Pelvic fracture reduction method and device based on deep learning, terminal, medium and product

Abstract

This application provides a method, device, terminal, medium, and product for pelvic fracture reduction based on deep learning. The method includes: acquiring an image of the pelvis to be reduced and performing coarse alignment processing; obtaining a pelvic reduction reference image based on a trained pelvic reduction model and the coarsely aligned pelvic image; and obtaining a pelvic reduction image using the ICP algorithm based on the pelvic reduction reference image and the image of the pelvis to be reduced, thereby providing data support for pelvic fracture surgical planning. This application not only improves the efficiency of pelvic fracture reduction but also improves its accuracy.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, device, terminal, medium and product for pelvic fracture reduction based on deep learning. Background Technology

[0002] Pelvic fractures are a serious injury with high morbidity and mortality rates. Typically, pelvic fractures require complex reconstructive surgery to restore the anatomical and functional integrity of the pelvis. Given the high technical demands and associated risks of this surgery, precise and detailed preoperative planning is essential. Computer-assisted surgery (CAS) systems can be used for planning clinical pelvic fracture surgeries. Surgeons can use CAS to design pelvic reduction strategies and determine surgical methods and steps based on the patient's preoperative CT scans. However, CAS-based pelvic fracture surgical planning is semi-automatic, requiring tedious and time-consuming manual intervention from the surgeon, and the planning outcome largely depends on the surgeon's clinical experience.

[0003] Early research focused on improving the operability of manual reduction. However, even for highly skilled surgeons, manual reduction still requires significant time and effort, especially in complex cases. Current methods focus on leveraging geometric principles and imaging information to achieve more automated approaches in surgical planning. These methods can be broadly categorized into two types. The first type is based on point cloud alignment, which extracts functionally relevant point clouds from the fracture surface of bone fragments and then aligns them through registration to achieve pelvic reduction. The main drawback of the first type is its high sensitivity to noise, making it prone to cumulative errors, especially in pelvic fractures with multiple fragments. The second type is based on template alignment, where the template is constructed as a simplified target using a prior model. The second type often ignores individual patient-specific imaging. Therefore, variations within the dataset, such as imaging characteristics, resolution, and individual variability, can render the second type ineffective. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the purpose of this application is to provide a deep learning-based method, device, terminal, medium and product for pelvic fracture reduction, in order to solve the problem of low efficiency in pelvic fracture reduction using existing methods.

[0005] To achieve the above and other related objectives, a first aspect of this application provides a deep learning-based method for pelvic fracture reduction, comprising: acquiring an image of the pelvis to be reduced and performing coarse alignment processing; obtaining a pelvic reduction reference image based on a trained pelvic reduction model and the coarsely aligned pelvic image; and performing registration processing on the pelvic image to be reduced using the ICP algorithm based on the pelvic reduction reference image to obtain a pelvic reduction image.

[0006] In some embodiments of the first aspect of this application, the specific method for training the pelvic reduction model includes: acquiring multiple paired fracture slice images and reduction slice images to establish a pelvic dataset; inputting the pelvic dataset into a GAN network for training to obtain a pelvic reduction model.

[0007] In some embodiments of the first aspect of this application, based on a trained pelvic repositioning model, a pelvic repositioning reference image is obtained from a coarsely aligned pelvic image, including: inputting multiple pelvic slice images corresponding to the coarsely aligned pelvic image into the pelvic repositioning model to obtain multiple corresponding pelvic repositioning slice images; integrating the multiple pelvic repositioning slice images, and performing median filtering on the integrated image to obtain the pelvic repositioning reference image.

[0008] In some embodiments of the first aspect of this application, the pelvic repositioning image is obtained by registering the pelvic image to be repositioned using the ICP algorithm based on the pelvic repositioning reference image, including: performing preliminary registration processing on the pelvic image to be repositioned based on the pelvic repositioning reference image to obtain a preliminary registered image; and performing fracture surface registration processing on the preliminary registered image using the ICP algorithm to obtain the pelvic repositioning image.

[0009] In some embodiments of the first aspect of this application, preliminary registration processing is performed on the pelvic image to be repositioned based on the pelvic repositioning reference image to obtain a preliminary registration image, including: performing image segmentation processing on the pelvic image to be repositioned to obtain multiple fracture tissues; keeping the position of the fracture tissue with the largest volume among the multiple fracture tissues unchanged, and aligning the remaining fracture tissues according to the pelvic repositioning reference image to obtain a preliminary registration image.

