Magnetic resonance eyeball imaging method and system based on resolution control

Abstract

The invention discloses a magnetic resonance eyeball imaging method and system based on resolution control, and the method comprises the steps: determining the pathological information of a target user in response to an imaging demand; determining a scanning resolution parameter of the target user based on a scanning prediction algorithm according to the pathological information; performing magnetic resonance scanning on the eyeball part of the target user according to the scanning resolution parameter to obtain a plurality of scanning images with different resolutions; and generating a three-dimensional eyeball model corresponding to the target user based on a three-dimensional reconstruction algorithm according to the plurality of scanning images with different resolutions. Therefore, accurate eyeball three-dimensional modeling based on pathological information and multi-resolution images can be realized, the eyeball imaging accuracy and diagnosis reliability are improved, and the model distortion risk caused by improper resolution is reduced.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a magnetic resonance ocular imaging method and system based on resolution control. Background Technology

[0002] With the rapid growth in demand for ophthalmic diagnosis and imaging technologies, medical institutions are increasingly emphasizing the improvement of the reliability and accuracy of ocular diagnosis through precise 3D modeling. Current technologies typically collect pathological information from users in response to imaging needs, generate ocular images using fixed-resolution scanning or standard image processing methods, and construct models based on simple 3D reconstruction algorithms to support diagnostic analysis. However, existing solutions lack dynamic analysis of pathological information and optimized prediction of scanning resolution, making it difficult to generate multi-resolution images adapted to pathological requirements. Commonly used uniform scanning and modeling strategies cannot adapt to complex ocular structures, resulting in insufficient model accuracy and susceptibility to distortion due to inappropriate resolution, thus limiting the diagnostic reliability and clinical application effectiveness of ocular imaging. Therefore, existing technologies have shortcomings that urgently need to be addressed. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a magnetic resonance ocular imaging method and system based on resolution control, which can realize accurate three-dimensional modeling of the eyeball based on pathological information and multi-resolution images, improve the accuracy of ocular imaging and diagnostic reliability, and reduce the risk of model distortion caused by inappropriate resolution.

[0004] To address the aforementioned technical problems, the first aspect of this invention discloses a magnetic resonance ocular imaging method based on resolution control, the method comprising: In response to imaging needs, determine the pathological information of the target user; Based on the pathological information, the scan resolution parameters of the target user are determined using a scan prediction algorithm. The target user's eyeball is subjected to magnetic resonance scanning according to the scanning resolution parameters to obtain multiple scan images of different resolutions; Based on the multiple scanned images of different resolutions, a three-dimensional eye model corresponding to the target user is generated using a three-dimensional reconstruction algorithm.

[0005] As an optional implementation, in the first aspect of the present invention, determining the pathological information of the target user includes: Obtain the target user's user parameters and historical medical records; The user parameters are input into the first prediction model to obtain the first pathological information; The historical medical records are input into the second prediction model to obtain the second pathological information; both the first prediction model and the second prediction model are jointly optimized and trained using a training dataset that includes multiple training user parameters, corresponding training medical records, and corresponding pathological information annotations. The intersection of the first pathological information and the second pathological information is calculated to obtain the pathological information of the target user.

[0006] As an optional implementation, in the first aspect of the present invention, the user parameters include user name, user height, user age, user gender, and user physiological indicators.

[0007] As an optional implementation, in the first aspect of the present invention, the pathological information includes disease information, disease prognosis, disease location, and disease medication plan.

[0008] As an optional implementation, in the first aspect of the invention, determining the scan resolution parameters of the target user based on the pathological information and a scan prediction algorithm includes: Based on the pre-defined correspondence between pathology and ocular tissue characteristics, the ocular tissue parameters corresponding to the target user are determined according to the pathological information; the ocular tissue parameters include ocular volume, relative position of the ocular, ocular diameter, and ocular tissue morphology. The pathological information and the ocular tissue parameters are input into a trained scan prediction neural network to obtain the scan resolution parameters of the target user; the scan prediction neural network is trained using a training dataset that includes multiple training pathological information and corresponding ocular tissue parameter annotations and optimal scan parameter annotations.

[0009] As an optional implementation, in the first aspect of the present invention, the scanning resolution parameters include the correspondence between scanning resolution and scanning position, scanning interval distance, and scanning interval time.

