first regions from the target tissue in the first segmentation result.

[0114] 202. Determine n second regions from the target tissue in the second segmentation result.

[0115] In the embodiment of the present application, the first region is a subregion of the target tissue in the first segmentation result. Optionally, the target tissue in the first segmentation result is composed of n first regions. The second region is a subregion of the target tissue in the second segmentation result. Optionally, the target tissue in the second segmentation result is composed of n second regions.

[0116] In an embodiment of the present application, the distribution of the n first regions in the target tissue in the first segmentation result is the same as the distribution of the n second regions in the target tissue in the second segmentation result. Optionally, the registration device divides the target tissue in the first segmentation result into the n first regions. The registration device divides the target tissue in the second segmentation result into the n second regions. Optionally, the registration device divides the target tissue in the first segmentation result and the target tissue in the second segmentation result respectively according to the same division method.

[0117] In one possible implementation, the target tissue belongs to a target object. For example, if the target tissue is the kidney of object O1, then object O1 is the target object. The target object's respiration causes the target tissue to move, resulting in a position of the target tissue in the first image that is different from a position of the target tissue in the second image. In other words, the difference between the position of the target tissue in the first image and the position of the target tissue in the second image is due to the target object's respiration. For example, because respiration causes the target tissue to move, resulting in different positions of the target tissue at different times, if the first image and the second image were acquired at different times, the position of the target tissue in the first image is different from the position of the target tissue in the second image.

[0118] Furthermore, since the target subject breathes through their lungs, the closer a part of the target tissue is to the lungs, the more affected it is by breathing, the greater the amplitude of its movement due to breathing, and accordingly, the greater the difference between its position in the first image and its position in the second image. Therefore, to more accurately determine the registration parameters for the regions corresponding to each part in the image, the target tissue can be divided into n regions based on their distance from the target subject's lungs.

[0119] Optionally, the registration device determines n first regions based on the target tissue in the first segmentation result, wherein the n first regions include a third region, the third region is any region among the n first regions, and the farther the third region is from the lung of the target subject, the larger the area of ​​the third region. The registration device determines n second regions based on the target tissue in the second segmentation result, wherein the n second regions include a fourth region, wherein the fourth region is any region among the n second regions, and the farther the fourth region is from the lung of the target subject, the larger the area of ​​the fourth region.

[0120] Since the farther the target tissue portion is from the lungs of the target subject, the less affected that portion is by respiration, and the smaller the amplitude of movement of that portion due to respiration, the farther the target tissue portion is from the lungs of the target subject, the smaller the difference in the amplitude of movement of different regions within that portion due to respiration. Conversely, the closer the target tissue portion is to the lungs of the target subject, the greater the difference in the amplitude of movement of different regions within that portion due to respiration. Therefore, the registration device determines n first regions from the target tissue in the first segmentation result, and n second regions from the target tissue in the second segmentation result, based on the principle that the farther the portion is from the lungs of the target subject, the larger the area of ​​the region determined from the target tissue. In this way, by registering the n first regions and the n second regions to determine the registration parameters of each portion in the target tissue, the accuracy of the registration parameters of each portion in the target tissue can be improved, the number of first regions and the number of second regions can be reduced, and the amount of data processing required to determine the registration parameters of each portion in the target tissue can be reduced.

[0121] 203. Obtain n second registration parameters by registering the n first regions and the n second regions, wherein the second registration parameters are used to register corresponding regions among the n first regions and the n second regions.

[0122] In an embodiment of the present application, for a region in the n first regions, its position in the first segmentation result is referred to as the first segmentation position. For a region in the n second regions, its position in the second segmentation result is referred to as the second segmentation position. If the first segmentation position is the same as the second segmentation position, then the region in the n first regions and the region in the n second regions are corresponding regions in the n first regions and the n second regions. For example, the n first regions include the first region q1 and the first region q2, and the n second regions include the second region q3 and the second region q4. If the position of the first region q1 in the first segmentation result is the same as the position of the second region q3 in the second segmentation result, then the first region q1 and the second region q3 are corresponding regions in the n first regions and the n second regions.

