Fracture minimally invasive navigation assisted reduction position identification method and system

Abstract

According to the fracture minimally invasive navigation auxiliary reduction position identification method and system provided by the invention, the injury boundary of the fracture description range in the fracture CT shot image can be intelligently detected, and the fracture mode and the bone residue distribution mode can be intelligently identified; by means of the augmented reality technology, the original positions of each bone and bone residues can be determined according to the crack features, so that when hospital workers conduct treatment, the augmented reality technology can be specifically applied to the medical field, and when a fracture patient is treated, the medical treatment efficiency is greatly improved. The falling positions and original positions of each bone and bone residue can be determined, the treatment efficiency can be improved, and the workload of medical staff is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of data evaluation, and in particular to a method and system for identifying the position of minimally invasive navigation-assisted reduction of fractures. Background Art

[0002] When a bone fracture occurs, there may be some small fragments after the bone is broken, which are called bone fragments in this application. When medical staff perform treatment, they first need to find the location where the bone fragments fell and the initial location of the bone fragments. This work is very cumbersome and it is not possible to accurately find each location. It takes a lot of time for medical workers to perform medical treatment and it is not possible to accurately restore the fracture. Therefore, a technical solution is urgently needed to improve the above technical problems. Summary of the invention

[0003] In order to improve the technical problems existing in the related technologies, the present application provides a method and system for identifying the position of fracture minimally invasive navigation-assisted reduction.

[0004] In a first aspect, a method for identifying a position of a fracture minimally invasive navigation-assisted reduction is provided, comprising: Obtain CT images of the fracture that needs to be treated; In response to the subject's fracture description range selection operation on the fracture CT image to be processed, generating an original injured boundary of a target fracture description range corresponding to the fracture description range selection operation; In combination with the original injury boundary, determining a boundary node between the target fracture description range and the normal description range in the fracture CT image to be processed; A fracture reduction position identification result of the target fracture description range is generated according to the boundary nodes.

[0005] In an independently implemented embodiment, before determining the target fracture description range and the normal description range in combination with the original injury boundary, the method further includes: If the fracture CT image to be processed is not a description image of the target feature attribute, the fracture CT image to be processed is processed into a target feature attribute description image, and the target feature attribute description image is a binary image or a grayscale image.

[0006] In an independently implemented embodiment, generating the fracture reduction position identification result of the target fracture description range according to the boundary node includes: For each direction of each dimension in the first dimension and the second dimension, combining the boundary nodes in the direction, obtaining a boundary description method corresponding to the direction; Each of the boundary description modes is processed to generate a fracture reduction position recognition result within the target fracture description range.

[0007] In an independently implemented embodiment, combining the boundary nodes of the direction to obtain a boundary description method corresponding to the direction includes: If there is one boundary node in the direction, a boundary description method corresponding to the direction is determined by combining the boundary node in the direction and the dimension to which the direction belongs, where the dimension to which the direction belongs is the first dimension or the second dimension; If there are at least two boundary nodes in the direction, a boundary description method corresponding to the direction is determined in combination with the at least two boundary nodes in the direction.

[0008] In an independently implemented embodiment, generating the fracture reduction position identification result of the target fracture description range according to the boundary node includes: According to the boundary nodes, generating a pending injury boundary corresponding to the target fracture description range; If there is one pending injury boundary, the pending injury boundary is used as the fracture reduction position recognition result of the target fracture description range; If there are at least two pending injured boundaries, the at least two pending injured boundaries are cleaned according to their heights to obtain a fracture reduction position identification result of the target fracture description range.

[0009] In an independent implementation embodiment, it also includes: The fracture description range is annotated on the fracture CT image to be processed according to the fracture reduction position recognition result, so that the annotated fracture CT image to be processed is used as a training sample image for the fracture feature processing thread.

[0010] In an independent implementation example, the fracture CT images to be processed are obtained by: Obtaining an image acquisition operation, wherein the image acquisition operation includes a method for obtaining a fracture CT image that needs to be processed; Accessing a fracture sample image database according to the acquisition method; The fracture CT images to be processed corresponding to the acquisition method are obtained from the fracture sample image database.

[0011] In an independently implemented embodiment, after generating the fracture reduction position identification result of the target fracture description range according to the boundary node, the method further includes: filtering the fracture description range from the fracture CT images that need to be processed according to the fracture reduction position identification result; Perform fracture feature recognition on the filtered fracture description range to obtain a fracture feature recognition result; Displaying the fracture feature recognition result in the fracture feature annotation area of ​​the description image to be processed; In response to the approval operation of the fracture feature recognition result in the fracture feature annotation area, the fracture CT image annotation content is generated according to the approved fracture feature recognition result and the fracture CT image to be processed.

