Space registration method for bone fracture navigation surgical instrument based on augmented reality technology

Abstract

According to the augmented reality technology-based spatial registration method for the fracture navigation surgical instrument, provided by the invention, the relationship between the actual fracture description condition and the sample fracture description condition covered by the fracture fragment distribution condition is analyzed by adopting at least two data analysis networks; and splicing the actual fracture description condition identification result and the actual fracture description condition relation distribution description result corresponding to each data analysis network obtained by analysis to construct a space registration description result corresponding to the fracture fragment description information set. Therefore, the fracture fragment distribution condition is processed by adopting multiple data analysis networks, and different data analysis networks are adapted to different types of data, so that the actual fracture description condition identification result and the actual fracture description condition relation distribution description result obtained through analysis are more accurate; therefore, recovery and registration can be accurately carried out according to each fragment, and medical staff can be assisted to complete work.

Description

Technical Field

[0001] The present application relates to the technical field of data registration, and in particular to a method for spatial registration of fracture navigation surgical equipment based on augmented reality technology. Background Art

[0002] There are many types of fractures. When encountering a comminuted fracture, the broken bones need to be found and restored, which brings great troubles and work pressure to medical staff. Therefore, a technical solution is continued to trace the broken bones to reduce the work pressure of medical workers. Summary of the invention

[0003] In order to improve the technical problems existing in the related technologies, the present application provides a method for spatial registration of fracture navigation surgical instruments based on augmented reality technology.

[0004] In a first aspect, a method for spatial registration of fracture navigation surgical instruments based on augmented reality technology is provided, the method comprising: Obtaining a description information set of fracture fragments covered by no less than one three-dimensional fracture data cluster; Extracting the distribution of fracture fragments in the fracture fragment description information set, wherein the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description; Using no less than two data analysis networks to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, to obtain the actual fracture description recognition result and the actual fracture description relationship distribution description result corresponding to each data analysis network; The actual fracture description situation recognition results and the actual fracture description situation relationship distribution description results corresponding to each data analysis network are spliced, and the spatial registration description results corresponding to the fracture fragment description information set are constructed based on the splicing results.

[0005] In the present application, the at least two data analysis networks include a first data analysis network; the at least two data analysis networks are used to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, and the actual fracture description recognition result and the actual fracture description relationship distribution description result corresponding to each data analysis network are obtained, including: Obtaining a predetermined association network, wherein the association network is generated based on the relationship between the actual fracture descriptions covered by the historical fracture fragment distribution and the sample fracture descriptions; The fracture fragment distribution is associated with the association network to obtain a first actual fracture description recognition result and a first actual fracture description relationship distribution description result corresponding to the first data analysis network.

[0006] In the present application, the bone fracture fragment distribution is associated with the association network to obtain a first actual bone fracture description recognition result and a first actual bone fracture description relationship distribution description result corresponding to the first data analysis network, including: Associating the fracture fragment distribution with the text processing rules in the association network, and determining the actual fracture description that meets the text processing rules as the first actual fracture description recognition result; The fracture fragment distribution is associated with the attribute analysis rules in the association network, and the actual fracture description relationship that meets the attribute analysis rules is determined as the first actual fracture description relationship distribution description result.

[0007] In the present application, the at least two data analysis networks include a second data analysis network; the at least two data analysis networks are used to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, and the actual fracture description identification result and the actual fracture description relationship distribution description result corresponding to each data analysis network are obtained, and also include: The artificial intelligence thread is obtained according to the relationship between the actual fracture descriptions marked in the example data and the sample fracture descriptions; The fracture fragment distribution is loaded into the artificial intelligence thread, and the actual fracture description and the relationship between the sample fracture description covered in the fracture fragment distribution are predicted by the artificial intelligence thread to obtain a second actual fracture description recognition result and a second actual fracture description relationship distribution description result corresponding to the second data analysis network.

[0008] In the present application, the artificial intelligence thread includes a first artificial intelligence thread and a second artificial intelligence thread; the loading of the fracture fragment distribution into the artificial intelligence thread predicts the actual fracture description situation and the sample fracture description situation relationship covered in the fracture fragment distribution situation, and obtains a second actual fracture description situation recognition result and a second actual fracture description situation relationship distribution description result corresponding to the second data analysis network, including: The fracture fragment distribution is loaded into the first artificial intelligence thread, and the actual fracture description in the fracture fragment distribution is identified by the first artificial intelligence thread to obtain the second actual fracture description identification result; The fracture fragment distribution is loaded into the second artificial intelligence thread, and the actual fracture description relationship in the fracture fragment distribution is extracted by the second artificial intelligence thread to obtain the second actual fracture description relationship distribution description result.

[0009] In the present application, the at least two data analysis networks include a third data analysis network; the at least two data analysis networks are used to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, and the actual fracture description identification result and the actual fracture description relationship distribution description result corresponding to each data analysis network are obtained, and also include: According to the actual fracture description of each range and the relationship between the sample fracture description, a bone distribution database corresponding to each range is established; The fracture fragment distribution is projected with the actual fracture description and the sample fracture description relationship in the bone distribution database to obtain a third actual fracture description recognition result and a third actual fracture description relationship distribution description result corresponding to the third data analysis network.

[0010] In the present application, the fracture fragment distribution is projected with the actual fracture description in the bone distribution database and the sample fracture description relationship to obtain a third actual fracture description recognition result and a third actual fracture description relationship distribution description result corresponding to the third data analysis network, including: Searching the bone distribution database for an actual fracture description corresponding to the fracture fragment distribution, and determining the searched actual fracture description as the third actual fracture description recognition result; The actual fracture description situation relationship corresponding to the fracture fragment distribution is searched from the bone distribution database, and the searched actual fracture description situation relationship is determined as the third actual fracture description situation relationship distribution description result.