[0123] In step 203, the registration device aligns the corresponding areas in the n first areas and the n second areas to obtain second registration parameters for the corresponding areas in the n first areas and the n second areas. Since for any area in the n first areas, there is a corresponding area in the n second areas, there are n groups of corresponding areas in the n first areas and the n second areas, where a group of corresponding areas includes one first area and one second area corresponding to the first area. The registration device aligns each group of corresponding areas in the n groups of corresponding areas to obtain n second registration parameters, where the n second registration parameters are used to align the first area and the second area in the n groups of corresponding areas, respectively, i.e., the second registration parameters correspond one-to-one with the first areas, or the second registration parameters correspond one-to-one with the second areas.

[0124] Optionally, the target tissue belongs to the target subject, and the target subject's respiration causes the target tissue to move, thereby causing the position of the target tissue in the first image to be different from the position of the target tissue in the second image. In other words, the difference between the position of the target tissue in the first image and the position of the target tissue in the second image is due to the target subject's respiration. For example, because respiration causes the target tissue to move, thereby causing the position of the target tissue to be different at different times, if the first image and the second image are acquired at different times, the position of the target tissue in the first image is different from the position of the target tissue in the second image.

[0125] Because different parts of the target tissue are affected differently by respiration, the amplitude of movement caused by respiration varies across different parts of the target tissue. Furthermore, because different regions within the n first regions correspond to different parts of the target tissue, the positional differences caused by respiration also vary across the n first regions. Therefore, among the n second registration parameters corresponding to the n first regions, different second registration parameters are different.

[0126] In step 203, the registration device may register the n first regions in the first segmentation result and the n second regions in the second segmentation result using any registration method to obtain n second registration parameters. In one possible implementation, the image registration device inputs corresponding regions of the n first regions and the n second regions into a registration model to obtain a second registration parameter. In another possible implementation, the registration device registers the n first regions in the first segmentation result and the n second regions in the second segmentation result based on the scale-invariant feature transform (SIFT) to obtain n second registration parameters.

[0127] 204. Determine a third registration parameter based on the first registration parameter and the n second registration parameters, wherein the third registration parameter is used to register the target tissue in the first image and the target tissue in the second image.

[0128] As described above, the first region is a subregion of the target tissue in the first segmentation result, and the second region is a subregion of the target tissue in the second segmentation result. Both the first region and the second region include the target tissue, so the second registration parameter obtained in step 203 is a local registration parameter of the target tissue. The first registration parameter is obtained by the registration model based on the first segmentation result and the second segmentation result, that is, the first registration parameter is a global registration parameter of the target tissue. Therefore, the registration device determines a third registration parameter based on the first registration parameter and the n second registration parameters, wherein the third registration parameter is used to register the target tissue in the first image with the target tissue in the second image. This can achieve the determination of the third registration parameter using the local registration parameter and the global registration parameter, thereby improving the accuracy of the third registration parameter.

[0129] Optionally, the target tissue is part of the target object, and the target object's respiration can cause the target tissue to move. As previously described, different parts of the target tissue experience different motion amplitudes due to respiration. Therefore, determining the third registration parameter for the target tissue in the first image and the target tissue in the second image based on the global registration parameter can easily overlook the differences in motion amplitudes between different parts of the target tissue, resulting in low accuracy of the third registration parameter. Therefore, the registration device determines the third registration parameter based on the first registration parameter and n second registration parameters, thereby improving the accuracy of the third registration parameter.

[0130] As an optional implementation, the registration device implements “determining the third registration parameter based on the first registration parameter and n second registration parameters” by performing the following steps:

[0131] 301. Based on the first registration parameter, determine m fourth registration parameters and t fifth registration parameters from the n second registration parameters, where m and t are both less than n, the sum of m and t is n, the difference between the fourth registration parameter and the first registration parameter is greater than or equal to a threshold, and the difference between the fifth registration parameter and the first registration parameter is less than a threshold.