[0010] As an optional implementation, in the first aspect of the present invention, generating a three-dimensional eye model corresponding to the target user based on a three-dimensional reconstruction algorithm from the plurality of scanned images of different resolutions includes: All the scanned images are grouped based on their resolution to obtain multiple image groups; each image group includes multiple scanned images whose resolution difference with each other is less than a preset difference threshold. All the image groups are sorted according to their average resolution to obtain an image group sequence; The image groups of the image group sequence are sequentially input into the three-dimensional reconstruction network until a three-dimensional eye model that meets the requirements is obtained.

[0011] As an optional implementation, in the first aspect of the invention, the step of sequentially inputting the image group sequence into a three-dimensional reconstruction network until a three-dimensional eyeball model meeting the requirements includes: The image groups in the image group sequence are input into the trained 3D reconstruction network from front to back, and the reconstructed eyeball models are output sequentially; the 3D reconstruction network is trained using a training dataset that includes multiple training image groups and corresponding 3D model annotations. The output reconstructed eye model is input into the accuracy recognition model to obtain the model accuracy; the accuracy recognition model is trained using a training dataset that includes multiple trained 3D models and corresponding accuracy annotations; Determine whether the model accuracy meets the preset accuracy requirements. If so, stop inputting the next sequence of the image group into the 3D reconstruction network and determine the current reconstructed eyeball model as the 3D eyeball model corresponding to the target user.

[0012] A second aspect of this invention discloses a magnetic resonance ocular imaging system based on resolution control, the system comprising: The determination module is used to determine the pathological information of the target user in response to imaging requirements; The prediction module is used to determine the scan resolution parameters of the target user based on the pathological information and a scan prediction algorithm. The scanning module is used to perform magnetic resonance scanning on the eye area of ​​the target user according to the scanning resolution parameters, so as to obtain multiple scan images with different resolutions; The reconstruction module is used to generate a three-dimensional eyeball model corresponding to the target user based on the multiple scanned images of different resolutions and a three-dimensional reconstruction algorithm.

[0013] As an optional implementation, in a second aspect of the invention, the specific method by which the determining module determines the pathological information of the target user includes: Obtain the target user's user parameters and historical medical records; The user parameters are input into the first prediction model to obtain the first pathological information; The historical medical records are input into the second prediction model to obtain the second pathological information; both the first prediction model and the second prediction model are jointly optimized and trained using a training dataset that includes multiple training user parameters, corresponding training medical records, and corresponding pathological information annotations. The intersection of the first pathological information and the second pathological information is calculated to obtain the pathological information of the target user.

[0014] As an optional implementation, in a second aspect of the present invention, the user parameters include user name, user height, user age, user gender, and user physiological indicators.

[0015] As an optional implementation, in a second aspect of the invention, the pathological information includes disease information, disease prognosis, disease location, and disease medication plan.

[0016] As an optional implementation, in a second aspect of the invention, the specific method by which the prediction module determines the scan resolution parameters of the target user based on the pathological information and a scan prediction algorithm includes: Based on the pre-defined correspondence between pathology and ocular tissue characteristics, the ocular tissue parameters corresponding to the target user are determined according to the pathological information; the ocular tissue parameters include ocular volume, relative position of the ocular, ocular diameter, and ocular tissue morphology. The pathological information and the ocular tissue parameters are input into a trained scan prediction neural network to obtain the scan resolution parameters of the target user; the scan prediction neural network is trained using a training dataset that includes multiple training pathological information and corresponding ocular tissue parameter annotations and optimal scan parameter annotations.

[0017] As an optional implementation, in a second aspect of the present invention, the scanning resolution parameters include the correspondence between scanning resolution and scanning position, scanning interval distance, and scanning interval time.

[0018] As an optional implementation, in a second aspect of the invention, the specific method by which the reconstruction module generates a three-dimensional eye model corresponding to the target user based on the plurality of scanned images of different resolutions and a three-dimensional reconstruction algorithm includes: All the scanned images are grouped based on their resolution to obtain multiple image groups; each image group includes multiple scanned images whose resolution difference with each other is less than a preset difference threshold. All the image groups are sorted according to their average resolution to obtain an image group sequence; The image groups of the image group sequence are sequentially input into the three-dimensional reconstruction network until a three-dimensional eye model that meets the requirements is obtained.