[0011] In the present application, if the actual fracture description corresponding to the fracture fragment distribution is searched from the bone distribution database, and the searched actual fracture description is determined as the third actual fracture description recognition result, it includes: If there are multiple bone distribution databases, and the actual fracture description in the fracture fragment distribution corresponds to the actual fracture description in the multiple bone distribution databases, the most relevant bone distribution database is screened from the multiple bone distribution databases, and the actual fracture description in the most relevant bone distribution database is determined as the third actual fracture description identification result.

[0012] In the present application, the actual fracture description situation recognition results and the actual fracture description situation relationship distribution description results corresponding to each data analysis network are spliced, and the spatial registration description results corresponding to the fracture fragment description information set are constructed based on the splicing results, including: Integrating the same actual fracture descriptions in the actual fracture description recognition results corresponding to each data analysis network, and fusing the integrated actual fracture descriptions with other actual fracture descriptions that have not been integrated to obtain a final actual fracture description recognition result; Integrating the same actual fracture description situation relationships in the actual fracture description situation relationship distribution description results corresponding to each data analysis network, and fusing the integrated actual fracture description situation relationships with other actual fracture description situation relationships that have not been integrated, to obtain a final actual fracture description situation relationship distribution description result; According to the final actual fracture description situation recognition result and the final actual fracture description situation relationship distribution description result, a spatial registration description result corresponding to the fracture fragment description information set is constructed.

[0013] In the present application, the spatial registration description result corresponding to the fracture fragment description information set is constructed according to the final actual fracture description recognition result and the final actual fracture description relationship distribution description result, including: Determining the actual fracture description in the final actual fracture description recognition result as an attribute of the network structure; Determine the relationship between different attributes according to the actual fracture description relationship in the final actual fracture description relationship distribution description result, and determine the constraint conditions of the network structure; The attributes of the network structure and the constraints of the network structure are integrated to obtain the spatial registration description result.

[0014] In the present application, the method further comprises: Extracting new actual fracture description situation recognition results and new relationship distribution description results from other three-dimensional fracture data clusters other than the at least one three-dimensional fracture data cluster; The new actual fracture description recognition result and the new relationship distribution description result are fused with the spatial registration description result to obtain an updated spatial registration description result.

[0015] In a second aspect, a fracture navigation surgical equipment spatial registration system based on augmented reality technology 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.

[0016] The fracture navigation surgical equipment spatial registration method based on augmented reality technology provided by the embodiment of the present application makes the obtained fracture fragment description information set richer and more comprehensive by obtaining the fracture fragment description information set covered by no less than one three-dimensional fracture data cluster. After obtaining the fracture fragment description information set, the fracture fragment distribution in the fracture fragment description information set is extracted, and then, no less than two data analysis networks are used to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, and the actual fracture description recognition result and the actual fracture description relationship distribution description result corresponding to each data analysis network obtained by the analysis are spliced ​​to build a spatial registration description result corresponding to the fracture fragment description information set. In this way, by using multiple data analysis networks to process the fracture fragment distribution, different data analysis networks adapt to different types of data, so that the actual fracture description recognition result and the actual fracture description relationship distribution description result obtained by the analysis are more accurate, so that each fragment can be accurately restored and registered, and medical staff can be assisted to complete their work. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] 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.

[0018] Figure 1 A flowchart of a method for spatial registration of fracture navigation surgical instruments based on augmented reality technology provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] 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.

[0020] See also Figure 1 , shows a method for spatial registration of fracture navigation surgical equipment based on augmented reality technology, which may include the technical solutions described in the following steps S201-S204.

[0021] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0022] Among them, the three-dimensional fracture data cluster can be obtained through CT and ultrasound. CT is equivalent to an image, and ultrasound is data information obtained by reflection.

[0023] In some optional embodiments, when obtaining the fracture fragment description information set included in at least one three-dimensional fracture data cluster, at least one three-dimensional fracture data cluster may be obtained first, and then the fracture fragment description information may be extracted from the three-dimensional fracture data cluster, thereby obtaining the fracture fragment description information set. Wherein, at least one three-dimensional fracture data cluster refers to including two or more three-dimensional fracture data clusters, and each three-dimensional fracture data cluster is independent of each other.

[0024] Optionally, after obtaining multiple three-dimensional fracture data clusters, a preprocessing operation may be performed on the three-dimensional fracture data clusters, wherein the preprocessing operation includes cleaning, classification, and identification of actual fracture description conditions. In this way, after obtaining multiple three-dimensional fracture data clusters, preprocessing the source data may be performed to facilitate extraction of the fracture fragment description information set covered in the three-dimensional fracture data clusters, thereby improving data processing efficiency.

[0025] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0026] Exemplarily, the distribution of bone fracture fragments may be understood as the distribution of bone fracture fragments in the body.

[0027] After obtaining the fracture fragment description information set, the fracture fragment distribution in the fracture fragment description information set is extracted. The method of extracting the fracture fragment distribution may include keyword extraction, actual fracture description identification, attribute extraction, etc. In this way, by extracting the fracture fragment distribution in the fracture fragment description information set, useful information can be extracted from the fracture fragment description information set to avoid interference from other useless information. Moreover, extracting the fracture fragment distribution from the fracture fragment description information set reduces the amount of subsequent data processing to a certain extent, and reduces the processing of the data volume.

[0028] In S203, at least two data analysis networks are used to analyze the relationship between the actual fracture description situations covered by the fracture fragment distribution and the sample fracture description situations, and obtain the actual fracture description situation recognition results and the actual fracture description situation relationship distribution description results corresponding to each data analysis network.

[0029] Exemplarily, the data parsing network can be one of CNN and the like, which is used to analyze fracture data in the present application.