[0132] In the embodiments of the present application, the threshold value is used to determine whether the difference between two registration parameters is large or small. Specifically, if the difference between the two registration parameters is greater than or equal to the threshold value, it indicates that the difference between the two registration parameters is large, and if the difference between the two registration parameters is less than the threshold value, it indicates that the difference between the two registration parameters is small. Therefore, if the difference between the fourth registration parameter and the first registration parameter is greater than or equal to the threshold value, it indicates that the difference between the fourth registration parameter and the first registration parameter is large, and if the difference between the fifth registration parameter and the first registration parameter is less than the threshold value, it indicates that the difference between the fifth registration parameter and the first registration parameter is small.

[0133] That is, the registration apparatus determines, by executing step 301 , m fourth registration parameters having large differences from the first registration parameters and t fifth registration parameters having small differences from the first registration parameters from the n second registration parameters.

[0134] Optionally, the target tissue belongs to the target object, and the breathing of the target object will cause the target tissue to move. As mentioned above, the movement amplitudes of different parts of the target tissue due to breathing are different. Specifically, the movement amplitudes of some parts of the target tissue due to breathing are large, which leads to a large difference between the movement amplitude of this part and the overall movement amplitude of the target tissue. The movement amplitudes of other parts of the target tissue due to breathing are small, and accordingly, the difference between the movement amplitude of this part and the overall movement amplitude of the target tissue is small. For example, the movement amplitude of the part of the target tissue close to the lung is large, and accordingly, the difference between the movement amplitude of the part close to the lung and the overall movement amplitude of the target tissue is large. The movement amplitude of the part of the target tissue far from the lung is small, and accordingly, the difference between the movement amplitude of the part far from the lung and the overall movement amplitude of the target tissue is small.

[0135] For ease of description, the target tissue portion with a large amplitude of motion due to respiration will be referred to below as the large-amplitude portion, and the target tissue portion with a small amplitude of motion due to respiration will be referred to below as the small-amplitude portion. Since the first registration parameter is a global registration parameter for the target tissue, the first registration parameter represents the overall amplitude of motion of the target tissue. Furthermore, since the second registration parameter is a local registration parameter for the target tissue, the second registration parameter represents the local amplitude of motion of the target tissue. Therefore, the second registration parameter corresponding to the large-amplitude portion differs significantly from the first registration parameter, while the second registration parameter corresponding to the small-amplitude portion differs slightly from the first registration parameter. At this point, the m fourth registration parameters determined by executing step 301 are the registration parameters corresponding to the large-amplitude portion, and the t fifth registration parameters determined by executing step 301 are the registration parameters corresponding to the small-amplitude portion. The m first regions and m second regions corresponding to the m fourth registration parameters are all regions corresponding to the large-amplitude portion. The t first regions and t second regions corresponding to the t fifth registration parameters are all regions corresponding to the small-amplitude portion.

[0136] 302. Determine a sixth registration parameter based on the t fifth registration parameters, where the sixth registration parameter is used to register regions corresponding to the t fifth registration parameters.

[0137] Because when the target tissue moves under the influence of breathing, the movements of different parts are related. For example, the difference in the movement amplitude between two adjacent parts is small, the translation directions of different parts are consistent, and the rotation directions of different parts are consistent. Therefore, for any part in the target tissue, the alignment parameters of other parts can be used to determine the alignment parameters of any part. For example, the target tissue includes part b1 and part b2, and the alignment parameters of part b2 can be used to determine the alignment parameters of part b1.

[0138] Considering that if the difference in the motion amplitudes of two parts is large, determining the registration parameters of one part based on the registration parameters of the other part can easily lead to large errors in the registration parameters of the other part. For example, if the target tissue includes parts b1 and b2, and the motion amplitudes of parts b1 and b2 differ significantly, determining the registration parameters of part b1 based on the registration parameters of part b2 can easily lead to large errors in the registration parameters of part b1. Therefore, when the difference in the motion amplitudes of the two parts is small, determining the registration parameters of one part based on the registration parameters of the other part can improve the accuracy of the registration parameters of the other part.