[0010] In some embodiments of the first aspect of this application, the ICP algorithm is used to perform fracture surface registration processing on the pre-registered image to obtain a pelvic reduction image, including: obtaining a point cloud dataset of each pelvic structure in the pre-registered image; calculating a relevant point set of each two adjacent pelvic structures based on the point cloud dataset of the fracture surfaces of each two adjacent pelvic structures; aligning the fracture surfaces of each two adjacent fractured tissues according to the relevant point sets of each two adjacent pelvic structures; and outputting the image obtained after registration as a pelvic reduction image.

[0011] To achieve the above and other related objectives, a second aspect of this application provides a deep learning-based pelvic fracture reduction device, comprising: an acquisition module for acquiring an image of the pelvis to be reduced and performing coarse alignment processing; a reference image generation module for obtaining a pelvic reduction reference image based on a trained pelvic reduction model and the coarsely aligned pelvic image; and a reduction module for obtaining a pelvic reduction image based on the pelvic reduction reference image and the image of the pelvis to be reduced, using an ICP algorithm.

[0012] To achieve the above and other related objectives, a third aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the deep learning-based pelvic fracture reduction method.

[0013] To achieve the above and other related objectives, a fourth aspect of this application provides a computer program product comprising computer program code that, when executed on a computer, causes the computer to implement the deep learning-based pelvic fracture reduction method.

[0014] To achieve the above and other related objectives, a fifth aspect of this application provides an electronic terminal, including a memory, a processor, and a computer program stored in the memory; the processor executes the computer program to implement the deep learning-based pelvic fracture reduction method.

[0015] As described above, the deep learning-based pelvic fracture reduction method, device, terminal, medium, and product of this application have the following beneficial effects:

[0016] This application not only improves the efficiency of pelvic fracture reduction, but also improves the accuracy of pelvic fracture reduction. Attached Figure Description

[0017] Figure 1 The diagram shown is a flowchart of a deep learning-based pelvic fracture reduction method in one embodiment of this application.

[0018] Figure 2 The diagram shown is a schematic representation of the structure of a GAN network in one embodiment of this application.

[0019] Figure 3 The diagram shown is a flowchart of pelvic fracture reduction in a specific embodiment of this application.

[0020] Figure 4 The diagram shown is a schematic block diagram of a deep learning-based pelvic fracture reduction device according to an embodiment of this application.

[0021] Figure 5The diagram shown is a structural schematic of an electronic terminal according to an embodiment of this application. Detailed Implementation

[0022] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0023] In the embodiments of this application, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that the terms "first" and "second" do not necessarily imply that they are different.

[0024] It should be noted that, in the embodiments of this application, the words "exemplary" or "for example" indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0025] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0026] To facilitate understanding of the embodiments of this application, firstly, in conjunction with Figure 1 Detailed explanation. Figure 1 This document illustrates a flowchart of a deep learning-based pelvic fracture reduction method according to an embodiment of the present invention. The deep learning-based pelvic fracture reduction method in this embodiment mainly includes the following steps:

[0027] Step S11: Obtain an image of the pelvis to be repositioned and perform coarse alignment processing.

[0028] In one embodiment, the image of the pelvis to be reduced is a CT image of a pelvic fracture. Specifically, the CT image of the pelvic fracture is a three-dimensional image.

[0029] In one embodiment, the specific process of coarse alignment processing of the pelvic image to be repositioned includes: performing image segmentation processing on the pelvic image to be repositioned to obtain multiple pelvic structures; and performing mirror registration processing on the multiple pelvic structures to obtain a coarsely aligned pelvic image.

[0030] Specifically, the types of pelvic structures include: the hip bone, sacrum, and coccyx. The specific process of mirror registration of the pelvic structures includes: for the hip bone, mirror aligning the healthy hip bone with the affected hip bone; for the sacrum, mirror aligning the healthy and affected sides of the sacrum according to the plane of symmetry; for the coccyx, mirror aligning the healthy and affected sides of the coccyx according to the plane of symmetry.