[0030] In some optional embodiments, the data analysis network is used to analyze the distribution of fracture fragments to obtain the actual fracture description covered by the distribution of fracture fragments and the relationship between the sample fracture description.

[0031] At least two data analysis networks refer to two or more data analysis networks, and at least two data analysis networks are used to analyze the distribution of fracture fragments, so as to obtain the actual fracture description recognition result and the actual fracture description relationship distribution description result corresponding to each data analysis network. For example, assuming that the data analysis network used includes a first data analysis network and a second data analysis network, the first data analysis network is used to process the distribution of fracture fragments, and the first actual fracture description recognition result and the first actual fracture description relationship distribution description result corresponding to the first data analysis network are obtained, and the second data analysis network is used to process the distribution of fracture fragments, and the second actual fracture description recognition result and the second actual fracture description relationship distribution description result corresponding to the second data analysis network are obtained. In this way, by using different data analysis networks to analyze the distribution of fracture fragments, the obtained actual fracture description recognition result and the actual fracture description relationship distribution description result can be made more accurate, avoiding the problem of inaccurate data caused by using a single data analysis network to analyze the distribution of fracture fragments.

[0032] In S204, the actual fracture description situation recognition results and the actual fracture description situation relationship distribution description results corresponding to each data analysis network are spliced, and a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing results.

[0033] Illustratively, the spatial registration description results in determining how the fractured bone fragments are put back into their original positions.

[0034] In some optional embodiments, after obtaining the actual fracture description situation recognition result and the actual fracture description situation relationship distribution description result corresponding to each data analysis network, the actual fracture description situation recognition result and the actual fracture description situation relationship distribution description result corresponding to each data analysis network are spliced, and then the spatial registration description result corresponding to the fracture fragment description information set is built based on the splicing result. For example, assuming that two data analysis networks are used to process the fracture fragment distribution, and the first actual fracture description situation recognition result and the first actual fracture description situation relationship distribution description result corresponding to the first data analysis network are obtained, and the second actual fracture description situation recognition result and the second actual fracture description situation relationship distribution description result corresponding to the second data analysis network are obtained, then it is necessary to splice the first actual fracture description situation recognition result and the second actual fracture description situation recognition result to obtain the final actual fracture description situation recognition result, splice the first actual fracture description situation relationship distribution description result and the second actual fracture description situation relationship distribution description result to obtain the final actual fracture description situation relationship distribution description result, and then build the spatial registration description result based on the splicing of the final actual fracture description situation recognition result and the final actual fracture description situation relationship distribution description result. In this way, by first splicing the actual fracture description recognition results corresponding to each data analysis network and the actual fracture description relationship distribution description results, and then building the spatial registration description results based on the splicing results, it is beneficial to build a more accurate spatial registration description result.

[0035] In the technical solution provided in the embodiment of the present application, by obtaining a fracture fragment description information set covered by no less than one three-dimensional fracture data cluster, the obtained fracture fragment description information set is made richer and more comprehensive. After obtaining the fracture fragment description information set, the fracture fragment distribution in the fracture fragment description information set is extracted, and then, no less than two data analysis networks are used to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, and the actual fracture description recognition result and the actual fracture description relationship distribution description result corresponding to each data analysis network obtained by the analysis are spliced ​​to build a spatial registration description result corresponding to the fracture fragment description information set. In this way, by using multiple data analysis networks to process the fracture fragment distribution, different data analysis networks adapt to different types of data, so that the actual fracture description recognition result and the actual fracture description relationship distribution description result obtained by the analysis are more accurate, so that each fragment can be accurately restored and registered, and medical staff can be assisted to complete their work.

[0036] In some optional embodiments, a method for spatial registration of fracture navigation surgery equipment based on augmented reality technology according to an embodiment of the present application may mainly include the following steps.

[0037] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0038] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0039] In S301, a predetermined association network is obtained, where the association network is generated based on the relationship between the actual fracture descriptions covered by the historical fracture fragment distribution and the sample fracture descriptions.

[0040] In some optional embodiments, at least two data analysis networks include a first data analysis network, and the first data analysis network is specifically: when analyzing the distribution of fracture fragments, the fracture fragment distribution can be analyzed through a predetermined association network, and the predetermined association network can also be considered as analyzing the fracture fragment distribution through preset rules. When using this data analysis network, specifically, a series of rules can be generated based on the actual fracture description covered by the historical fracture fragment distribution and the relationship between the sample fracture description.

[0041] In S302, the distribution of fracture fragments is associated with the association network to obtain a first actual fracture description recognition result and a first actual fracture description relationship distribution description result corresponding to the first data analysis network.

[0042] In some optional embodiments, after obtaining the predetermined association network, the distribution of fracture fragments is associated with the association network, so as to obtain the first actual fracture description situation recognition result and the first actual fracture description situation relationship distribution description result corresponding to the first data analysis network. The first actual fracture description situation recognition result covers each actual fracture description situation, and the first actual fracture description situation relationship distribution description result covers the relationship between the actual fracture description situation and the actual fracture description situation.

[0043] In S303, the first actual fracture description situation recognition result is spliced ​​with the actual fracture description situation recognition results corresponding to other data analysis networks, and the first actual fracture description situation relationship distribution description result is spliced ​​with the actual fracture description situation relationship distribution description results corresponding to other data analysis networks, and a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing result.

[0044] In some optional embodiments, after obtaining the first actual fracture description recognition result and the first actual fracture description relationship distribution description result, the first actual fracture description recognition result and the actual fracture description recognition result corresponding to other data analysis networks, and the first actual fracture description relationship distribution description result and the actual fracture description relationship distribution description result corresponding to other data analysis networks are spliced, and then a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing result. Among them, the other data analysis networks can be, for example, the second data analysis network and the third data analysis network.

[0045] In this way, the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description is analyzed by adopting the rule association method, which is conducive to extracting the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description. This method is simple to implement and can accurately capture the target information.