[0139] As described in step 301, the t fifth registration parameters all differ slightly from the first registration parameter. Therefore, the difference in the motion amplitude of the parts corresponding to the t fifth registration parameters is relatively small. Therefore, the registration device determines a sixth registration parameter based on the t fifth registration parameters. The sixth registration parameter is used to register the areas corresponding to the t fifth registration parameters, thereby improving the accuracy of the sixth registration parameter.

[0140] Optionally, the target tissue belongs to the target subject, and the target subject's breathing can cause the target tissue to move. The t fifth registration parameters are registration parameters corresponding to small-amplitude areas. Because the difference in the amplitude of movement of small-amplitude areas is small, the registration device determines, based on the t fifth registration parameters, sixth registration parameters for the t first regions corresponding to the t fifth registration parameters and the t second regions corresponding to the t fifth registration parameters, thereby improving the accuracy of the sixth registration parameters. It should be understood that the registration parameters of any of the t first regions are the sixth registration parameters.

[0141] 303. Determine a third registration parameter based on the first registration parameter, m fourth registration parameters, and the sixth registration parameter.

[0142] In one possible implementation, the registration device determines, based on the first registration parameter and the sixth registration parameter, registration parameters for t first regions corresponding to the t fifth registration parameters, where the registration parameters for the t first regions are used to register the t first regions with the t second regions. A third registration parameter is determined based on the registration parameters for the t first regions and the m fourth registration parameters.

[0143] Optionally, the registration device obtains the registration parameters of the t first regions by performing weighted averaging on the first registration parameter and the sixth registration parameter.

[0144] In another possible implementation, the registration device determines a seventh registration parameter based on the m fourth registration parameters and the sixth registration parameter, wherein the seventh registration parameter is used to register the target tissue in the first image with the target tissue in the second image. A third registration parameter is determined based on the first registration parameter and the seventh registration parameter.

[0145] In this implementation, because the m first regions corresponding to the m fourth registration parameters correspond to regions with large motion amplitudes, the difference in motion amplitude between any two of the m first regions is significant. In this case, determining the registration parameters of another of the m first regions based on the registration parameters of one of the m first regions can easily lead to significant errors in the registration parameters of the other first region. Therefore, the registration device determines the seventh registration parameter based on the m fourth registration parameters of the m first regions and the sixth registration parameter, thereby improving the accuracy of the seventh registration parameter.

[0146] Optionally, the registration device determines the seventh registration parameter by fitting the m fourth registration parameters and the sixth registration parameter, thereby fitting the local registration parameters into the global registration parameters, thereby making the seventh registration parameter smoother.

[0147] Since the m fourth registration parameters and the sixth registration parameters are both registration parameters obtained by local registration of the target tissue, the seventh registration parameter is the global registration parameter of the target tissue. Therefore, by determining the seventh registration parameter based on the m fourth registration parameters and the sixth registration parameter, the global registration parameter of the target tissue can be determined based on the local information of the target tissue.

[0148] Because the first registration parameter is determined based on the first and second segmentation results, it is a global registration parameter determined based on global information about the target tissue. Therefore, the registration device determines the third registration parameter based on the first and seventh registration parameters. This allows the global registration parameter of the target tissue to be determined based on both local and global information about the target tissue, thereby improving the accuracy of the third registration parameter.

[0149] Optionally, the registration device obtains the third registration parameter by performing weighted averaging on the first registration parameter and the seventh registration parameter.

[0150] In another possible implementation, the registration device corrects m fourth registration parameters based on the first registration parameters to obtain m corrected fourth registration parameters. Based on the first registration parameters, the sixth registration parameters are corrected to obtain the corrected sixth registration parameters. Based on the m corrected fourth registration parameters and the corrected sixth registration parameters, the third registration parameters are determined, wherein the third registration parameters include m corrected fourth registration parameters and the corrected sixth registration parameters. At this time, based on the m corrected fourth registration parameters in the third registration parameters, the m first areas in the first image and the m second areas in the second image can be registered respectively. Based on the corrected sixth registration parameters in the third registration parameters, the t first areas in the first image and the t second areas in the second image can be registered. In this way, the registration of the target tissue in the first image and the target tissue in the second image can be achieved.