[0012] In an independently implemented embodiment, the combining the original injury boundary and determining the boundary node between the target fracture description range and the normal description range in the fracture CT image to be processed includes: For the target feature attribute description image of the fracture CT image to be processed, determining the target fracture description range and the normal description range in combination with the original injury boundary; Determine the pixel value of each fracture image pixel point in the target fracture description range and the normal description range; Determine each original boundary node between the target fracture description range and the normal description range according to the comparison result between the pixel values ​​of each fracture image pixel point; The boundary nodes include boundary nodes of a first dimension and boundary nodes of a second dimension, the first dimension is a description dimension of fracture features in the target fracture description range, the second dimension is a dimension corresponding to the first dimension, and the boundary nodes of each dimension in the first dimension and the second dimension include boundary nodes located in two directions of the fracture description range on the dimension; For each direction of each dimension in the first dimension and the second dimension, according to the difference of each original boundary node belonging to the direction in the dimension, if the difference between each original boundary node in the same direction is not greater than the target value, then at least one original boundary node is used as the final boundary node; if the difference between each original boundary node in the same direction is greater than the difference of the target value, then at least two final boundary nodes are determined from each original boundary node, and the at least two boundary nodes include the original boundary node with the farthest difference from another direction corresponding to the direction.

[0013] In a second aspect, a fracture minimally invasive navigation-assisted reduction position identification system is provided, comprising a processor and a memory communicating with each other, wherein the processor is used to read a computer program from the memory and execute it to implement the above method.

[0014] The embodiment of the present application provides a method and system for identifying the position of fracture reduction by minimally invasive navigation-assisted reduction, which can intelligently detect the injured boundary of the fracture description range in the fracture CT image, can intelligently identify the fracture mode of the fracture and the distribution mode of bone residues, and can determine the restoration of each bone and bone residue to its original position according to the crack characteristics through augmented reality technology. In this way, hospital staff can apply augmented reality technology to the medical field when conducting treatment. When treating fracture patients, they can determine the falling position and original position of each bone and bone residue, which can improve the efficiency of treatment and reduce the workload of medical staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A flowchart of a method for identifying the position of a fracture minimally invasive navigation-assisted reduction provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to better understand the above technical scheme, the technical scheme of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0018] See also Figure 1 , shows a method for identifying the position of minimally invasive navigation-assisted reduction of fractures, which may include the technical solutions described in the following steps S101-S104.

[0019] Step S101, obtaining a fracture CT image to be processed.

[0020] Among them, CT (Computed Tomography) is a medical imaging technology that uses X-ray beams to perform layer-by-layer scans of the human body and uses computer processing to produce detailed images of the body's internal structure.

[0021] The fracture CT image to be processed may refer to a description image that needs to be annotated with fracture features, and the fracture CT image to be processed includes a fracture description range. Optionally, after obtaining the fracture CT image to be processed, the fracture CT image to be processed may be displayed by the terminal device.

[0022] In an alternative embodiment, the fracture CT images to be processed are obtained by: Obtaining an image acquisition operation, wherein the image acquisition operation includes a method for obtaining a fracture CT image to be processed; accessing a fracture sample image database according to an acquisition method; A fracture CT image that needs to be processed corresponding to the acquisition method is obtained from the fracture sample image database.

[0023] Among them, when the object (such as a patient) wants to annotate the fracture features of the fracture CT image that needs to be processed, the fracture CT image that needs to be processed can be obtained first. When the object obtains the fracture CT image that needs to be processed, it can trigger a description of the image acquisition operation that includes a method for obtaining the fracture CT image that needs to be processed; accordingly, when the terminal device receives the description of the image acquisition operation triggered by the object, it can access the fracture sample image database in combination with the acquisition method included in the image acquisition operation, and obtain the fracture CT image that needs to be processed corresponding to the acquisition method from the fracture sample image database, and display the obtained fracture CT image that needs to be processed in the interface.

[0024] Among them, the embodiment of the present application does not limit the source of the fracture CT images that need to be processed. For example, the fracture sample image database can be a fracture sample image database describing the server side. At this time, the fracture CT images that need to be processed are the fracture CT images stored on the server side.

[0025] Optionally, if the fracture CT image that needs to be processed is a fracture CT image stored locally on the terminal device, and the fracture feature annotation of the fracture CT image that needs to be processed is performed by the server, the server does not know the specific fracture CT image that needs to be processed. Therefore, after the terminal device obtains the fracture CT image that needs to be processed according to the local storage method, it can send the fracture CT image that needs to be processed to the server.

[0026] Step S102, in response to the subject's fracture description range selection operation on the fracture CT image to be processed, generating an original injured boundary of a target fracture description range corresponding to the fracture description range selection operation.

