Feature guidance generation method and system for inner ear lymphatic effusion lesion

Abstract

The invention provides a feature guidance generation method and system for inner ear lymphatic effusion lesion, and the method comprises the steps: obtaining image data of a patient in a lesion region, and marking a lesion range image of the lesion region based on the image data; on the basis of the lesion range image, lesion parameter information of the lesion area is recognized through an image recognition model, and lesion feature information of the lesion area is recognized on the basis of the lesion parameter information of the lesion area; and generating a lesion feature guide image of the patient in the lesion range image based on the lesion feature information, and generating a feature guide report of the patient based on the lesion feature guide image of the patient and the lesion feature information. By adopting the scheme, the feature recognition efficiency of the inner ear lymphatic effusion can be improved.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence and image feature recognition technology, and in particular to a method and system for generating feature guidance for lesions of endolymphatic effusion in the inner ear. Background Technology

[0002] Inner ear effusion is most commonly seen in Meniere's disease, a common inner ear disorder that typically presents with symptoms such as hearing loss, tinnitus, and vertigo. Diagnosing inner ear effusion requires careful examination of medical imaging images to analyze its condition and severity. However, traditional images often reveal complex features of inner ear effusion, making it difficult to accurately determine its specific state. Therefore, improving the accuracy of feature identification for inner ear effusion is a current research focus.

[0003] Traditional diagnosis of inner ear lymph effusion relies primarily on medical imaging (such as MRI or CT) and the physician's experience. However, diagnostic results depend on the physician's experience, which may lead to misdiagnosis or missed diagnosis. Furthermore, due to the complex characteristics of inner ear lymph effusion, physicians need to spend a significant amount of time analyzing imaging data, resulting in a slow diagnostic process and low efficiency in identifying the characteristics of inner ear lymph effusion. Summary of the Invention

[0004] The main objective of this invention is to provide a method and system for generating feature guidance for endolymphatic effusion lesions in the inner ear. This aims to address the problems in the prior art, where diagnostic results rely on the doctor's experience, which may lead to misdiagnosis or missed diagnosis. Furthermore, the characteristics of endolymphatic effusion are complex, requiring doctors to spend a significant amount of time analyzing image data, resulting in slow diagnostic speed and low efficiency in identifying the features of endolymphatic effusion.

[0005] To achieve the above objectives, the present invention provides a method for generating characteristic indicators for endolymphatic effusion lesions in the inner ear, the method comprising:

[0006] Acquire image data of the patient in the lesion area, and based on the image data, mark the lesion extent in the lesion area;

[0007] Based on the lesion range image, the lesion parameter information of the lesion area is identified through an image recognition model, and based on the lesion parameter information of the lesion area, the lesion feature information of each lesion area is identified.

[0008] Based on the lesion feature information, a lesion feature guidance image of the patient is generated in the lesion range image, and a feature guidance report of the patient is generated based on the lesion feature guidance image of the patient and the lesion feature information.

[0009] Optionally, the step of marking the lesion extent image based on the image data includes:

[0010] The image edge recognition network identifies the edge information of the lesion range in the image data;

[0011] In the image data, the edge information of the lesion range is processed by image marking using a preset marking method to obtain the lesion range image of the lesion area.

[0012] Optionally, the step of identifying lesion parameter information of the lesion area based on the lesion range image using an image recognition model includes:

[0013] Identify sub-region images within the lesion area image, and use an image recognition model to identify the image parameter information corresponding to the sub-region images;

[0014] Based on the image parameter information, the lesion parameter values ​​of each lesion parameter type corresponding to the sub-range image are identified through a lesion parameter identification strategy, and the lesion parameter values ​​of all lesion parameter types are used as the lesion parameter information of the lesion region.

[0015] Optionally, identifying various lesion feature information of the lesion region based on the lesion parameter information of the lesion region includes:

[0016] Based on the image parameter information corresponding to the sub-range image, a two-dimensional model of the lesion is constructed using a two-dimensional modeling strategy;

[0017] Based on the image parameter information, the two-dimensional model of the lesion is subjected to model assignment processing to obtain the lesion structure model of the lesion region;

[0018] Based on the lesion structure model, a two-dimensional feature extraction network is used to extract the regional structure features of the lesion region. Based on the regional structure features, the lesion parameter values ​​of each lesion parameter type are adjusted using the lesion parameter identification strategy to obtain the target lesion parameter values ​​of each lesion parameter type.