[0019] As an optional implementation, in a second aspect of the invention, the specific method by which the reconstruction module sequentially inputs the image group sequence into a three-dimensional reconstruction network until a three-dimensional eyeball model meeting the requirements is as follows: The image groups in the image group sequence are input into the trained 3D reconstruction network from front to back, and the reconstructed eyeball models are output sequentially; the 3D reconstruction network is trained using a training dataset that includes multiple training image groups and corresponding 3D model annotations. The output reconstructed eye model is input into the accuracy recognition model to obtain the model accuracy; the accuracy recognition model is trained using a training dataset that includes multiple trained 3D models and corresponding accuracy annotations; Determine whether the model accuracy meets the preset accuracy requirements. If so, stop inputting the next sequence of the image group into the 3D reconstruction network and determine the current reconstructed eyeball model as the 3D eyeball model corresponding to the target user.

[0020] A third aspect of the present invention discloses another magnetic resonance ocular imaging system based on resolution control, the system comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the resolution-controlled magnetic resonance ocular imaging method disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the resolution-controlled magnetic resonance ocular imaging method disclosed in the first aspect of the present invention.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention determines the pathological information of the target user by responding to imaging needs, determines the scanning resolution parameters based on the scanning prediction algorithm and obtains multi-resolution scanning images, and generates a three-dimensional eye model by combining a three-dimensional reconstruction algorithm. This enables accurate three-dimensional modeling of the eye based on pathological information and multi-resolution images, improves the accuracy of eye imaging and diagnostic reliability, and reduces the risk of model distortion caused by inappropriate resolution. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1This is a schematic flowchart of a magnetic resonance ocular imaging method based on resolution control disclosed in an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of the structure of a magnetic resonance ocular imaging system based on resolution control disclosed in an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of another magnetic resonance ocular imaging system based on resolution control disclosed in an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] This invention discloses a magnetic resonance ocular imaging method and system based on resolution control. By responding to imaging requirements to determine the target user's pathological information, a scan resolution parameter is determined based on a scan prediction algorithm, and multi-resolution scan images are obtained. Combined with a 3D reconstruction algorithm, a 3D ocular model is generated. This enables accurate 3D ocular modeling based on pathological information and multi-resolution images, improving the accuracy and diagnostic reliability of ocular imaging and reducing the risk of model distortion due to inappropriate resolution. Detailed explanations follow.

[0031] Example 1 Please see Figure 1 , Figure 1 This is a schematic flowchart of a magnetic resonance ocular imaging method based on resolution control, as disclosed in an embodiment of the present invention. Wherein, Figure 1 The described resolution-controlled magnetic resonance ocular imaging method can be applied to data processing systems / data processing devices / data processing servers (wherein, the server includes local processing servers or cloud processing servers). For example... Figure 1 As shown, this resolution-controlled magnetic resonance ocular imaging method may include the following operations: 101. In response to imaging needs, determine the pathological information of the target user.

[0032] Optionally, the imaging requirement can be for ophthalmic diagnosis, surgical planning, or medical research; this invention does not limit this requirement.

[0033] Optionally, the pathological information may include the type of eye disease, the degree of lesion, or abnormal tissue parameters, which are not limited in this invention.

[0034] Optionally, this determination process can be based on user input, medical record queries, or automatic analysis; this invention does not impose any limitations.

[0035] 102. Based on pathological information and a scan prediction algorithm, determine the scan resolution parameters for the target user. 103. Perform magnetic resonance scanning on the target user's eyeball according to the scanning resolution parameters to obtain multiple scan images with different resolutions. 104. Based on multiple scanned images of different resolutions, generate a 3D eyeball model corresponding to the target user using a 3D reconstruction algorithm.

[0036] Optionally, the 3D reconstruction algorithm can be a stereoscopic reconstruction algorithm, a surface reconstruction algorithm, or a deep learning reconstruction algorithm; this invention does not limit the specific algorithm.

[0037] Optionally, the three-dimensional eye model may include the cornea, lens, retina, or the overall structure of the eyeball; this invention does not limit this.

[0038] As can be seen, the above-described embodiments of the invention determine the pathological information of the target user in response to imaging requirements, determine the scanning resolution parameters based on the scanning prediction algorithm and obtain multi-resolution scanning images, and generate a three-dimensional eye model by combining a three-dimensional reconstruction algorithm. This enables accurate three-dimensional modeling of the eye based on pathological information and multi-resolution images, improves the accuracy of eye imaging and diagnostic reliability, and reduces the risk of model distortion caused by inappropriate resolution.