[0027] This application displays CT images through augmented reality technology, which enables medical staff to obtain more real and reliable data, so that they can accurately reset to each position during resetting.

[0028] Among them, the fracture description range selection operation refers to the action of an object selecting a target fracture description range, and the specific form of triggering the fracture description range selection operation can be configured as needed, and is not limited in the embodiment of the present application.

[0029] Step S103, combining the original injury boundary, determining the boundary node between the target fracture description range and the normal description range in the fracture CT image to be processed.

[0030] Step S104, generating a fracture reduction position recognition result of a target fracture description range according to the boundary nodes.

[0031] Among them, the fracture reduction position recognition result can be understood as the position where the bone and bone debris are broken and fallen, as well as the initial position.

[0032] Wherein, the boundary node refers to the fracture image pixel point between the target fracture description range and the normal description range, and the boundary node can be determined according to the value of each fracture image pixel point within the original injured boundary of the target fracture description range. For example, when the pixel values ​​of two adjacent fracture image pixels of a fracture image pixel point within the original injured boundary are different, it means that the fracture image pixel point is between the target fracture description range and the normal description range, and the fracture image pixel point is the boundary node. Further, the fracture reduction position recognition result of the target fracture description range can be generated according to the determined boundary node. Wherein, when the fracture reduction position recognition result of the target fracture description range is generated according to the determined boundary node, it can be determined by the pixel value of the fracture image pixel point belonging to the fracture feature. Optionally, the object can trigger the fracture feature content selection operation, which refers to the action of the object wanting to determine the fracture reduction position recognition result of the target fracture description range and the action of informing the position of the fracture image pixel point of the fracture feature in the fracture CT image to be processed, and the specific form of triggering the fracture feature content selection operation can be configured as needed, which is not limited by the embodiment of the present application. For example, when an object clicks on any fracture feature included in the target fracture description range, it can be regarded as the object triggering the fracture feature content selection operation, and the fracture image pixel point at the clicked position is the fracture image pixel point of the fracture feature.

[0033] Accordingly, when the terminal device receives the fracture feature content selection operation of the object for the target fracture description range, the fracture reduction position recognition result of the target fracture description range can be generated according to the determined boundary node in response to the fracture feature content selection operation and displayed to the object. Optionally, the fracture reduction position recognition result is not greater than the original injury boundary.

[0034] It can be understood that when the fracture reduction position identification result of the target fracture description range is determined by the server, the terminal device can send a fracture feature content selection request to the server in response to the object's fracture feature content selection operation for the target fracture description range. When the server receives the fracture feature content selection request, it can know the position of the fracture image pixel point of the fracture feature in the fracture CT image that needs to be processed, and determine the fracture reduction position identification result of the target fracture description range, and return the determined fracture reduction position identification result to the terminal device, so that the terminal device displays the fracture reduction position identification result to the object.

[0035] In an optional embodiment of the present application, the method further includes: According to the fracture reduction position recognition result, the fracture description range is annotated on the fracture CT image to be processed, so that the annotated fracture CT image to be processed is used as a training sample image for the fracture feature processing thread.

[0036] As an optional embodiment, after obtaining the fracture reduction position recognition result of the target fracture description range, the fracture feature annotation can be performed on the fracture CT image to be processed according to the obtained fracture reduction position recognition result, and the annotated fracture CT image to be processed is used as a training sample image to train the original fracture feature processing thread, thereby obtaining the final fracture feature processing thread. Among them, the specific implementation method of annotating the fracture feature of the fracture CT image to be processed in combination with the fracture reduction position recognition result can be configured according to actual needs, and the embodiment of the present application is not limited. For example, when the fracture feature processing thread is a fracture feature recognition thread, the fracture feature included in the fracture reduction position recognition result in the fracture CT image to be processed can be annotated to obtain the annotated fracture CT image to be processed, and then the original fracture feature recognition thread is trained in combination with the annotated fracture CT image to be processed until the corresponding training conditions are met.

[0037] Optionally, when the scheme provided by the embodiment of the present application is interactively executed by the terminal device and the server, the above-mentioned obtaining of the fracture CT image to be processed can be that the terminal obtains the fracture CT image to be processed, and then sends the fracture CT image to be processed to the server, and the server responds to the object's fracture description range selection operation for the fracture CT image to be processed, generates the original injured boundary of the target fracture description range corresponding to the fracture description range selection operation, and determines the boundary node of the target fracture description range in the fracture CT image to be processed in combination with the original injured boundary, and then generates the fracture reduction position recognition result of the target fracture description range according to the boundary node. Optionally, when the fracture CT image to be processed is obtained, the fracture CT image to be processed can be displayed in the terminal device, and after the original injured boundary of the target fracture description range and the fracture reduction position recognition result of the target fracture description range are generated, the generated original injured boundary and the fracture reduction position recognition result of the target fracture description range can also be displayed to the object through the terminal device, so that the object can perform subsequent processing in combination with the displayed original injured boundary or fracture reduction position recognition result. Among them, when the method is executed by the server, the fracture CT image that needs to be processed, the original injured boundary of the target fracture description range, and the fracture reduction position identification result of the target fracture description range can be displayed through the display device corresponding to the server.