[0046] In some optional embodiments, a method for spatial registration of fracture navigation surgery equipment based on augmented reality technology according to an embodiment of the present application may mainly include the following steps.

[0047] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0048] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0049] In S301, a predetermined association network is obtained, where the association network is generated based on the relationship between the actual fracture descriptions covered by the historical fracture fragment distribution and the sample fracture descriptions.

[0050] In S401, the distribution of fracture fragments is associated with the text processing rules in the association network, and the actual fracture description that meets the text processing rules is determined as the first actual fracture description recognition result.

[0051] In S402, the distribution of fracture fragments is associated with the attribute analysis rules in the association network, and the actual fracture description relationship that meets the attribute analysis rules is determined as the first actual fracture description relationship distribution description result.

[0052] In some optional embodiments, after generating an association network and preprocessing the fracture fragment description information set, the association network can be applied to the text in the fracture fragment description information set. This process can generally include pattern association, rule triggering, etc. For example, text processing rules can be used to associate actual fracture descriptions, and attribute analysis rules can be used to describe the relationships between actual fracture descriptions. Finally, actual fracture descriptions and relationships that meet the rules can be extracted from the text based on the association network.

[0053] In S303, the first actual fracture description situation recognition result is spliced ​​with the actual fracture description situation recognition results corresponding to other data analysis networks, and the first actual fracture description situation relationship distribution description result is spliced ​​with the actual fracture description situation relationship distribution description results corresponding to other data analysis networks, and a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing result.

[0054] In this way, the relationship between the actual fracture descriptions covered by the fracture fragment distribution and the sample fracture descriptions is analyzed by adopting the rule association method. Specifically, the actual fracture descriptions are analyzed by setting text processing rules, and the relationship between the actual fracture descriptions is analyzed by attribute analysis rules. Different rules are set specifically for the analysis of the relationship between the actual fracture descriptions and the sample fracture descriptions, so that the relationship between the actual fracture descriptions and the sample fracture descriptions obtained by analysis is more accurate.

[0055] In some optional embodiments, a method for spatial registration of fracture navigation surgery equipment based on augmented reality technology according to an embodiment of the present application may mainly include the following steps.

[0056] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0057] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0058] In S501, an artificial intelligence thread is obtained according to the relationship configuration between the actual fracture descriptions marked in the example data and the sample fracture descriptions.

[0059] In some optional embodiments, no less than two data analysis networks include a second data analysis network, and the second data analysis network is specifically: when analyzing the distribution of fracture fragments, the distribution of fracture fragments is analyzed by machine learning. Specifically, supervised learning, unsupervised learning or semi-supervised learning methods can be used to analyze the distribution of fracture fragments to identify the relationship between the actual fracture description and the sample fracture description. When machine learning is used to analyze the distribution of fracture fragments, these methods usually require a configuration data set for configuring an artificial intelligence thread. The configuration process of the artificial intelligence thread is specifically to configure the artificial intelligence thread according to the relationship between the actual fracture description marked in the example data and the sample fracture description. In this way, by using a machine learning method to analyze the distribution of fracture fragments, it can automatically adapt to different types of data and scenarios, and has good generalization performance.

[0060] In S502, the fracture fragment distribution is loaded into the artificial intelligence thread, and the actual fracture description situation and the relationship between the sample fracture description situation covered in the fracture fragment distribution situation are predicted through the artificial intelligence thread to obtain a second actual fracture description situation recognition result and a second actual fracture description situation relationship distribution description result corresponding to the second data analysis network.

[0061] In some optional embodiments, after obtaining the pre-configured artificial intelligence thread, the distribution of fracture fragments is loaded into the artificial intelligence thread, and the distribution of fracture fragments is parsed by using a machine learning method, thereby obtaining a second actual fracture description recognition result and a second actual fracture description relationship distribution description result corresponding to the second data parsing network. The second actual fracture description recognition result covers each actual fracture description, and the second actual fracture description relationship distribution description result covers the relationship between the actual fracture description and the actual fracture description.

[0062] In S503, the second actual fracture description recognition result is spliced ​​with the actual fracture description recognition result corresponding to other data analysis networks, and the second actual fracture description relationship distribution description result is spliced ​​with the actual fracture description relationship distribution description result corresponding to other data analysis networks, and a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing result.

[0063] In some optional embodiments, after obtaining the second actual fracture description recognition result and the second actual fracture description relationship distribution description result, the second actual fracture description recognition result is spliced ​​with the actual fracture description recognition result corresponding to other data analysis networks, and the second actual fracture description relationship distribution description result is spliced ​​with the actual fracture description relationship distribution description result corresponding to other data analysis networks, and then a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing result. Among them, the other data analysis networks can be, for example, the first data analysis network and the third data analysis network.

[0064] In this way, by using machine learning methods to analyze the distribution of fracture fragments, this method can automatically adapt to different types of data and scenarios and has good generalization performance.

[0065] In some optional embodiments, a method for spatial registration of fracture navigation surgery equipment based on augmented reality technology according to an embodiment of the present application may mainly include the following steps.

[0066] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0067] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0068] In S501, an artificial intelligence thread is obtained according to the relationship configuration between the actual fracture descriptions marked in the example data and the sample fracture descriptions.

[0069] In S601, the fracture fragment distribution is loaded into the first artificial intelligence thread, and the actual fracture description in the fracture fragment distribution is identified by the first artificial intelligence thread to obtain a second actual fracture description identification result.

[0070] In S602, the fracture fragment distribution is loaded into the second artificial intelligence thread, and the second artificial intelligence thread extracts the relationship between the actual fracture description situation in the fracture fragment distribution to obtain a second actual fracture description situation relationship distribution description result.