[0039] As an optional embodiment, the step of determining the target user's pathological information in the above steps includes: Obtain the target user's user parameters and historical medical records; Input the user parameters into the first prediction model to obtain the first pathological information; Historical medical records are input into the second prediction model to obtain the second pathological information; optionally, both the first and second prediction models are jointly optimized and trained using a training dataset that includes multiple training user parameters and corresponding training medical records and corresponding pathological information annotations. The intersection of the first pathological information and the second pathological information is calculated to obtain the target user's pathological information.

[0040] Optionally, the user parameters may include age, gender, vision data, or eye biometrics, which are not limited in this invention.

[0041] Optionally, the historical medical records may include diagnostic reports, treatment records, or imaging data, and the present invention does not limit them.

[0042] Optionally, this acquisition process can be implemented based on an electronic medical record system, user input, or database query; this invention does not limit this process.

[0043] Optionally, the first prediction model can be a convolutional neural network, a random forest model, or a support vector machine model; this invention does not limit the model.

[0044] Optionally, the first pathological information can be a disease classification, lesion probability, or risk score, and the present invention does not limit it.

[0045] Optionally, the second prediction model can be a recurrent neural network, a Transformer model, or a regression model; this invention does not limit the specific model.

[0046] Optionally, the intersection calculation can be implemented based on set operations, probability fusion, or logical reasoning, and this invention does not limit it.

[0047] Optionally, the pathological information may be a comprehensive diagnostic result, lesion area, or risk level; this invention does not limit this.

[0048] As can be seen, through the above optional embodiments, by inputting the target user's user parameters and historical medical records into the first and second prediction models respectively to obtain pathological information and taking the intersection to determine the final pathological information, accurate pathological information extraction based on multi-source data collaborative prediction is achieved, thereby improving the accuracy and reliability of target user's ocular pathology analysis and reducing the risk of pathological misjudgment.

[0049] As an optional embodiment, the user parameters in the above steps include user name, user height, user age, user gender, and user physiological indicators.

[0050] As can be seen, the above optional embodiments limit the content of user parameters to comprehensively characterize the user's parameter features, assist in achieving accurate three-dimensional modeling of the eyeball based on pathological information and multi-resolution images, improve the accuracy of eyeball imaging and diagnostic reliability, and reduce the risk of model distortion caused by inappropriate resolution.

[0051] As an optional embodiment, the pathological information in the above steps includes disease information, disease prognosis, disease location, and disease medication plan.

[0052] As can be seen, the content of pathological information is defined through the above optional embodiments to comprehensively characterize the user's pathological features, assist in the realization of accurate three-dimensional modeling of the eyeball based on pathological information and multi-resolution images, improve the accuracy of eyeball imaging and diagnostic reliability, and reduce the risk of model distortion caused by inappropriate resolution.

[0053] As an optional embodiment, the step described above, determining the scan resolution parameters of the target user based on pathological information and a scan prediction algorithm, includes: Based on the pre-defined correspondence between pathology and ocular tissue characteristics, the ocular tissue parameters corresponding to the target user are determined according to the pathological information; optionally, the ocular tissue parameters include ocular volume, relative position of the ocular, ocular diameter, and ocular tissue morphology. The pathological information and ocular tissue parameters are input into the trained scan prediction neural network to obtain the scan resolution parameters of the target user. The scan prediction neural network is trained using a training dataset that includes multiple training pathological information and corresponding ocular tissue parameter annotations and optimal scan parameter annotations.

[0054] Optionally, the correspondence can be a lookup table, a mathematical mapping, or a data-driven model; this invention does not limit the type of correspondence.

[0055] Optionally, the ocular tissue parameters may include additional parameters, such as ocular curvature, corneal thickness, or retinal density, which are not limited in this invention.

[0056] Optionally, the determination of these ocular tissue parameters can be based on feature matching, rule reasoning, or machine learning models, and this invention does not limit the scope of the determination.

[0057] Optionally, the scanning prediction neural network can be a deep neural network, a convolutional neural network, or a regression neural network; the present invention does not limit this.

[0058] Optionally, the scanning resolution parameter can be a resolution level, pixel density, or scanning accuracy value; this invention does not limit this.