[0038] In combination with the solution provided in the embodiment of the present application, the injured boundary of the fracture description range in the fracture CT image can be intelligently detected, the fracture mode and the distribution mode of bone residues can be intelligently identified, and the augmented reality technology can be used to determine the restoration of each bone and bone residue to its original position according to the crack characteristics. In this way, hospital staff can apply augmented reality technology to the medical field when conducting treatment. When treating patients with fractures, the falling position and original position of each bone and bone residue can be determined, which can improve the efficiency of treatment and reduce the workload of medical staff.

[0039] In an optional embodiment of the present application, combining the original injury boundary, determining the boundary node between the target fracture description range and the normal description range in the fracture CT image to be processed includes: If the fracture CT image to be processed is not a description image of the target feature attribute, the fracture CT image to be processed is processed into a description image of the target feature attribute, and the description image of the target feature attribute is a binary image or a grayscale image; For the target feature attribute description image, the target fracture description range and the normal description range are determined in combination with the original injury boundary; According to the pixel value of each fracture image pixel point in the target fracture description range and the normal description range, the boundary node between the target fracture description range and the normal description range is determined.

[0040] Optionally, when generating the fracture reduction position recognition result of the target fracture description range, it can be determined whether the fracture CT image to be processed is a description image of the target feature attribute. If the fracture CT image to be processed is not a description image of the target feature attribute, the fracture CT image to be processed can be converted into a description image of the target feature attribute first, and then the fracture reduction position recognition result of the target fracture description range is generated by combining the processed fracture CT image to be processed. The target feature attribute description image can be a binary image or a grayscale image, that is, when the target feature attribute description image is a binary image, if the fracture CT image to be processed is not a binary image, the fracture CT image to be processed is converted into a binary image; and when the target feature attribute description image is a grayscale image, if the fracture CT image to be processed is not a grayscale image, the fracture CT image to be processed is converted into a grayscale image.

[0041] Processing the fracture CT image to be processed into a target feature attribute description image refers to adjusting the fracture image pixels in the fracture CT image to be processed.

[0042] Optionally, when converting the fracture CT image to be processed into a target feature attribute description image and obtaining the original injured boundary of the target fracture description range, the target fracture description range and normal description range can be determined in combination with the determined original injured boundary. For example, the target fracture description range and normal description range within the original injured boundary can be determined. At this time, compared with determining the entire fracture description range and normal description range in the fracture CT image to be processed, the data processing amount can be effectively reduced and the data processing efficiency can be improved. Optionally, when determining the target fracture description range and normal description range, the boundary node between the target fracture description range and the normal description range can be determined based on the pixel value of each fracture image pixel point in the target fracture description range and the normal description range, and then the fracture reduction position recognition result of the target fracture description range can be generated based on the boundary node.

[0043] In an optional embodiment of the present application, the boundary nodes include boundary nodes of a first dimension and boundary nodes of a second dimension, the first dimension is a description dimension of the fracture feature in the target fracture description range, the second dimension is a dimension corresponding to the first dimension, and the boundary nodes of each dimension in the first dimension and the second dimension include boundary nodes located in two directions of the fracture description range on the dimension; According to the boundary nodes, the fracture reduction position recognition results of the target fracture description range are generated, including: For each direction of each dimension in the first dimension and the second dimension, combining the boundary nodes in the direction, obtaining a boundary description method corresponding to the direction; Each boundary description method is processed to generate a fracture reduction position recognition result within the target fracture description range.

[0044] Optionally, for the boundary nodes of the first dimension and the boundary nodes of the second dimension, the boundary nodes of each dimension of the first dimension and the second dimension include the boundary nodes on the dimension located in both directions of the fracture description range, that is, the boundary nodes of each dimension include the boundary nodes in both directions of the fracture description range.

[0045] Correspondingly, when determining the fracture reduction position identification result of the target fracture description range based on the boundary nodes, for each direction of each dimension in the first dimension and the second dimension, the boundary nodes belonging to the direction can be first determined from each boundary node, and combined with the boundary nodes belonging to the direction, the boundary description method corresponding to the direction is obtained. Furthermore, when the boundary description method corresponding to each direction in the first dimension and the second dimension is obtained, each boundary description method can be processed to generate the fracture reduction position identification result of the target fracture description range.