[0151] As an optional implementation, both the first image and the second image are acquired during the breathing process of the target object, and the acquisition time of the first image is different from the acquisition time of the second image.

[0152] In one possible implementation scenario, the relative positional relationship between the image acquisition device and the target object is the target relative positional relationship. For example, the target object lies on a surgical table, the ultrasound probe is in a fixed position, and the relative positional relationship between the ultrasound probe and the target object is also fixed. The first image and the second image are acquired by the image acquisition device at different times.

[0153] In this implementation, after determining the third registration parameter, the registration device further performs the following steps: first, based on the third registration parameter, transforms the target tissue in the first image to obtain a third image. Then, based on the second and third images, the lesion in the target tissue is determined, thereby improving the accuracy of the determined lesion.

[0154] Optionally, when the third registration parameters include m corrected fourth registration parameters and a corrected sixth registration parameter, the registration device transforms the m first regions in the first image based on the m corrected fourth registration parameters in the third registration parameters, respectively, to obtain m first transformed regions. Based on the corrected sixth registration parameters in the third registration parameters, the registration device transforms the t first regions in the first image to obtain second transformed regions. Based on the m first transformed regions and the second transformed regions, a transformed target tissue is determined. The target tissue in the first image is replaced with the transformed target tissue to obtain a third image.

[0155] In a possible implementation scenario, the first image and the second image are both ultrasound images including the target tissue, and the third image obtained by transforming the target tissue in the first image is also an ultrasound image including the target tissue. Figure 6a、 Figure 6b 、 Figure 6c ,in, Figure 6a A schematic diagram of a first image provided in an embodiment of the present application, Figure 6b A schematic diagram of a second image provided in an embodiment of the present application, Figure 6c A schematic diagram of a third image provided in an embodiment of the present application.

[0156] Based on Figure 6a The first image shown and Figure 6b After determining the third registration parameter for the second image shown, based on the third registration parameter, Figure 6a By transforming the target tissue in Figure 6c The third image shown. Figure 6a The first image shown and Figure 6c As can be seen from the third image, the position of the target tissue in the first image is different from the position of the target tissue in the third image. Figure 6a The first image shown, Figure 6b The second image shown and Figure 6c It can be seen from the third image shown that the difference between the position of the target tissue in the first image and the position of the target tissue in the second image is greater than the difference between the position of the target tissue in the third image and the position of the target tissue in the second image. That is to say, by transforming the target tissue in the first image based on the third alignment parameter, the target tissue in the first image can be aligned with the target tissue in the second image.

[0157] Those skilled in the art will understand that in the above-mentioned method of the specific implementation method, the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0158] If the technical solution of this application involves personal information, the product that applies the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing personal information. If the technical solution of this application involves sensitive personal information, the product that applies the technical solution of this application has obtained the individual's separate consent before processing sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, a clear and prominent sign is set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that they agree to the collection of their personal information; or on the personal information processing device, when the personal information processing rules are notified by obvious signs / information, the individual's authorization is obtained through pop-up information or by asking the individual to upload their personal information; among which, personal information processing may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

[0159] The above describes in detail the method of the embodiment of the present application, and the following provides an apparatus of the embodiment of the present application.

[0160] See also Figure 7 , Figure 7 This is a structural diagram of a tissue image registration device provided in an embodiment of the present application. The tissue image registration device 1 includes: an acquisition unit 11, a segmentation unit 12, and a processing unit 13. Optionally, the tissue image registration device 1 also includes: a determination unit 14. Specifically:

[0161] An acquisition unit 11 is configured to acquire a first image and a second image, wherein both the first image and the second image include target tissue;

[0162] a segmentation unit 12, configured to obtain a first segmentation result by segmenting the target tissue in the first image;

[0163] The segmentation unit 12 is further configured to obtain a second segmentation result by segmenting the target tissue in the second image;

[0164] The processing unit 13 is used to input the first segmentation result and the second segmentation result into a registration model to obtain a first registration parameter, wherein the first registration parameter is used to align the target tissue in the first image and the target tissue in the second image, and the registration model has the ability to determine the registration parameters of the target tissue in the two images.