[0059] Optionally, the training dataset may include historical scan data, simulation parameter data, or experimental data, and this invention does not impose any limitations.

[0060] As can be seen, through the above optional embodiments, by determining the ocular tissue parameters based on pathological information and preset correspondence, and combining them with a trained scan prediction neural network to generate scan resolution parameters, accurate scan parameter optimization based on pathological and tissue features can be achieved, thereby improving the quality of ocular scan images and the accuracy of 3D modeling, and reducing the risk of imaging errors caused by improper parameters.

[0061] As an optional embodiment, the scanning resolution parameters in the above steps include the correspondence between scanning resolution and scanning position, scanning interval distance, and scanning interval time.

[0062] As can be seen, the above optional embodiments define the content of the scanning resolution parameters to comprehensively characterize the user's scanning resolution parameter features, assist in achieving accurate three-dimensional modeling of the eyeball based on pathological information and multi-resolution images, improve the accuracy of eyeball imaging and diagnostic reliability, and reduce the risk of model distortion caused by inappropriate resolution.

[0063] As an optional embodiment, the step described above, generating a 3D eye model corresponding to the target user based on multiple scanned images of different resolutions and a 3D reconstruction algorithm, includes: All scanned images are grouped based on resolution to obtain multiple image groups; optionally, each image group includes multiple scanned images whose resolution difference is less than a preset difference threshold. All image groups are sorted according to their average resolution to obtain an image group sequence; The image groups in the image group sequence are input into the 3D reconstruction network in sequence until a 3D eye model that meets the requirements is obtained.

[0064] Optionally, the grouping process can be implemented based on K-means clustering, hierarchical clustering, or threshold segmentation, and this invention does not limit it.

[0065] Optionally, the gap threshold can be a fixed threshold, a dynamic threshold, or a threshold adjusted based on the performance of the scanning device; the present invention does not limit this.

[0066] As can be seen, through the above optional embodiments, by grouping scanned images based on resolution size and sorting them according to average resolution to form an image group sequence, and then inputting them sequentially into a 3D reconstruction network to generate an eye model, accurate 3D eye modeling based on resolution grouping and sequence processing can be achieved, thereby improving model generation efficiency and quality and reducing the risk of reconstruction errors caused by improper image selection.

[0067] As an optional embodiment, the step described above, which involves sequentially inputting the image group sequence into the 3D reconstruction network until a 3D eyeball model that meets the requirements, includes: The image groups in the image group sequence are input into the trained 3D reconstruction network from front to back, and the reconstructed eyeball models are output sequentially; optionally, the 3D reconstruction network is trained using a training dataset that includes multiple training image groups and corresponding 3D model annotations. The output reconstructed eye model is input into the accuracy recognition model to obtain the model accuracy; optionally, the accuracy recognition model is trained using a training dataset that includes multiple trained 3D models and corresponding accuracy annotations. Determine whether the model accuracy meets the preset accuracy requirements. If so, stop inputting the next sequence of image groups into the 3D reconstruction network and determine the current reconstructed eyeball model as the 3D eyeball model corresponding to the target user.

[0068] Optionally, the 3D reconstruction network can be a U-Net model, a VoxelNet model, or a generative adversarial network model; this invention does not limit the specific model.

[0069] Optionally, the reconstructed eyeball model may include an eyeball surface model, a stereoscopic model, or a multi-layered tissue model; the present invention does not impose any limitations on this.

[0070] Optionally, the training dataset may include historical scanned images, simulated image data, or labeled data; this invention does not impose any limitations.

[0071] Optionally, the accuracy recognition model can be a classification model, a regression model, or a deep learning model; this invention does not limit the specific model.

[0072] Optionally, the model accuracy can be an accuracy score, an error value, or a confidence score; this invention does not impose any limitations on this.

[0073] Optionally, the accuracy requirement can be a fixed accuracy threshold, a dynamic accuracy threshold, or a threshold adjusted based on clinical needs; this invention does not limit the specific accuracy requirement.

[0074] As can be seen, through the above optional embodiments, by sequentially inputting image groups into a 3D reconstruction network to generate a reconstructed eyeball model and using a precision recognition model to evaluate the model's precision, models that meet the precision requirements are selected as the final 3D eyeball model, thereby achieving accurate model selection based on precision evaluation, improving the reliability and diagnostic applicability of 3D eyeball modeling, and reducing the risk of insufficient model precision.