[0071] In some optional embodiments, a second artificial intelligence thread can be used to extract the actual fracture description relationship covered by the fracture fragment distribution, and the second artificial intelligence thread can use a relational classification model, such as a convolutional neural network (CNN), a recurrent neural network (RNN), a graph neural network (GNN), etc.

[0072] In S503, the second actual fracture description recognition result is spliced ​​with the actual fracture description recognition result corresponding to other data analysis networks, and the second actual fracture description relationship distribution description result is spliced ​​with the actual fracture description relationship distribution description result corresponding to other data analysis networks, and a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing result.

[0073] In this way, by using the first artificial intelligence thread to identify the actual fracture description covered in the fracture fragment distribution, and using the second artificial intelligence thread to extract the relationship between the actual fracture description covered in the fracture fragment distribution, the actual fracture description identification and relationship extraction are analyzed using different artificial intelligence threads respectively, so that the second actual fracture description identification result and the second actual fracture description relationship distribution description result can be more accurate.

[0074] In some optional embodiments, a method for spatial registration of fracture navigation surgery equipment based on augmented reality technology according to an embodiment of the present application may mainly include the following steps.

[0075] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0076] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0077] In S701, a bone distribution database corresponding to each range is built according to the actual fracture description of each range and the relationship between the sample fracture description.

[0078] In some optional embodiments, at least two data analysis networks include a third data analysis network, and the third data analysis is specifically to analyze the distribution of fracture fragments by means of a bone distribution database when analyzing the distribution of fracture fragments. When the third data analysis network is used to analyze the distribution of fracture fragments, a bone distribution database needs to be built in advance. The building of the bone distribution database can be to build a bone distribution database corresponding to each range based on the actual fracture description of each range and the relationship between the sample fracture description.

[0079] It is understandable that the bone distribution database can also use the existing bone distribution database to analyze the distribution of fracture fragments to identify the actual fracture description and its relationship. These bone distribution databases usually contain rich information on actual fracture description and relationship, and therefore can help improve the accuracy of identifying the actual fracture description and finding the relationship between the actual fracture description.

[0080] In S702, the distribution of fracture fragments is projected with the actual fracture description in the bone distribution database and the relationship between the sample fracture description to obtain a third actual fracture description recognition result and a third actual fracture description relationship distribution description result corresponding to the third data analysis network.

[0081] In some optional embodiments, after pre-building a bone distribution database, the distribution of fracture fragments is projected with the actual fracture description in the bone distribution database and the relationship between the sample fracture description to obtain a third actual fracture description recognition result and a third actual fracture description relationship distribution description result corresponding to the third data analysis network.

[0082] In S703, the third actual fracture description situation recognition result is spliced ​​with the actual fracture description situation recognition results corresponding to other data analysis networks, and the third actual fracture description situation relationship distribution description result is spliced ​​with the actual fracture description situation relationship distribution description results corresponding to other data analysis networks, and a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing result.

[0083] In some optional embodiments, after obtaining the third actual fracture description recognition result and the third actual fracture description relationship distribution description result, the third actual fracture description recognition result is spliced ​​with the actual fracture description recognition result corresponding to other data analysis networks, and the third actual fracture description relationship distribution description result is spliced ​​with the actual fracture description relationship distribution description result corresponding to other data analysis networks, and then a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing result. Among them, the other data analysis networks can be, for example, the first data analysis network and the second data analysis network.

[0084] In this way, since the bone distribution database usually covers a wealth of actual fracture descriptions and sample fracture description relationships, by projecting the fracture fragment distribution with the actual fracture descriptions and sample fracture description relationships in the bone distribution database, it is helpful to improve the accuracy of actual fracture description identification and relationship discovery.

[0085] In some optional embodiments, a method for spatial registration of fracture navigation surgery equipment based on augmented reality technology according to an embodiment of the present application may mainly include the following steps.

[0086] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0087] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0088] In S701, a bone distribution database corresponding to each range is built according to the actual fracture description of each range and the relationship between the sample fracture description.

[0089] In S801, the actual fracture description corresponding to the fracture fragment distribution is searched from the bone distribution database, and the searched actual fracture description is determined as the third actual fracture description recognition result.

[0090] In S802, the actual fracture description situation relationship corresponding to the fracture fragment distribution is searched from the bone distribution database, and the searched actual fracture description situation relationship is determined as the third actual fracture description situation relationship distribution description result.

[0091] In some optional embodiments, in order to utilize the information in the bone distribution database, it is necessary to align and map the data to be processed with the actual fracture description and relationship in the bone distribution database. This process generally includes actual fracture description linking, relationship mapping, etc. The purpose of the actual fracture description linking is to find the corresponding actual fracture description of the actual fracture description in the data to be processed in the bone distribution database, and the purpose of the relationship mapping is to find the corresponding relationship of the relationship in the data to be processed in the bone distribution database. Specifically, the actual fracture description linking and relationship mapping can be implemented by string association, similarity calculation, machine learning and other methods. For example, the actual fracture description in the data to be processed can be compared with the actual fracture description in the bone distribution database using a string association method, and the most similar actual fracture description is found to be determined as the link result. Finally, the third actual fracture description recognition result and the third actual fracture description relationship distribution description result are obtained, and the third actual fracture description recognition result covers each actual fracture description, and the third actual fracture description relationship distribution description result covers the relationship between the actual fracture description and the actual fracture description.

[0092] In S703, the third actual fracture description situation recognition result is spliced ​​with the actual fracture description situation recognition results corresponding to other data analysis networks, and the third actual fracture description situation relationship distribution description result is spliced ​​with the actual fracture description situation relationship distribution description results corresponding to other data analysis networks, and a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing result.

[0093] In this way, since the bone distribution database usually covers a wealth of actual fracture descriptions and sample fracture description relationships, by projecting the fracture fragment distribution with the actual fracture descriptions and sample fracture description relationships in the bone distribution database, it is helpful to improve the accuracy of actual fracture description identification and relationship discovery.