[0019] The target lesion parameter values ​​of each of the lesion parameter types are used as the lesion feature information of each lesion region.

[0020] Optionally, generating a lesion feature guidance image for the patient based on each of the lesion feature information within the lesion area image includes:

[0021] In the lesion identification database, query the image identification method corresponding to each lesion parameter type, and the image identification range corresponding to each lesion parameter type;

[0022] Based on the target lesion parameter value for each lesion parameter type, image feature marking processing is performed within the image marking range corresponding to each lesion parameter type using the image marking method corresponding to each lesion parameter type, to obtain the lesion feature guidance image of the patient.

[0023] Optionally, generating a feature guidance report for the patient based on the patient's lesion feature guidance image and each of the lesion feature information includes:

[0024] Obtain a lesion feature report template, and based on the target lesion parameter values ​​for each of the lesion parameter types, identify the patient's lesion abnormality information through a lesion abnormality evaluation network;

[0025] The target lesion parameter values ​​for each of the lesion parameter types, the patient's lesion abnormality information, and the patient's lesion feature guidance image are filled into the lesion feature report template to obtain the patient's feature guidance report.

[0026] Furthermore, to achieve the above objectives, the present invention also provides a feature guidance generation system for inner ear endolymphatic effusion lesions, the feature guidance generation system for inner ear endolymphatic effusion lesions comprising:

[0027] The acquisition module is used to acquire image data of the patient in the lesion area, and mark the lesion range image of the lesion area based on the image data;

[0028] The identification module is used to identify lesion parameter information of the lesion area based on the lesion range image and through an image recognition model, and to identify various lesion feature information of the lesion area based on the lesion parameter information of the lesion area.

[0029] The generation module is used to generate a lesion feature guidance image of the patient in the lesion range image based on the lesion feature information, and to generate a feature guidance report of the patient based on the lesion feature guidance image and the lesion feature information.

[0030] Optionally, the acquisition module is specifically used for:

[0031] The image edge recognition network identifies the edge information of the lesion range in the image data;

[0032] In the image data, the edge information of the lesion range is processed by image marking using a preset marking method to obtain the lesion range image of the lesion area.

[0033] Optionally, the identification module is specifically used for:

[0034] Identify sub-region images within the lesion area image, and use an image recognition model to identify the image parameter information corresponding to the sub-region images;

[0035] Based on the image parameter information, the lesion parameter values ​​of each lesion parameter type corresponding to the sub-range image are identified through a lesion parameter identification strategy, and the lesion parameter values ​​of all lesion parameter types are used as the lesion parameter information of the lesion region.

[0036] Optionally, the identification module is specifically used for:

[0037] Based on the image parameter information corresponding to the sub-range image, a two-dimensional model of the lesion is constructed using a two-dimensional modeling strategy;

[0038] Based on the image parameter information, the two-dimensional model of the lesion is subjected to model assignment processing to obtain the lesion structure model of the lesion region;

[0039] Based on the lesion structure model, a two-dimensional feature extraction network is used to extract the regional structure features of the lesion region. Based on the regional structure features, the lesion parameter values ​​of each lesion parameter type are adjusted using the lesion parameter identification strategy to obtain the target lesion parameter values ​​of each lesion parameter type.

[0040] The target lesion parameter values ​​of each of the lesion parameter types are used as the lesion feature information of each lesion region.

[0041] Optionally, the generation module is specifically used for:

[0042] In the lesion identification database, query the image identification method corresponding to each lesion parameter type, and the image identification range corresponding to each lesion parameter type;

[0043] Based on the target lesion parameter value for each lesion parameter type, image feature marking processing is performed within the image marking range corresponding to each lesion parameter type using the image marking method corresponding to each lesion parameter type, to obtain the lesion feature guidance image of the patient.

[0044] Optionally, the generation module is specifically used for:

[0045] Obtain a lesion feature report template, and based on the target lesion parameter values ​​for each of the lesion parameter types, identify the patient's lesion abnormality information through a lesion abnormality evaluation network;

[0046] The target lesion parameter values ​​for each of the lesion parameter types, the patient's lesion abnormality information, and the patient's lesion feature guidance image are filled into the lesion feature report template to obtain the patient's feature guidance report.

[0047] Thirdly, this application provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method described in any one of the first aspects.