[0031] Step S12: Based on the trained pelvic repositioning model, obtain a pelvic repositioning reference image according to the coarsely aligned pelvic image.

[0032] In one embodiment, the specific method for training the pelvic reduction model includes: acquiring multiple paired fracture slice images and reduction slice images to establish a pelvic dataset; inputting the pelvic dataset into a GAN network for training to obtain a pelvic reduction model.

[0033] Specifically, each coarsely aligned fracture slice image and its corresponding real repositioned slice image are paired fracture slice images and repositioned slice images. It should be noted that the specific process of coarse alignment is the same as in the above embodiments, and will not be repeated here.

[0034] Furthermore, such as Figure 2 As shown, the GAN network includes a generator and a discriminator. The generator generates simulated repositioned slice images, and the discriminator determines whether the input image is simulated or real. The generator uses a CNN model. The discriminator consists of three identical feature extraction layers and one fully connected layer; each feature extraction layer consists of sequentially connected 3×3 convolutional layers, Leaky ReLU layers, and Batch Normalization layers.

[0035] Furthermore, each coarsely aligned fracture slice image is input into the generator to generate a simulated repositioned slice image and an image deformation field. The simulated repositioned slice image and the real repositioned slice image are then input into the discriminator for authenticity assessment. It should be understood that the deformation field describes how to transform the shape of one image into another, making them as similar as possible.

[0036] Furthermore, the objective function of the GAN network is shown in Equation 1 below. The objective function defines the adversarial training process between the generator and the discriminator.

[0037]

[0038] Where E(·) is the expected distribution of the data, P frac(x) P represents the distribution of fracture slice image samples. reduct(z) Let D(x) represent the distribution of real reset slice image samples, D(x) represent the probability that the discriminator classifies the real reset slice image x as real, and D(G(z)) represent the probability that the discriminator classifies the simulated reset slice image G(z) as real.

[0039] Furthermore, during the training process of the pelvic reduction model, the discriminator adopts the discriminator loss function D shown in Formula 2. loss The generator uses the generator loss function G shown in Formula 3 below. loss :

[0040]

[0041]

[0042] Where, x i Represents a fracture section image, z i The image represents a slice of fracture reduction, m represents the number of images, D(·) represents the discriminator, and G(·) represents the generator.

[0043] In one embodiment, such as Figure 3 As shown, based on the trained pelvic repositioning model, a pelvic repositioning reference image is obtained from the coarsely aligned pelvic image. This includes: inputting multiple pelvic slice images corresponding to the coarsely aligned pelvic image into the pelvic repositioning model to obtain multiple corresponding pelvic repositioning slice images; integrating the multiple pelvic repositioning slice images and performing median filtering on the integrated image to obtain the pelvic repositioning reference image.

[0044] Specifically, CT (computed tomography) equipment can obtain multiple two-dimensional cross-sectional images of the pelvis (pelvic slice images), each of which is a two-dimensional grayscale image. These pelvic slice images are then combined to obtain a three-dimensional pelvic image. The Marching Cubes algorithm can be used to combine these slice images. Marching Cubes is an algorithm that renders isosurfaces from volume data.

[0045] It should be noted that the basic principle of median filtering is to replace the value of a point in a digital image or digital sequence with the median value of all points in its neighborhood, so that the surrounding pixel values ​​are close to the true value, thereby eliminating isolated noise points.

[0046] In one specific embodiment, the Marching Cubes algorithm is used to integrate the multiple pelvic repositioning slice images.

[0047] Step S13: Based on the pelvic repositioning reference image, the ICP algorithm is used to register the pelvic image to be repositioned to obtain the pelvic repositioning image.

[0048] In one embodiment, such as Figure 3 As shown, based on the pelvic repositioning reference image, the ICP algorithm is used to register the pelvic image to be repositioned to obtain a pelvic repositioning image. This includes: performing preliminary registration processing on the pelvic image to be repositioned based on the pelvic repositioning reference image to obtain a preliminary registered image; and using the ICP algorithm to perform fracture surface registration processing on the preliminary registered image to obtain a pelvic repositioning image.

[0049] In one embodiment, such as Figure 3 As shown, based on the pelvic reduction reference image, the pelvic image to be reduced is initially registered to obtain an initial registered image, including: performing image segmentation processing on the pelvic image to be reduced to obtain multiple fracture tissues; keeping the position of the fracture tissue with the largest volume unchanged among the multiple fracture tissues, and aligning the remaining fracture tissues according to the pelvic reduction reference image to obtain an initial registered image.