[0165] In combination with any embodiment of the present application, the tissue image registration device further includes:

[0166] a determining unit 14, configured to determine n first regions from the target tissue in the first segmentation result;

[0167] The determining unit 14 is further configured to determine n second regions from the target tissue in the second segmentation result;

[0168] The processing unit 13 is further configured to obtain n second registration parameters by registering the n first regions with the n second regions, wherein the second registration parameters are used to register corresponding regions among the n first regions and the n second regions;

[0169] The determining unit 14 is further configured to determine a third registration parameter based on the first registration parameter and the n second registration parameters, wherein the third registration parameter is used to register the target tissue in the first image and the target tissue in the second image.

[0170] In combination with any embodiment of the present application, the determining unit 14 is further configured to:

[0171] determining, based on the first registration parameter, m fourth registration parameters and t fifth registration parameters from the n second registration parameters, where m and t are both less than n, the sum of m and t is n, a difference between the fourth registration parameter and the first registration parameter is greater than or equal to a threshold, and a difference between the fifth registration parameter and the first registration parameter is less than the threshold;

[0172] Determining a sixth registration parameter based on the t fifth registration parameters, wherein the sixth registration parameter is used to register the areas corresponding to the t fifth registration parameters;

[0173] The third registration parameter is determined based on the first registration parameter, the m fourth registration parameters, and the sixth registration parameter.

[0174] In combination with any embodiment of the present application, the determining unit 14 is further configured to:

[0175] determining a seventh registration parameter based on the m fourth registration parameters and the sixth registration parameter, wherein the seventh registration parameter is used to register the target tissue in the first image and the target tissue in the second image;

[0176] The third registration parameter is determined based on the first registration parameter and the seventh registration parameter.

[0177] In conjunction with any embodiment of the present application, the target tissue belongs to the target object;

[0178] The determining unit 14 is further configured to:

[0179] determining, based on the target tissue in the first segmentation result, n first regions, where the n first regions include a third region, and the third region is any region among the n first regions, and the farther the third region is from the lung of the target subject, the larger the area of ​​the third region;

[0180] Based on the target tissue in the second segmentation result, n second regions are determined, where the n second regions include a fourth region, and the fourth region is any region among the n second regions. The farther the fourth region is from the lung of the target object, the larger the area of ​​the fourth region.

[0181] In conjunction with any embodiment of the present application, the first image and the second image are both acquired during the breathing process of the target object, and the acquisition time of the first image is different from the acquisition time of the second image;

[0182] The processing unit 13 is further configured to transform the target tissue in the first image based on the third registration parameter to obtain a third image;

[0183] The tissue image registration device further includes: a determination unit 14, configured to determine a lesion in the target tissue based on the second image and the third image.

[0184] In an embodiment of the present application, both the first image and the second image include target tissue. After acquiring the first image and the second image, the registration device segments the target tissue in the first image to obtain a first segmentation result, and segments the target tissue in the second image to obtain a second segmentation result. Because the registration model is capable of determining registration parameters for the target tissues in the two images, the registration device inputs the first segmentation result and the second segmentation result into the registration model to obtain first registration parameters, wherein the first registration parameters are used to register the target tissue in the first image and the target tissue in the second image. Because the target tissue in the first image can be determined based on the first segmentation result, and the target tissue in the second image can be determined based on the second segmentation result, after the first segmentation result and the second segmentation result are input into the registration model, the registration model can determine the target tissue in the first image based on the first segmentation result, and determine the target tissue in the second image based on the second segmentation result, and then determine the first registration parameter based on the target tissue in the first image and the target tissue in the second image. This can reduce the interference of the image content other than the target tissue in the first image and the image content other than the target tissue in the second image on the determination of the first registration parameter, thereby improving the accuracy of the first registration parameter.