[0046] In an optional embodiment of the present application, determining the boundary node between the target fracture description range and the normal description range according to the pixel value of each fracture image pixel point in the target fracture description range and the normal description range includes: Determine the pixel value of each fracture image pixel point in the target fracture description range and the normal description range; According to the comparison result between the pixel values ​​of the pixels of each fracture image, each original boundary node between the target fracture description range and the normal description range is determined; According to each original boundary node, the boundary node between the target fracture description range and the normal description range is determined.

[0047] Among them, since the fracture CT image to be processed is the fracture CT image of the target feature attribute description image, the pixel values ​​of the fracture image pixels in the normal description range in the fracture CT image to be processed are the same, and the pixel values ​​of the fracture image pixels in the fracture description range in the fracture CT image to be processed are the same, but the pixel values ​​of the fracture image pixels in the normal description range are different from the pixel values ​​of the fracture image pixels in the fracture description range. When determining the comparison result between the pixel values ​​of each fracture image pixel, if each fracture image pixel belongs to the same area (that is, the normal description range or the fracture description range), the comparison result between the pixel values ​​is 0 at this time, and when each fracture image pixel does not belong to the same area (such as one fracture image pixel belongs to the normal description range, and one fracture image pixel belongs to the fracture description range), the comparison result between the pixel values ​​will not be 0.

[0048] Accordingly, when determining the boundary nodes, the pixel values ​​of each fracture image pixel point in the target fracture description range and the normal description range can be determined, and then the original boundary nodes between the target fracture description range and the normal description range can be determined according to the comparison results between the pixel values ​​of each fracture image pixel point, and then the boundary nodes can be determined from the original boundary nodes. For example, when the comparison result between the pixel values ​​of two fracture image pixels is not 0, it means that the two fracture image pixels are respectively in the normal description range and the target fracture description range, and because the position of the fracture image pixel point of the fracture feature in the fracture CT image to be processed can be known by combining the fracture feature content selection operation, the pixel value of the fracture image pixel point belonging to the fracture feature can be known according to the pixel value of the fracture image pixel point at this position, and then the fracture image pixel point belonging to the fracture feature in the two fracture image pixels can be determined according to the pixel value of the fracture image pixel point belonging to the fracture feature, and the fracture image pixel point belonging to the fracture feature is used as the original adjacent fracture image pixel point.

[0049] In an optional embodiment of the present application, determining the boundary nodes between the target fracture description range and the normal description range according to each original boundary node includes: For each direction of each dimension in the first dimension and the second dimension, at least one final fracture image pixel point is determined from each original boundary node according to the difference of each original boundary node belonging to the direction in the dimension.

[0050] Optionally, for each direction of each dimension in the first dimension and the second dimension, the difference between each original boundary node belonging to the direction in the dimension can be determined, and then based on the difference between each original boundary node in the dimension, at least one final boundary node can be determined from each original boundary node.

[0051] In an optional embodiment of the present application, determining a final boundary node from each original boundary node includes: If the differences between the original boundary nodes in the same direction are not greater than the target value, then at least one of the original boundary nodes will be used as the final boundary node; If the difference between the original boundary nodes in the same direction is greater than the target value, then at least two final boundary nodes are determined from the original boundary nodes, and the at least two boundary nodes include the original boundary node with the farthest difference from another direction corresponding to the direction among the original boundary nodes.

[0052] Optionally, for each original boundary node belonging to the same direction, if it is determined that the difference between each original boundary node in this dimension is not greater than the target value, it means that each original boundary node is close to each other and can be regarded as being on the same horizontal line. At this time, any one or more original boundary nodes can be selected from each original boundary node as the final boundary node. On the contrary, if the difference between each original boundary node in this dimension is greater than the target value, it means that there are original boundary nodes that are far away from each other. At this time, at least two final boundary nodes can be determined from each original boundary node to ensure the accuracy of the fracture reduction position identification result determined by each boundary node.

[0053] Among them, the original boundary node with the farthest difference from the other direction corresponding to the direction in each original boundary node among the at least two boundary nodes determined. For example, if the two original boundary nodes corresponding to the difference greater than the target value belong to the left direction of the fracture description range, the difference between the original boundary nodes belonging to the left direction of the fracture description range and the right direction of the fracture description range can be determined respectively at this time, and then the original boundary node with the farthest difference in the right direction of the difference area is used as one of the at least two final boundary nodes. Of course, in actual applications, the two original boundary nodes corresponding to the difference greater than the target value can also be used as boundary nodes.

[0054] In an optional embodiment of the present application, combining the boundary nodes of the direction to obtain the boundary description method corresponding to the direction includes: If there is only one boundary node in the direction, then the boundary description method corresponding to the direction is determined by combining the boundary node in the direction and the dimension to which the direction belongs, and the dimension to which the direction belongs is the first dimension or the second dimension; If there are at least two boundary nodes in the direction, a boundary description method corresponding to the direction is determined in combination with the at least two boundary nodes in the direction.