[0094] In some optional embodiments, a method for spatial registration of fracture navigation surgery equipment based on augmented reality technology according to an embodiment of the present application may mainly include the following steps.

[0095] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0096] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0097] In S701, a bone distribution database corresponding to each range is built according to the actual fracture description of each range and the relationship between the sample fracture description.

[0098] In S901, if there are multiple bone distribution databases, and the actual fracture description in the fracture fragment distribution corresponds to the actual fracture description in multiple bone distribution databases, the most relevant bone distribution database is screened from the multiple bone distribution databases, and the actual fracture description in the most relevant bone distribution database is determined as the third actual fracture description identification result.

[0099] In some optional embodiments, the process of linking the actual fracture description may cause ambiguity, that is, the actual fracture description in one to-be-processed data may correspond to the actual fracture description in multiple bone distribution databases. In order to ensure the accuracy of the actual fracture description recognition result, it is necessary to disambiguate the actual fracture description. The method of disambiguating the actual fracture description may include context-based similarity calculation, classification-based method, network structure-based method, etc. These methods are aimed at finding the actual fracture description in the most relevant bone distribution database.

[0100] In S802, the actual fracture description situation relationship corresponding to the fracture fragment distribution is searched from the bone distribution database, and the searched actual fracture description situation relationship is determined as the third actual fracture description situation relationship distribution description result.

[0101] In S703, the third actual fracture description situation recognition result is spliced ​​with the actual fracture description situation recognition results corresponding to other data analysis networks, and the third actual fracture description situation relationship distribution description result is spliced ​​with the actual fracture description situation relationship distribution description results corresponding to other data analysis networks, and a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing result.

[0102] In this way, by screening the most relevant bone distribution database from multiple bone distribution databases, the interference of actual fracture descriptions of other bone distribution databases can be avoided, thereby improving the accuracy of the actual fracture description recognition results.

[0103] In some optional embodiments, a method for spatial registration of fracture navigation surgery equipment based on augmented reality technology according to an embodiment of the present application may mainly include the following steps.

[0104] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0105] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0106] In S203, at least two data analysis networks are used to analyze the relationship between the actual fracture description situations covered by the fracture fragment distribution and the sample fracture description situations, and obtain the actual fracture description situation recognition results and the actual fracture description situation relationship distribution description results corresponding to each data analysis network.

[0107] In S1001, the same actual fracture description situations in the actual fracture description situation recognition results corresponding to each data analysis network are integrated, and the integrated actual fracture description situations are fused with other actual fracture description situations that have not been integrated to obtain the final actual fracture description situation recognition results.

[0108] In some optional embodiments, after obtaining the actual fracture description recognition result corresponding to each data analysis network, the same actual fracture description is integrated, and then the integrated actual fracture description is fused with other actual fracture descriptions that have not been integrated, so as to obtain the final actual fracture description recognition result. For example, assuming that two data analysis networks are used, and the first actual fracture description recognition result corresponding to the first data analysis network and the second actual fracture description recognition result corresponding to the second data analysis network are obtained respectively, when the first actual fracture description recognition result and the second actual fracture description recognition result are integrated, the same actual fracture description in the two recognition results is integrated, and then the integrated actual fracture description is obtained. The integrated actual fracture description and other actual fracture descriptions that have not been integrated are fused to obtain the final actual fracture description recognition result.

[0109] In S1002, the same actual fracture description situation relationships in the actual fracture description situation relationship distribution description results corresponding to each data analysis network are integrated, and the integrated actual fracture description situation relationships are fused with other actual fracture description situation relationships that have not been integrated to obtain the final actual fracture description situation relationship distribution description results.

[0110] In some optional embodiments, after obtaining the actual fracture description situation relationship distribution description result corresponding to each data analysis network, the same actual fracture description situation relationship is integrated, and then the integrated actual fracture description situation relationship is fused with other actual fracture description situation relationships that have not been integrated, so as to obtain the final actual fracture description situation relationship distribution description result. For example, assuming that two data analysis networks are used, and the first actual fracture description situation relationship distribution description result corresponding to the first data analysis network and the second actual fracture description situation relationship distribution description result corresponding to the second data analysis network are obtained respectively, when the first actual fracture description situation relationship distribution description result and the second actual fracture description situation relationship distribution description result are integrated, the same actual fracture description situation relationship in the two actual fracture description situation relationship distribution description results is integrated, and then the integrated actual fracture description situation relationship is obtained. The integrated actual fracture description situation relationship is fused with other actual fracture description situation relationships that have not been integrated, so as to obtain the final actual fracture description situation relationship distribution description result.

[0111] In S1003, a spatial registration description result corresponding to the fracture fragment description information set is constructed according to the final actual fracture description recognition result and the final actual fracture description relationship distribution description result.

[0112] In this way, by first splicing the actual fracture description recognition results corresponding to each data analysis network and the actual fracture description relationship distribution description results, and then building the spatial registration description results based on the splicing results, it is beneficial to build a more comprehensive and accurate spatial registration description result.

[0113] In some optional embodiments, a method for spatial registration of fracture navigation surgery equipment based on augmented reality technology according to an embodiment of the present application may mainly include the following steps.

[0114] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0115] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0116] In S203, at least two data analysis networks are used to analyze the relationship between the actual fracture description situations covered by the fracture fragment distribution and the sample fracture description situations, and obtain the actual fracture description situation recognition results and the actual fracture description situation relationship distribution description results corresponding to each data analysis network.

[0117] In S1001, the same actual fracture description situations in the actual fracture description situation recognition results corresponding to each data analysis network are integrated, and the integrated actual fracture description situations are fused with other actual fracture description situations that have not been integrated to obtain the final actual fracture description situation recognition results.