[0185] In some embodiments, the functions or modules included in the device provided in the embodiments of the present application can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments. For the sake of brevity, it will not be repeated here.

[0186] Figure 8 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. The electronic device 2 includes a processor 21 and a memory 22. Optionally, the electronic device 2 also includes an input device 23 and an output device 24. The processor 21, the memory 22, the input device 23 and the output device 24 are coupled via a connector, and the connector includes various interfaces, transmission lines or buses, etc., which are not limited in the embodiments of the present application. It should be understood that in each embodiment of the present application, coupling refers to mutual connection in a specific manner, including direct connection or indirect connection through other devices, for example, connection through various interfaces, transmission lines, buses, etc.

[0187] The processor 21 may be one or more graphics processing units (GPUs). If the processor 21 is a GPU, the GPU may be a single-core GPU or a multi-core GPU. Alternatively, the processor 21 may be a processor group consisting of multiple GPUs, with the multiple processors coupled to each other via one or more buses. Alternatively, the processor may be another type of processor, and the present embodiment is not limiting.

[0188] The memory 22 can be used to store computer program instructions and various computer program codes, including program codes for executing the solution of the present application. Optionally, the memory includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or compact disc read-only memory (CD-ROM), and is used for related instructions and data.

[0189] The input device 23 is used to input data and / or signals, and the output device 24 is used to output data and / or signals. The input device 23 and the output device 24 can be independent devices or an integrated device.

[0190] It can be understood that in the embodiment of the present application, the memory 22 can be used not only to store relevant instructions, but also to store relevant data. The embodiment of the present application does not limit the specific data stored in the memory.

[0191] It is understandable that Figure 8 Only a simplified design of an electronic device is shown. In actual applications, the electronic device may further include other necessary components, including but not limited to any number of input / output devices, processors, memories, etc., and all electronic devices that can implement the embodiments of the present application are within the scope of protection of the present application.

[0192] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0193] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here. Those skilled in the art will also clearly understand that the descriptions of the various embodiments of this application have different focuses. For the convenience and brevity of description, the same or similar parts may not be repeated in different embodiments. Therefore, for parts not described or not described in detail in a certain embodiment, reference can be made to the descriptions of other embodiments.

[0194] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0195] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0196] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0197] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted via the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0198] Those skilled in the art will appreciate that all or part of the process steps in the above-described method embodiments can be implemented by a computer program instructing the relevant hardware. The program can be stored in a computer-readable storage medium. When executed, the program can include the process steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A tissue image registration method, characterized in that: The method comprises: Acquiring a first image and a second image, wherein both the first image and the second image include target tissue; Obtaining a first segmentation result by segmenting the target tissue in the first image; obtaining a second segmentation result by segmenting the target tissue in the second image; Inputting the first segmentation result and the second segmentation result into a registration model to obtain first registration parameters, wherein the first registration parameters are used to register the target tissue in the first image and the target tissue in the second image, wherein the registration model is capable of determining the registration parameters of the target tissue in the two images; determining n first regions from the target tissue in the first segmentation result; determining n second regions from the target tissue in the second segmentation result; Obtaining n second registration parameters by registering the n first regions with the n second regions, wherein the second registration parameters are used to register corresponding regions among the n first regions and the n second regions; determining a third registration parameter based on the first registration parameter and the n second registration parameters, wherein the third registration parameter is used to register the target tissue in the first image and the target tissue in the second image; The determining a third registration parameter based on the first registration parameter and the n second registration parameters includes: determining, based on the first registration parameter, m fourth registration parameters and t fifth registration parameters from the n second registration parameters, where m and t are both less than n, the sum of m and t is n, a difference between the fourth registration parameter and the first registration parameter is greater than or equal to a threshold, and a difference between the fifth registration parameter and the first registration parameter is less than the threshold; Determining a sixth registration parameter based on the t fifth registration parameters, wherein the sixth registration parameter is used to register the areas corresponding to the t fifth registration parameters; The third registration parameter is determined based on the first registration parameter, the m fourth registration parameters, and the sixth registration parameter.