[0050] Specifically, following the outline of the pelvis in the pelvic reduction reference image, the remaining fractured tissues are aligned with the pelvic reduction reference image.

[0051] It should be noted that all image segmentation processes in the above embodiments can use existing image segmentation models, and the present invention does not limit them.

[0052] In one embodiment, such as Figure 3As shown, the ICP algorithm is used to perform fracture surface registration processing on the initially registered image to obtain a pelvic reduction image. This includes: obtaining a point cloud dataset of each fractured tissue in the initially registered image; calculating the relevant point set of each two adjacent fractured tissues based on the point cloud dataset of the fracture surfaces of each two adjacent fractured tissues; aligning the fracture surfaces of each two adjacent fractured tissues according to the relevant point set of each two adjacent fractured tissues; and outputting the image obtained after registration as a pelvic reduction image.

[0053] Specifically, the point cloud datasets of the fracture surfaces of two adjacent fractured tissues are P f P m Among them, P f ={f i |f i ∈R 3 i = 1, 2, ..., N f};P m ={m i |m i ∈R 3 i = 1, 2, ..., N m Point cloud dataset P f and point cloud dataset P m The relevant points are It should be noted that the distribution of points at the fracture surface of fractured tissue is denser than that on a smooth surface. In point cloud datasets, related points typically refer to points of the same physical object in different views or scans.

[0054] Furthermore, the calculation satisfies Relevant points Relevant points For point cloud dataset P m In point cloud dataset P f The following are relevant points;

[0055] Furthermore, the calculation satisfies rotation matrix R k and the translation matrix T k ;

[0056] Furthermore, the relevant point set S is calculated. k+1 ,

[0057] Furthermore, the relevant point set represents the correspondence between each point on the fracture surface of the fractured tissue and each point on the fracture surface of the adjacent fractured tissue. This correspondence allows the fracture surfaces of the two fractured tissues to be aligned.

[0058] It should be understood that registration is considered complete when every two adjacent fracture tissues in all fracture tissues are spliced ​​together.

[0059] Figure 4 This is a schematic block diagram of a deep learning-based pelvic fracture reduction device provided in an embodiment of this application. Figure 4 As shown, the deep learning-based pelvic fracture reduction device 4 includes:

[0060] The acquisition module 41 is used to acquire an image of the pelvis to be repositioned and perform coarse alignment processing.

[0061] The reference image generation module 42 is used to obtain a pelvic repositioning reference image based on the trained pelvic repositioning model and the coarsely aligned pelvic image.

[0062] The reset module 43 is used to perform registration processing on the pelvic image to be reset using the ICP algorithm based on the pelvic reset reference image to obtain a pelvic reset image.

[0063] It should be understood that the specific process of each module performing the above-mentioned steps has been described in detail in the above method embodiments, and will not be repeated here for the sake of brevity.

[0064] It should also be understood that the module division in the embodiments of this application is illustrative and only represents a logical functional division; in actual implementation, there may be other division methods. Furthermore, the functional modules in the various embodiments of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0065] In one embodiment, the specific method for training the pelvic reduction model includes: acquiring multiple paired fracture slice images and reduction slice images to establish a pelvic dataset; inputting the pelvic dataset into a GAN network for training to obtain a pelvic reduction model.

[0066] In one embodiment, based on a trained pelvic repositioning model, a pelvic repositioning reference image is obtained from a coarsely aligned pelvic image. This includes: inputting multiple pelvic slice images corresponding to the coarsely aligned pelvic image into the pelvic repositioning model to obtain multiple corresponding pelvic repositioning slice images; integrating the multiple pelvic repositioning slice images and performing median filtering on the integrated image to obtain the pelvic repositioning reference image.

[0067] In one embodiment, based on the pelvic repositioning reference image, the ICP algorithm is used to register the pelvic image to be repositioned to obtain a pelvic repositioning image, including: performing preliminary registration processing on the pelvic image to be repositioned based on the pelvic repositioning reference image to obtain a preliminary registered image; and using the ICP algorithm to perform fracture surface registration processing on the preliminary registered image to obtain a pelvic repositioning image.