[0075] Example 2 Please see Figure 2 , Figure 2 This is a schematic diagram of a magnetic resonance ocular imaging system based on resolution control, as disclosed in an embodiment of the present invention. Figure 2The described resolution-controlled magnetic resonance ocular imaging system can be applied to data processing systems / data processing devices / data processing servers (wherein, the server includes a local processing server or a cloud processing server). For example... Figure 2 As shown, the resolution-controlled magnetic resonance ocular imaging system may include: The determination module 201 is used to determine the pathological information of the target user in response to imaging requirements.

[0076] The prediction module 202 is used to determine the scan resolution parameters of the target user based on the pathological information and the scan prediction algorithm. The scanning module 203 is used to perform magnetic resonance scanning on the eye of the target user according to the scanning resolution parameters, so as to obtain multiple scan images of different resolutions. The reconstruction module 204 is used to generate a three-dimensional eyeball model corresponding to the target user based on a three-dimensional reconstruction algorithm, using multiple scanned images of different resolutions.

[0077] As can be seen, the above-described embodiments of the invention determine the pathological information of the target user in response to imaging requirements, determine the scanning resolution parameters based on the scanning prediction algorithm and obtain multi-resolution scanning images, and generate a three-dimensional eye model by combining a three-dimensional reconstruction algorithm. This enables accurate three-dimensional modeling of the eye based on pathological information and multi-resolution images, improves the accuracy of eye imaging and diagnostic reliability, and reduces the risk of model distortion caused by inappropriate resolution.

[0078] As an optional embodiment, the method by which the determining module determines the pathological information of the target user includes: Obtain the target user's user parameters and historical medical records; Input the user parameters into the first prediction model to obtain the first pathological information; Historical medical records are input into the second prediction model to obtain the second pathological information; optionally, both the first and second prediction models are jointly optimized and trained using a training dataset that includes multiple training user parameters and corresponding training medical records and corresponding pathological information annotations. The intersection of the first pathological information and the second pathological information is calculated to obtain the target user's pathological information.

[0079] As can be seen, through the above optional embodiments, by inputting the target user's user parameters and historical medical records into the first and second prediction models respectively to obtain pathological information and taking the intersection to determine the final pathological information, accurate pathological information extraction based on multi-source data collaborative prediction is achieved, thereby improving the accuracy and reliability of target user's ocular pathology analysis and reducing the risk of pathological misjudgment.

[0080] As an optional implementation, user parameters include user name, user height, user age, user gender, and user physiological indicators.

[0081] As can be seen, the above optional embodiments limit the content of user parameters to comprehensively characterize the user's parameter features, assist in achieving accurate three-dimensional modeling of the eyeball based on pathological information and multi-resolution images, improve the accuracy of eyeball imaging and diagnostic reliability, and reduce the risk of model distortion caused by inappropriate resolution.

[0082] As an optional embodiment, pathological information includes disease information, disease prognosis, disease location, and disease medication plan.

[0083] As can be seen, the content of pathological information is defined through the above optional embodiments to comprehensively characterize the user's pathological features, assist in the realization of accurate three-dimensional modeling of the eyeball based on pathological information and multi-resolution images, improve the accuracy of eyeball imaging and diagnostic reliability, and reduce the risk of model distortion caused by inappropriate resolution.

[0084] As an optional embodiment, the prediction module determines the specific method by which it determines the scan resolution parameters of the target user based on pathological information and a scan prediction algorithm, including: Based on the pre-defined correspondence between pathology and ocular tissue characteristics, the ocular tissue parameters corresponding to the target user are determined according to the pathological information; optionally, the ocular tissue parameters include ocular volume, relative position of the ocular, ocular diameter, and ocular tissue morphology. The pathological information and ocular tissue parameters are input into the trained scan prediction neural network to obtain the scan resolution parameters of the target user. The scan prediction neural network is trained using a training dataset that includes multiple training pathological information and corresponding ocular tissue parameter annotations and optimal scan parameter annotations.

[0085] As can be seen, through the above optional embodiments, by determining the ocular tissue parameters based on pathological information and preset correspondence, and combining them with a trained scan prediction neural network to generate scan resolution parameters, accurate scan parameter optimization based on pathological and tissue features can be achieved, thereby improving the quality of ocular scan images and the accuracy of 3D modeling, and reducing the risk of imaging errors caused by improper parameters.