[0055] Optionally, if there is one boundary node belonging to the same direction, the boundary description method corresponding to the direction can be determined in combination with the one boundary node and the dimension to which the direction belongs; correspondingly, if there are no less than two boundary nodes belonging to the same direction, the boundary description method corresponding to the direction can be determined in combination with the no less than two boundary nodes.

[0056] In an optional embodiment of the present application, a fracture reduction position recognition result of a target fracture description range is generated according to the boundary nodes, including: According to the boundary nodes, a pending injury boundary corresponding to the target fracture description range is generated; If there is only one pending injury boundary, the pending injury boundary is used as the fracture reduction position recognition result of the target fracture description range; If there are at least two pending injured boundaries, the at least two pending injured boundaries are cleaned according to their heights to obtain a fracture reduction position recognition result within the target fracture description range.

[0057] Optionally, when the original injured boundary of the target fracture description range corresponding to the fracture description range selection operation only includes the fracture description range that requires fracture feature annotation, the first pending injured boundary corresponding to the target fracture description range determined according to the boundary node will be one. At this time, the pending injured boundary can be directly used as the fracture reduction position identification result of the target fracture description range.

[0058] Optionally, when at least two pending injured boundaries are cleaned according to their heights to obtain the fracture reduction position identification result of the target fracture description range, the pending injured boundary with the largest height can be retained as the fracture reduction position identification result of the target fracture description range.

[0059] In an optional embodiment of the present application, after generating the fracture reduction position identification result of the target fracture description range according to the boundary node, it also includes: Filtering the fracture description range from the fracture CT images that need to be processed according to the fracture reduction position recognition result; Perform fracture feature recognition on the filtered fracture description range to obtain a fracture feature recognition result; The fracture feature recognition result is displayed in the fracture feature annotation area of ​​the description image that needs to be processed; In response to the approval operation of the fracture feature recognition result in the fracture feature annotation area, the fracture CT image annotation content is generated according to the approved fracture feature recognition result and the fracture CT image to be processed.

[0060] Optionally, after generating the fracture reduction position recognition result of the target fracture description range, the fracture reduction position recognition result may be used as a filter frame to filter from the fracture CT images to be processed to obtain a filtered fracture description range.

[0061] In order to better understand the minimally invasive navigation-assisted reduction position recognition method for fractures provided in the embodiment of the present application, the minimally invasive navigation-assisted reduction position recognition method for fractures is described in detail below in combination with a specific application scenario. The application scenario in this example is to determine the fracture reduction position recognition result of the target fracture description range in the fracture CT image to be processed according to the minimally invasive navigation-assisted reduction position recognition method for fractures, and then annotate the fracture description range of the fracture CT image to be processed according to the generated fracture reduction position recognition result, and use the annotated fracture CT image to be processed as a training sample image for the fracture feature processing thread.

[0062] Step S401, the terminal device obtains and displays the fracture CT image to be processed; Step S402, the terminal device displays the original injured boundary of the target fracture description range corresponding to the fracture description range selection operation; Specifically, when the object wants to annotate the fracture features in the fracture description range that needs to be processed, the fracture description range selection operation can be triggered to select the specific fracture description range that needs to be annotated (i.e., the target fracture description range). The terminal device responds to the fracture description range selection operation, determines and displays the original injured boundary of the target fracture description range corresponding to the fracture description range selection operation, and at the same time, the original injured boundary of the target fracture description range can be sent to the server to inform the server that the object wants to annotate the fracture features within the original injured boundary. When the terminal device sends the original injured boundary of the target fracture description range to the server, the coordinates of the original injured boundary of the target fracture description range in the fracture CT image that needs to be processed can be sent to the server, and the server can determine the original injured boundary of the target fracture description range based on the received coordinates.

[0063] Step S403, the terminal device receives the object's fracture feature content selection operation for the target fracture description range, and sends a fracture feature content selection request to the server; Specifically, after the original injured boundary of the target fracture description range is displayed to the object, the object can trigger a fracture feature content selection operation for the target fracture description range. When the terminal device receives the fracture feature content selection operation, it means that the object wants to determine the fracture reduction position recognition result of the target fracture description range. At this time, a fracture feature content selection request can be sent to the server. When the server receives the fracture feature content selection request, it can know the position of the fracture image pixel point of the fracture feature in the fracture CT image that needs to be processed. Step S404, the server responds to the fracture feature content selection request and determines whether the fracture CT image to be processed is a binary image, if so, executes step S406, otherwise, executes step S405; Step S405: converting the fracture CT image to be processed into a binary image; Specifically, when the server receives a fracture feature content selection request sent by a terminal device, it can determine whether the fracture CT image to be processed is a black and white image. If the fracture CT image to be processed is not a black and white image, each fracture image pixel in the fracture CT image to be processed can be binarized to obtain the processed fracture CT image to be processed, and then the fracture reduction position recognition result of the target fracture description range is determined according to the processed description image.