[0068] In one embodiment, a preliminary registration process is performed on the pelvic image to be repositioned based on the pelvic repositioning reference image to obtain a preliminary registration image. This includes: performing image segmentation processing on the pelvic image to be repositioned to obtain multiple fracture tissues; keeping the position of the fracture tissue with the largest volume among the multiple fracture tissues unchanged, and aligning the remaining fracture tissues according to the pelvic repositioning reference image to obtain a preliminary registration image.

[0069] In one embodiment, the ICP algorithm is used to perform fracture surface registration processing on the initially registered image to obtain a pelvic reduction image, including: obtaining a point cloud dataset of each pelvic structure in the initially registered image; calculating a relevant point set for each two adjacent pelvic structures based on the point cloud dataset of the fracture surfaces of each two adjacent pelvic structures; aligning the fracture surfaces of each two adjacent fractured tissues according to the relevant point sets of each two adjacent pelvic structures; and outputting the image obtained after registration as a pelvic reduction image.

[0070] Figure 5 This is a schematic block diagram of the electronic terminal provided in an embodiment of this application. Figure 5 As shown, the electronic terminal includes at least one processor 501, a memory 502, at least one network interface 503, and a user interface 505. The various components in the device are coupled together via a bus system 504. It is understood that the bus system 504 is used to implement communication between these components. In addition to a data bus, the bus system 504 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in… Figure 5 The general will label all buses as bus systems.

[0071] The user interface 505 may include a monitor, keyboard, mouse, trackball, clicker, button, touchpad, or touch screen.

[0072] It is understood that memory 502 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM) or programmable read-only memory (PROM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM) and synchronous static random access memory (SSRAM). The memories described in the embodiments of this invention are intended to include, but are not limited to, these and any other suitable categories of memory.

[0073] In this embodiment of the invention, the memory 502 is used to store various types of data to support the operation of the electronic terminal 500. Examples of this data include: any executable program for operation on the electronic terminal 500, such as the operating system 5021 and application programs 5022; the operating system 5021 contains various system programs, such as the framework layer, core library layer, driver layer, etc., for implementing various basic services and handling hardware-based tasks. The application program 5022 may contain various applications, such as a media player, browser, etc., for implementing various application services. The methods provided in this embodiment of the invention can be included in the application program 5022.

[0074] The methods disclosed in the above embodiments of the present invention can be applied to processor 501, or implemented by processor 501. Processor 5401 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in processor 501 or by instructions in the form of software. The processor 501 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 501 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. General-purpose processor 501 may be a microprocessor or any conventional processor, etc. The steps of the accessory optimization method provided in the embodiments of the present invention can be directly reflected as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium, which is located in a memory. The processor reads the information in the memory and combines it with its hardware to complete the steps of the aforementioned method.

[0075] In an exemplary embodiment, the electronic terminal 500 may be used by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), or complex programmable logic devices (CPLDs) to execute the aforementioned method.

[0076] According to the method provided in the embodiments of this application, this application also provides a computer program product, which includes: computer program code, which, when run on a computer, causes the computer to execute... Figure 1 The embodiment shown illustrates a deep learning-based method for pelvic fracture reduction.

[0077] According to the method provided in the embodiments of this application, this application also provides a computer-readable storage medium storing program code, which, when executed on a computer, causes the computer to perform... Figure 1 The embodiment shown illustrates a deep learning-based method for pelvic fracture reduction.

[0078] As used in this specification, the terms "component," "module," "system," etc., are used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. As illustrated, applications running on computing devices and computing devices can both be components. One or more components may reside in a process and / or an execution thread, and components may be located on a single computer and / or distributed among two or more computers. Furthermore, these components can be executed from various computer-readable media on which various data structures are stored. Components can communicate, for example, via local and / or remote processes based on signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system, and / or a network, such as the Internet interacting with other systems via signals).

[0079] Those skilled in the art will recognize that the various illustrative logical blocks and steps 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 implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0080] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0081] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0082] The units described as separate components may or may not be physically separate. 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 the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0083] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0084] In the above embodiments, the functions of each functional unit can be implemented entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. A computer program product includes one or more computer instructions (programs). When the computer program instructions (programs) are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, 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, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., high-density digital video discs, DVDs), or semiconductor media (e.g., solid-state disks, SSDs, etc.).