[0086] As an optional embodiment, the scanning resolution parameters include the correspondence between scanning resolution and scanning position, scanning interval distance, and scanning interval time.

[0087] As can be seen, the above optional embodiments define the content of the scanning resolution parameters to comprehensively characterize the user's scanning resolution parameter features, assist in achieving accurate three-dimensional modeling of the eyeball based on pathological information and multi-resolution images, improve the accuracy of eyeball imaging and diagnostic reliability, and reduce the risk of model distortion caused by inappropriate resolution.

[0088] As an optional embodiment, the specific method by which the reconstruction module generates a 3D eyeball model corresponding to the target user based on multiple scanned images of different resolutions and a 3D reconstruction algorithm includes: All scanned images are grouped based on resolution to obtain multiple image groups; optionally, each image group includes multiple scanned images whose resolution difference is less than a preset difference threshold. All image groups are sorted according to their average resolution to obtain an image group sequence; The image groups in the image group sequence are input into the 3D reconstruction network in sequence until a 3D eye model that meets the requirements is obtained.

[0089] As can be seen, through the above optional embodiments, by grouping scanned images based on resolution size and sorting them according to average resolution to form an image group sequence, and then inputting them sequentially into a 3D reconstruction network to generate an eye model, accurate 3D eye modeling based on resolution grouping and sequence processing can be achieved, thereby improving model generation efficiency and quality and reducing the risk of reconstruction errors caused by improper image selection.

[0090] As an optional embodiment, the specific method by which the reconstruction module sequentially inputs the image group sequence into the 3D reconstruction network until a 3D eyeball model that meets the requirements is as follows: The image groups in the image group sequence are input into the trained 3D reconstruction network from front to back, and the reconstructed eyeball models are output sequentially; optionally, the 3D reconstruction network is trained using a training dataset that includes multiple training image groups and corresponding 3D model annotations. The output reconstructed eye model is input into the accuracy recognition model to obtain the model accuracy; optionally, the accuracy recognition model is trained using a training dataset that includes multiple trained 3D models and corresponding accuracy annotations. Determine whether the model accuracy meets the preset accuracy requirements. If so, stop inputting the next sequence of image groups into the 3D reconstruction network and determine the current reconstructed eyeball model as the 3D eyeball model corresponding to the target user.

[0091] As can be seen, through the above optional embodiments, by sequentially inputting image groups into a 3D reconstruction network to generate a reconstructed eyeball model and using a precision recognition model to evaluate the model's precision, models that meet the precision requirements are selected as the final 3D eyeball model, thereby achieving accurate model selection based on precision evaluation, improving the reliability and diagnostic applicability of 3D eyeball modeling, and reducing the risk of insufficient model precision.

[0092] Example 3 Please see Figure 3 , Figure 3 This is another magnetic resonance ocular imaging system based on resolution control disclosed in the embodiments of the present invention. Figure 3 The described resolution-controlled magnetic resonance ocular imaging system is applied in a data processing system / data processing device / data processing server (wherein, the server includes a local processing server or a cloud processing server). For example... Figure 3 As shown, the resolution-controlled magnetic resonance ocular imaging system may include: Memory 301 storing executable program code; Processor 302 coupled to memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the resolution-controlled magnetic resonance ocular imaging method described in Embodiment 1.

[0093] Example 4 This invention discloses a computer read storage medium that stores a computer program for electronic data exchange, wherein the computer program causes a computer to execute the steps of the resolution-controlled magnetic resonance ocular imaging method described in Embodiment 1.

[0094] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of the resolution-controlled magnetic resonance ocular imaging method described in Embodiment 1.

[0095] The foregoing has described specific embodiments of this specification; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0096] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0097] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0098] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0099] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0100] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0101] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0102] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0103] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0104] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0105] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0106] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0107] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0108] Finally, it should be noted that the magnetic resonance ocular imaging method and system based on resolution control disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, and not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A magnetic resonance ocular imaging method based on resolution control, characterized in that, The method includes: In response to imaging needs, determine the pathological information of the target user; Based on the pathological information, the scan resolution parameters of the target user are determined using a scan prediction algorithm. The target user's eyeball is subjected to magnetic resonance scanning according to the scanning resolution parameters to obtain multiple scan images of different resolutions; Based on the multiple scanned images of different resolutions, a three-dimensional eye model corresponding to the target user is generated using a three-dimensional reconstruction algorithm.