[0064] Step S406, the server determines each original boundary node between the target fracture description range and the normal description range; Furthermore, the original boundary nodes between the target fracture description range and the normal description range can be determined according to the comparison result between the pixel values ​​of the fracture image pixels. For example, when the comparison result between the pixel values ​​of two fracture image pixels is not 0, it means that the two fracture image pixels are respectively in the normal description range and the target fracture description range, and then the fracture image pixel belonging to the fracture feature can be determined according to the pixel value of the fracture image pixel belonging to the fracture feature, and the fracture image pixel belonging to the fracture feature is used as the original boundary node.

[0065] Step S407, the server determines the boundary nodes between the target fracture description range and the normal description range according to each original boundary node; Optionally, the boundary nodes may include boundary nodes of a first dimension and boundary nodes of a second dimension, and the first dimension may refer to the description dimension of the fracture feature in the target fracture description range, and the second dimension is the dimension corresponding to the first dimension. Further, for each direction of each dimension in the first dimension and the second dimension, the difference between each original boundary node in the dimension may be determined, and then the boundary node of the direction may be determined from each original boundary node according to the difference between each original boundary node in the dimension.

[0066] Among them, when determining the boundary nodes from the original boundary nodes according to the difference of each original boundary node in this dimension, for each original boundary node belonging to the same direction, if it is determined that the difference between each original boundary node in this dimension is not greater than the target value, it means that the original boundary nodes are close to each other and can be regarded as being on the same horizontal line. At this time, no less than one original boundary node can be selected from the original boundary nodes as the boundary node. On the contrary, if the difference between each original boundary node in this dimension is greater than the target value, it means that there are original boundary nodes that are far away from each other. At this time, no less than two boundary nodes can be determined from each original boundary node to ensure the accuracy of the fracture reduction position identification result finally determined according to each boundary node.

[0067] Step S408: The server generates a fracture reduction position identification result for the target fracture description range according to the boundary nodes.

[0068] Specifically, for the same direction in each dimension of the first dimension and the second dimension, the boundary nodes belonging to the same direction can be first determined from the adjacent fracture image pixels, and the boundary nodes belonging to the same direction can be combined to obtain the boundary description method corresponding to the direction. Furthermore, when the boundary description method corresponding to each direction in the first dimension and the second dimension is obtained, each boundary description method can be processed to obtain the fracture reduction position recognition result of the target fracture description range.

[0069] Step S409, sending the fracture reduction position recognition result of the target fracture description range to the terminal device, so that the terminal device displays the fracture reduction position recognition result of the target fracture description range to the subject.

[0070] On the basis of the above, a fracture minimally invasive navigation-assisted reduction position identification device is provided, the device comprising: An image acquisition module is used to obtain a fracture CT image that needs to be processed; A boundary determination module, configured to generate an original injured boundary of a target fracture description range corresponding to the fracture description range selection operation in response to the subject's fracture description range selection operation on the fracture CT image to be processed; A node determination module, used to determine the boundary node between the target fracture description range and the normal description range in the fracture CT image to be processed in combination with the original injury boundary; A position identification module is used to generate a fracture reduction position identification result of the target fracture description range according to the boundary node.

[0071] Based on the above, a minimally invasive navigation-assisted reduction position identification system for fractures is shown, including a processor and a memory that communicate with each other, and the processor is used to read a computer program from the memory and execute it to implement the above method.

[0072] Based on the above, a computer-readable storage medium is also provided, on which a computer program stored implements the above method when running.

[0073] In summary, based on the above scheme, the injured boundary of the fracture description range in the fracture CT image can be intelligently detected, the fracture mode and the distribution mode of bone residues can be intelligently identified, and the augmented reality technology can be used to determine the restoration of each bone and bone residue to its original position according to the crack characteristics. In this way, hospital staff can apply augmented reality technology to the medical field when conducting treatment. When treating patients with fractures, the falling position and original position of each bone and bone residue can be determined, which can improve the efficiency of treatment and reduce the workload of medical staff.

[0074] It should be understood that the system and its modules shown above can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art will understand that the above methods and systems can be implemented using computer executable instructions and / or included in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software such as executed by various types of processors, and can also be implemented by a combination of the above hardware circuits and software (e.g., firmware).

[0075] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or any other beneficial effects that may be obtained.