2. The tissue image registration method according to claim 1, characterized in that: The determining the third registration parameter based on the first registration parameter, the m fourth registration parameters and the sixth registration parameter includes: determining a seventh registration parameter based on the m fourth registration parameters and the sixth registration parameter, wherein the seventh registration parameter is used to register the target tissue in the first image and the target tissue in the second image; The third registration parameter is determined based on the first registration parameter and the seventh registration parameter.

3. The tissue image registration method according to claim 1 or 2, characterized in that: The target tissue belongs to the target object, and the target tissue is one of the following: a kidney in the target object, a cartilage in the target object, or a blood vessel in the target object; The determining n first regions from the target tissue in the first segmentation result includes: Determining n first regions based on the target tissue in the first segmentation result, where the n first regions include a third region, where the third region is any one of the n first regions, and the farther the third region is from the lung in the target object, the larger the area of ​​the third region; The determining n second regions from the target tissue in the second segmentation result includes: Based on the target tissue in the second segmentation result, n second regions are determined, where the n second regions include a fourth region, and the fourth region is any region among the n second regions. The farther the fourth region is from the lung in the target object, the larger the area of ​​the fourth region.

4. The tissue image registration method according to claim 1 or 2, characterized in that: The first image and the second image are both acquired during the breathing process of the target object, and the acquisition time of the first image is different from the acquisition time of the second image; After determining the third registration parameter, the method further includes: transforming the target tissue in the first image based on the third registration parameter to obtain a third image; A lesion in the target tissue is determined based on the second image and the third image.

5. A tissue image registration device, characterized in that: The tissue image registration device comprises: an acquisition unit, configured to acquire a first image and a second image, wherein both the first image and the second image include target tissue; a segmentation unit, configured to obtain a first segmentation result by segmenting the target tissue in the first image; The segmentation unit is further configured to obtain a second segmentation result by segmenting the target tissue in the second image; a processing unit, configured to input the first segmentation result and the second segmentation result into a registration model to obtain first registration parameters, wherein the first registration parameters are used to register the target tissue in the first image and the target tissue in the second image, wherein the registration model is capable of determining the registration parameters of the target tissue in the two images; a determining unit, configured to determine n first regions from the target tissue in the first segmentation result; The determining unit is further configured to determine n second regions from the target tissue in the second segmentation result; The determining unit is further configured to obtain n second registration parameters by registering the n first regions with the n second regions, wherein the second registration parameters are used to register corresponding regions among the n first regions and the n second regions; The processing unit is further configured to determine a third registration parameter based on the first registration parameter and the n second registration parameters, wherein the third registration parameter is used to register the target tissue in the first image with the target tissue in the second image; The determining of the third registration parameter based on the first registration parameter and the n second registration parameters includes: determining m fourth registration parameters and t fifth registration parameters from the n second registration parameters based on the first registration parameter, wherein both m and t are smaller than n, the sum of m and t is n, the difference between the fourth registration parameter and the first registration parameter is greater than or equal to a threshold, and the difference between the fifth registration parameter and the first registration parameter is smaller than the threshold; determining a sixth registration parameter based on the t fifth registration parameters, wherein the sixth registration parameter is used to register the areas corresponding to the t fifth registration parameters; and determining the third registration parameter based on the first registration parameter, the m fourth registration parameters, and the sixth registration parameter.

6. An electronic device, characterized in that: include: A processor and a memory, wherein the memory is used to store computer program code, the computer program code includes computer instructions, and when the processor executes the computer instructions, the electronic device executes the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions. When the program instructions are executed by a processor, the processor is caused to execute the method according to any one of claims 1 to 4.

8. A computer program product, characterized in that The computer program product comprises a computer program; when the computer program is run on a computer, the computer is caused to execute the method according to any one of claims 1 to 4.

Images