2. The magnetic resonance ocular imaging method based on resolution control according to claim 1, characterized in that, The determination of the target user's pathological information includes: Obtain the target user's user parameters and historical medical records; The user parameters are input into the first prediction model to obtain the first pathological information; The historical medical records are input into the second prediction model to obtain the second pathological information; both the first prediction model and the second prediction model are jointly optimized and trained using a training dataset that includes multiple training user parameters, corresponding training medical records, and corresponding pathological information annotations. The intersection of the first pathological information and the second pathological information is calculated to obtain the pathological information of the target user.

3. The magnetic resonance ocular imaging method based on resolution control according to claim 2, characterized in that, The user parameters include user name, user height, user age, user gender, and user physiological indicators.

4. The magnetic resonance ocular imaging method based on resolution control according to claim 1, characterized in that, The pathological information includes disease information, disease prognosis, disease location, and disease treatment plan.

5. The magnetic resonance ocular imaging method based on resolution control according to claim 1, characterized in that, The step of determining the scan resolution parameters of the target user based on the pathological information and a scan prediction algorithm includes: Based on the pre-defined correspondence between pathology and ocular tissue characteristics, the ocular tissue parameters corresponding to the target user are determined according to the pathological information; the ocular tissue parameters include ocular volume, relative position of the ocular, ocular diameter, and ocular tissue morphology. The pathological information and the ocular tissue parameters are input into a trained scan prediction neural network to obtain the scan resolution parameters of the target user; the scan prediction neural network is trained using a training dataset that includes multiple training pathological information and corresponding ocular tissue parameter annotations and optimal scan parameter annotations.

6. The magnetic resonance ocular imaging method based on resolution control according to claim 1, characterized in that, The scanning resolution parameters include the correspondence between scanning resolution and scanning position, scanning interval distance, and scanning interval time.

7. The magnetic resonance ocular imaging method based on resolution control according to claim 6, characterized in that, The step of generating a 3D eye model corresponding to the target user based on the multiple scanned images of different resolutions and a 3D reconstruction algorithm includes: All the scanned images are grouped based on their resolution to obtain multiple image groups; each image group includes multiple scanned images whose resolution difference with each other is less than a preset difference threshold. All the image groups are sorted according to their average resolution to obtain an image group sequence; The image groups of the image group sequence are sequentially input into the three-dimensional reconstruction network until a three-dimensional eye model that meets the requirements is obtained.

8. The magnetic resonance ocular imaging method based on resolution control according to claim 7, characterized in that, The step of sequentially inputting the image groups of the image group sequence into the 3D reconstruction network until a 3D eyeball model that meets the requirements is obtained includes: The image groups in the image group sequence are input into the trained 3D reconstruction network from front to back, and the reconstructed eyeball models are output sequentially; the 3D reconstruction network is trained using a training dataset that includes multiple training image groups and corresponding 3D model annotations. The output reconstructed eye model is input into the accuracy recognition model to obtain the model accuracy; the accuracy recognition model is trained using a training dataset that includes multiple trained 3D models and corresponding accuracy annotations; Determine whether the model accuracy meets the preset accuracy requirements. If so, stop inputting the next sequence of the image group into the 3D reconstruction network and determine the current reconstructed eyeball model as the 3D eyeball model corresponding to the target user.

9. A magnetic resonance ocular imaging system based on resolution control, characterized in that, The system includes: The determination module is used to determine the pathological information of the target user in response to imaging requirements; The prediction module is used to determine the scan resolution parameters of the target user based on the pathological information and a scan prediction algorithm. The scanning module is used to perform magnetic resonance scanning on the eye area of ​​the target user according to the scanning resolution parameters, so as to obtain multiple scan images with different resolutions; The reconstruction module is used to generate a three-dimensional eyeball model corresponding to the target user based on the multiple scanned images of different resolutions and a three-dimensional reconstruction algorithm.

10. A magnetic resonance ocular imaging system based on resolution control, characterized in that, The system includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the magnetic resonance ophthalmic imaging method based on resolution control as described in any one of claims 1-8.

Images