Claims

1. A method for identifying the position of minimally invasive navigation-assisted fracture reduction, characterized in that: include: Obtain CT images of the fracture that needs to be treated; In response to the subject's fracture description range selection operation on the fracture CT image to be processed, generating an original injured boundary of a target fracture description range corresponding to the fracture description range selection operation; In combination with the original injury boundary, determining a boundary node between the target fracture description range and the normal description range in the fracture CT image to be processed; A fracture reduction position identification result of the target fracture description range is generated according to the boundary nodes.

2. The method according to claim 1, characterized in that Before determining the target fracture description range and the normal description range in combination with the original injury boundary, the method further includes: If the fracture CT image to be processed is not a description image of the target feature attribute, the fracture CT image to be processed is processed into a target feature attribute description image, and the target feature attribute description image is a binary image or a grayscale image.

3. The method according to claim 2, characterized in that The step of generating the fracture reduction position recognition result of the target fracture description range according to the boundary node includes: For each direction of each dimension in the first dimension and the second dimension, combining the boundary nodes in the direction, obtaining a boundary description method corresponding to the direction; Each of the boundary description modes is processed to generate a fracture reduction position recognition result within the target fracture description range.

4. The method according to claim 3, characterized in that The step of combining the boundary nodes in the direction to obtain the boundary description method corresponding to the direction includes: If there is one boundary node in the direction, a boundary description method corresponding to the direction is determined by combining the boundary node in the direction and the dimension to which the direction belongs, where the dimension to which the direction belongs is the first dimension or the second dimension; If there are at least two boundary nodes in the direction, a boundary description method corresponding to the direction is determined in combination with the at least two boundary nodes in the direction.

5. The method according to claim 3, characterized in that: The step of generating the fracture reduction position recognition result of the target fracture description range according to the boundary node includes: According to the boundary nodes, generating a pending injury boundary corresponding to the target fracture description range; If there is one pending injury boundary, the pending injury boundary is used as the fracture reduction position recognition result of the target fracture description range; If there are at least two pending injured boundaries, the at least two pending injured boundaries are cleaned according to their heights to obtain a fracture reduction position identification result of the target fracture description range.

6. The method according to claim 3, characterized in that Also includes: The fracture description range is annotated on the fracture CT image to be processed according to the fracture reduction position recognition result, so that the annotated fracture CT image to be processed is used as a training sample image for the fracture feature processing thread.

7. The method according to claim 1, characterized in that The fracture CT images to be processed are obtained by: Obtaining an image acquisition operation, wherein the image acquisition operation includes a method for obtaining a fracture CT image that needs to be processed; Accessing a fracture sample image database according to the acquisition method; The fracture CT images to be processed corresponding to the acquisition method are obtained from the fracture sample image database.

8. The method according to claim 1, characterized in that After the fracture reduction position recognition result of the target fracture description range is generated according to the boundary node, the method further includes: filtering the fracture description range from the fracture CT images that need to be processed according to the fracture reduction position identification result; Perform fracture feature recognition on the filtered fracture description range to obtain a fracture feature recognition result; Displaying the fracture feature recognition result in the fracture feature annotation area of ​​the description image to be processed; In response to the approval operation of the fracture feature recognition result in the fracture feature annotation area, the fracture CT image annotation content is generated according to the approved fracture feature recognition result and the fracture CT image to be processed.

9. The method according to claim 1, characterized in that: The combining the original injury boundary and determining the boundary node between the target fracture description range and the normal description range in the fracture CT image to be processed includes: For the target feature attribute description image of the fracture CT image to be processed, determining the target fracture description range and the normal description range in combination with the original injury boundary; Determine the pixel value of each fracture image pixel point in the target fracture description range and the normal description range; Determine each original boundary node between the target fracture description range and the normal description range according to the comparison result between the pixel values ​​of each fracture image pixel point; The boundary nodes include boundary nodes of a first dimension and boundary nodes of a second dimension, the first dimension is a description dimension of fracture features in the target fracture description range, the second dimension is a dimension corresponding to the first dimension, and the boundary nodes of each dimension in the first dimension and the second dimension include boundary nodes located in two directions of the fracture description range on the dimension; For each direction of each dimension in the first dimension and the second dimension, according to the difference of each original boundary node belonging to the direction in the dimension, if the difference between each original boundary node in the same direction is not greater than the target value, then at least one original boundary node is used as the final boundary node; if the difference between each original boundary node in the same direction is greater than the difference of the target value, then at least two final boundary nodes are determined from each original boundary node, and the at least two boundary nodes include the original boundary node with the farthest difference from another direction corresponding to the direction.

10. A minimally invasive navigation-assisted fracture reduction position recognition system, characterized in that: The invention comprises a processor and a memory communicating with each other, wherein the processor is used to read a computer program from the memory and execute the computer program to implement the method according to any one of claims 1 to 9.

Images