[0118] In S1002, the same actual fracture description situation relationships in the actual fracture description situation relationship distribution description results corresponding to each data analysis network are integrated, and the integrated actual fracture description situation relationships are fused with other actual fracture description situation relationships that have not been integrated to obtain the final actual fracture description situation relationship distribution description results.

[0119] In this way, by using a variety of data analysis networks to process the distribution of fracture fragments, different data analysis networks adapt to different types of data, so that the actual fracture description identification results and the actual fracture description relationship distribution description results obtained by analysis are more accurate, so that they can accurately restore and align each fragment to assist medical staff in completing their work.

[0120] In some optional embodiments, a method for spatial registration of fracture navigation surgery equipment based on augmented reality technology according to an embodiment of the present application may mainly include the following steps.

[0121] In S201, a fracture fragment description information set covered by at least one three-dimensional fracture data cluster is obtained.

[0122] In S202, the distribution of fracture fragments in the fracture fragment description information set is extracted, and the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description.

[0123] In S203, at least two data analysis networks are used to analyze the relationship between the actual fracture description situations covered by the fracture fragment distribution and the sample fracture description situations, and obtain the actual fracture description situation recognition results and the actual fracture description situation relationship distribution description results corresponding to each data analysis network.

[0124] In S204, the actual fracture description situation recognition results and the actual fracture description situation relationship distribution description results corresponding to each data analysis network are spliced, and a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing results.

[0125] In S1401, new actual fracture description situation recognition results and new relationship distribution description results are extracted from other three-dimensional fracture data clusters other than at least one three-dimensional fracture data cluster.

[0126] In S1402, the new actual fracture description recognition result and the new relationship distribution description result are fused with the spatial registration description result to obtain an updated spatial registration description result.

[0127] In some optional embodiments, when the spatial registration description result is updated, the incremental learning and text mining methods can be used to automatically extract knowledge points from the new fracture fragment description information and update the spatial registration description result. Specifically, a new fracture fragment description information set is first extracted from other three-dimensional fracture data clusters other than at least one three-dimensional fracture data cluster, and then the actual fracture description situation and the actual fracture description situation relationship are obtained from the new fracture fragment description information set, and preprocessed. Then, the actual fracture description situation in the text is identified using natural language processing technology, and it is marked as a known named actual fracture description situation. The new actual fracture description situation and the new actual fracture description situation relationship are extracted from the text using natural language processing and machine learning technology, and the new actual fracture description situation is represented as an attribute of the spatial registration description result, and the new actual fracture description situation relationship is represented as an edge in the spatial registration description result. Finally, the new actual fracture description situation and the new actual fracture description situation relationship are added to the spatial registration description result to obtain an updated spatial registration description result.

[0128] In this way, incremental learning and text mining methods are used to automatically extract new actual fracture descriptions and new actual fracture description relationships from new fracture fragment description information and update spatial registration description results, thereby maintaining the real-time and comprehensiveness of spatial registration description results. Based on the above, a fracture navigation surgical equipment spatial registration system based on augmented reality technology 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.

[0129] Based on the above, a computer-readable storage medium is also provided, on which a computer program stored implements the above method when running.

[0130] In summary, based on the above scheme, by obtaining a fracture fragment description information set covered by no less than one three-dimensional fracture data cluster, the obtained fracture fragment description information set is made richer and more comprehensive. After obtaining the fracture fragment description information set, the fracture fragment distribution in the fracture fragment description information set is extracted, and then, no less than two data analysis networks are used to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, and the actual fracture description recognition results and the actual fracture description relationship distribution description results corresponding to each data analysis network obtained by the analysis are spliced ​​to build a spatial registration description result corresponding to the fracture fragment description information set. In this way, by using multiple data analysis networks to process the fracture fragment distribution, different data analysis networks adapt to different types of data, so that the actual fracture description recognition results and the actual fracture description relationship distribution description results obtained by the analysis are more accurate, so that each fragment can be accurately restored and registered, and medical staff can be assisted in completing their work.

[0131] 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).

[0132] 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 spatial registration of fracture navigation surgical equipment based on augmented reality technology, characterized in that: The method comprises: Obtaining a description information set of fracture fragments covered by no less than one three-dimensional fracture data cluster; Extracting the distribution of fracture fragments in the fracture fragment description information set, wherein the distribution of fracture fragments includes the relationship between the actual fracture description and the sample fracture description; Using no less than two data analysis networks to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, to obtain the actual fracture description recognition result and the actual fracture description relationship distribution description result corresponding to each data analysis network; The actual fracture description situation recognition results and the actual fracture description situation relationship distribution description results corresponding to each data analysis network are spliced, and the spatial registration description results corresponding to the fracture fragment description information set are constructed based on the splicing results.

2. The method for spatial registration of fracture navigation surgical instruments based on augmented reality technology according to claim 1, characterized in that: The at least two data analysis networks include a first data analysis network; the at least two data analysis networks are used to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, and the actual fracture description recognition result and the actual fracture description relationship distribution description result corresponding to each data analysis network are obtained, including: Obtaining a predetermined association network, wherein the association network is generated based on the relationship between the actual fracture descriptions covered by the historical fracture fragment distribution and the sample fracture descriptions; The fracture fragment distribution is associated with the association network to obtain a first actual fracture description recognition result and a first actual fracture description relationship distribution description result corresponding to the first data analysis network.

3. The method for spatial registration of fracture navigation surgical instruments based on augmented reality technology according to claim 2, characterized in that: The step of associating the fracture fragment distribution with the association network to obtain a first actual fracture description recognition result and a first actual fracture description relationship distribution description result corresponding to the first data analysis network includes: Associating the fracture fragment distribution with the text processing rules in the association network, and determining the actual fracture description that meets the text processing rules as the first actual fracture description recognition result; The fracture fragment distribution is associated with the attribute analysis rules in the association network, and the actual fracture description relationship that meets the attribute analysis rules is determined as the first actual fracture description relationship distribution description result.