[0085] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0086] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0087] In summary, this application provides a method, device, terminal, medium, and product for pelvic fracture reduction based on deep learning. The method includes: acquiring an image of the pelvis to be reduced and performing coarse alignment processing; obtaining a pelvic reduction reference image based on a trained pelvic reduction model and the coarsely aligned pelvic image; and obtaining a pelvic reduction image using the ICP algorithm based on the pelvic reduction reference image and the image of the pelvis to be reduced, thereby providing data support for pelvic fracture surgical planning. This application not only improves the efficiency of pelvic fracture reduction but also enhances its accuracy. Therefore, this application effectively overcomes the various shortcomings of existing technologies and has high industrial applicability.

[0088] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A deep learning-based method for pelvic fracture reduction, characterized in that, include: Acquire an image of the pelvis to be repositioned and perform coarse alignment processing; Based on the trained pelvic repositioning model, a pelvic repositioning reference image is obtained from the coarsely aligned pelvic image. Based on the pelvic repositioning reference image, the ICP algorithm is used to register the pelvic image to be repositioned to obtain the pelvic repositioning image.

2. The deep learning-based pelvic fracture reduction method according to claim 1, characterized in that, Specific methods for training the pelvic reduction model include: Multiple paired fracture slice images and reduction slice images were acquired to build a pelvic dataset; The pelvic dataset is input into a GAN network for training to obtain a pelvic repositioning model.

3. The deep learning-based pelvic fracture reduction method according to claim 2, characterized in that, Based on the trained pelvic repositioning model, a pelvic repositioning reference image is obtained from the coarsely aligned pelvic image, including: Multiple pelvic slice images corresponding to the coarsely aligned pelvic image are input into the pelvic repositioning model to obtain multiple corresponding pelvic repositioning slice images. The multiple pelvic repositioning slice images are integrated, and the integrated image is subjected to median filtering to obtain a pelvic repositioning reference image.

4. The deep learning-based pelvic fracture reduction method according to claim 1, characterized in that, Based on the pelvic repositioning reference image, the ICP algorithm is used to register the pelvic image to be repositioned, resulting in a pelvic repositioning image, including: Based on the pelvic repositioning reference image, the pelvic image to be repositioned is preliminarily registered to obtain a preliminarily registered image. The ICP algorithm is used to perform fracture surface registration processing on the pre-registered image to obtain a pelvic repositioning image.

5. The deep learning-based pelvic fracture reduction method according to claim 4, characterized in that, Based on the pelvic reduction reference image, the pelvic image to be reduced is preliminarily registered to obtain a preliminarily registered image, including: The image of the pelvis to be repositioned is segmented to obtain multiple fracture tissues; The position of the largest fracture tissue among the multiple fracture tissues is kept unchanged, and the remaining fracture tissues are aligned according to the pelvic reduction reference image to obtain a preliminary registration image.

6. The deep learning-based pelvic fracture reduction method according to claim 5, characterized in that, Using the ICP algorithm, the pre-registered image is subjected to fracture surface registration processing to obtain a pelvic repositioning image, including: Obtain the point cloud dataset for each fractured tissue in the image after initial registration; Based on the point cloud dataset of the fracture surfaces of every two adjacent fracture tissues, calculate the relevant point set of every two adjacent fracture tissues; Align the fracture surfaces of each pair of adjacent fracture tissues based on the relevant point set of each pair of adjacent fracture tissues. The image obtained after registration is output as a pelvic repositioning image.

7. A pelvic fracture reduction device based on deep learning, characterized in that, include: The acquisition module is used to acquire images of the pelvis to be repositioned and perform coarse alignment processing. The reference image generation module is used to obtain a pelvic repositioning reference image based on the trained pelvic repositioning model and the coarsely aligned pelvic image. The repositioning module is used to perform registration processing on the pelvic image to be repositioned using the ICP algorithm based on the pelvic repositioning reference image, so as to obtain a pelvic repositioning image.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.

9. A computer program product, characterized in that, The computer program product includes computer program code that, when run on a computer, causes the computer to implement the method as described in any one of claims 1 to 6.

10. An electronic terminal, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the method according to any one of claims 1 to 6.

Images