4. The method for spatial registration of fracture navigation surgical instruments based on augmented reality technology according to claim 1, characterized in that: The at least two data analysis networks include a second data analysis network; the at least two data analysis networks are used to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, and the actual fracture description recognition result and the actual fracture description relationship distribution description result corresponding to each data analysis network are obtained, and the method also includes: The artificial intelligence thread is obtained according to the relationship between the actual fracture descriptions marked in the example data and the sample fracture descriptions; The fracture fragment distribution is loaded into the artificial intelligence thread, and the actual fracture description and the relationship between the sample fracture description covered in the fracture fragment distribution are predicted by the artificial intelligence thread to obtain a second actual fracture description recognition result and a second actual fracture description relationship distribution description result corresponding to the second data analysis network.

5. The method for spatial registration of fracture navigation surgical instruments based on augmented reality technology according to claim 4, characterized in that: The artificial intelligence thread includes a first artificial intelligence thread and a second artificial intelligence thread; the fracture fragment distribution is loaded into the artificial intelligence thread to predict the actual fracture description situation and the sample fracture description situation relationship covered by the fracture fragment distribution situation, and obtain a second actual fracture description situation recognition result and a second actual fracture description situation relationship distribution description result corresponding to the second data analysis network, including: The fracture fragment distribution is loaded into the first artificial intelligence thread, and the actual fracture description in the fracture fragment distribution is identified by the first artificial intelligence thread to obtain the second actual fracture description identification result; The fracture fragment distribution is loaded into the second artificial intelligence thread, and the actual fracture description relationship in the fracture fragment distribution is extracted by the second artificial intelligence thread to obtain the second actual fracture description relationship distribution description result.

6. The method for spatial registration of fracture navigation surgical instruments based on augmented reality technology according to claim 1, characterized in that: The at least two data analysis networks include a third data analysis network; the at least two data analysis networks are used to analyze the relationship between the actual fracture description covered by the fracture fragment distribution and the sample fracture description, and the actual fracture description recognition result and the actual fracture description relationship distribution description result corresponding to each data analysis network are obtained, and the following further includes: According to the actual fracture description of each range and the relationship between the sample fracture description, a bone distribution database corresponding to each range is established; The fracture fragment distribution is projected with the actual fracture description and the sample fracture description relationship in the bone distribution database to obtain a third actual fracture description recognition result and a third actual fracture description relationship distribution description result corresponding to the third data analysis network.

7. The method for spatial registration of fracture navigation surgical instruments based on augmented reality technology according to claim 6, characterized in that: The projecting of the fracture fragment distribution with the actual fracture description and the sample fracture description relationship in the bone distribution database to obtain a third actual fracture description recognition result and a third actual fracture description relationship distribution description result corresponding to the third data parsing network includes: Searching the bone distribution database for an actual fracture description corresponding to the fracture fragment distribution, and determining the searched actual fracture description as the third actual fracture description recognition result; The actual fracture description situation relationship corresponding to the fracture fragment distribution is searched from the bone distribution database, and the searched actual fracture description situation relationship is determined as the third actual fracture description situation relationship distribution description result.

8. The method for spatial registration of fracture navigation surgical instruments based on augmented reality technology according to claim 7, characterized in that: If the actual fracture description corresponding to the fracture fragment distribution is searched from the bone distribution database, and the searched actual fracture description is determined as the third actual fracture description recognition result, it includes: If there are multiple bone distribution databases, and the actual fracture description in the fracture fragment distribution corresponds to the actual fracture description in the multiple bone distribution databases, the most relevant bone distribution database is screened from the multiple bone distribution databases, and the actual fracture description in the most relevant bone distribution database is determined as the third actual fracture description identification result.

9. The method for spatial registration of fracture navigation surgical instruments based on augmented reality technology according to claim 1, characterized in that: The actual fracture description situation recognition results and the actual fracture description situation relationship distribution description results corresponding to each data analysis network are spliced, and a spatial registration description result corresponding to the fracture fragment description information set is constructed based on the splicing results, including: Integrating the same actual fracture descriptions in the actual fracture description recognition results corresponding to each data analysis network, and fusing the integrated actual fracture descriptions with other actual fracture descriptions that have not been integrated to obtain a final actual fracture description recognition result; Integrating the same actual fracture description situation relationships in the actual fracture description situation relationship distribution description results corresponding to each data analysis network, and fusing the integrated actual fracture description situation relationships with other actual fracture description situation relationships that have not been integrated, to obtain a final actual fracture description situation relationship distribution description result; According to the final actual fracture description recognition result and the final actual fracture description relationship distribution description result, construct the spatial registration description result corresponding to the fracture fragment description information set; Wherein, the spatial registration description result corresponding to the fracture fragment description information set is constructed according to the final actual fracture description recognition result and the final actual fracture description relationship distribution description result, including: Determining the actual fracture description in the final actual fracture description recognition result as an attribute of the network structure; Determine the relationship between different attributes according to the actual fracture description relationship in the final actual fracture description relationship distribution description result, and determine the constraint conditions of the network structure; The attributes of the network structure and the constraints of the network structure are integrated to obtain the spatial registration description result.

10. The method for spatial registration of fracture navigation surgical instruments based on augmented reality technology according to claim 1, characterized in that: The method further comprises: Extracting new actual fracture description situation recognition results and new relationship distribution description results from other three-dimensional fracture data clusters other than the at least one three-dimensional fracture data cluster; The new actual fracture description recognition result and the new relationship distribution description result are fused with the spatial registration description result to obtain an updated spatial registration description result.

Images