[0048] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0049] Fifthly, this application provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0050] This invention provides a method and system for generating feature guidance for inner ear endolymphatic effusion lesions. The method includes: acquiring image data of a patient in the lesion area, and marking the lesion extent image of the lesion area based on the image data; identifying lesion parameter information of the lesion area based on the lesion extent image using an image recognition model, and identifying various lesion feature information of the lesion area based on the lesion parameter information; generating a lesion feature guidance image of the patient in the lesion extent image based on each of the lesion feature information, and generating a feature guidance report of the patient based on the patient's lesion feature guidance image and each of the lesion feature information. This solution first marks the extent of the lesion area, then uses an influence recognition model to identify the lesion parameters, transforming complex image data into parameterizable and data-driven analysis. This solution then combines the transformed lesion parameters with intelligent lesion feature identification, avoiding the limitations, biases, and errors of human identification. This effectively improves the comprehensiveness and accuracy of lesion feature identification. Finally, based on the identified lesion features, the solution generates lesion feature guidance images and reports. These visually demonstrate the image basis for lesion feature identification to medical staff, while also providing comprehensive feature analysis results for the lesion area. Compared to manual feature identification, this solution effectively improves the identification speed and comprehensiveness, and avoids identification anomalies caused by complex feature environments, thus comprehensively improving the feature identification efficiency for inner ear lymphatic effusion. Attached Figure Description

[0051] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart of a method for generating feature guidelines for endolymphatic effusion lesions in the inner ear, provided in an embodiment of the present invention.

[0053] Figure 2 This is a schematic diagram of the feature guidance generation system for inner ear endolymphatic effusion lesions provided in an embodiment of the present invention;

[0054] Figure 3 An internal structural diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0055] The feature guidance generation method for inner ear endolymphatic effusion lesions provided in this invention is applied to a feature guidance generation system for inner ear endolymphatic effusion lesions. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this application. The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or accompanying drawings of this application are used to distinguish different objects, not to describe a particular order.

[0056] 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 this application. 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.

[0057] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0058] The feature guidance generation method for inner ear endolymphatic effusion lesions provided in this application embodiment can be applied in a feature guidance generation application environment for inner ear endolymphatic effusion lesions. This method can be applied to a terminal, a server, or a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, etc. The terminal first marks the lesion area and then uses an influence recognition model to identify the lesion parameters, transforming complex image data into parameterizable and data-driven analysis. This solution then combines the transformed lesion parameters with intelligent lesion feature recognition, avoiding the limitations, biases, and errors of human identification. This effectively improves the comprehensiveness and accuracy of lesion feature recognition. Finally, based on the identified lesion features, the solution generates lesion feature guidance images and reports. These visually demonstrate the image basis for lesion feature recognition to medical staff and include comprehensive feature analysis results for the lesion area. Compared to manual feature recognition, this solution effectively improves recognition speed and comprehensiveness, and avoids recognition anomalies caused by complex feature environments, thus comprehensively improving the efficiency of feature recognition for inner ear lymphatic effusion.

[0059] In one embodiment, Figure 1 As shown, a method for generating feature guidelines for inner ear lymphatic effusion lesions is provided. Taking the application of this method to a terminal as an example, the method includes the following steps:

[0060] Step S101: Obtain image data of the patient in the lesion area, and mark the lesion range image based on the image data.

[0061] In this embodiment, the terminal receives MRI (Magnetic Resonance Imaging) image data transmitted from a magnetic resonance imaging (MRI) device, thereby acquiring image data of the patient's inner ear lymphatic effusion lesion area. The terminal then performs standardization processing on the image data, including denoising, normalization, and data enhancement (such as rotation and flipping). It also labels the image data, distinguishing between normal and lesion areas. This yields the image data of the patient in the lesion area. Based on the image data, the terminal then marks the lesion extent image within the lesion area. This lesion extent image is the image encompassed by the edge information of the lesion extent within the lesion area. The specific labeling process will be described in detail later.

[0062] Step S102: Based on the lesion range image, the lesion parameter information of the lesion area is identified through the image recognition model, and based on the lesion parameter information of the lesion area, the lesion feature information of each lesion area is identified.

[0063] In this embodiment, the terminal identifies lesion parameter information of the lesion area based on the lesion range image using an image recognition model, and identifies various lesion feature information of the lesion area based on the lesion parameter information. The image recognition model uses a Convolutional Neural Network (CNN) as the base model, combined with an attention mechanism and a Feature Pyramid Network (FPN) for image feature recognition. This image recognition model incorporates transfer learning technology, utilizing pre-trained models (such as ResNet and EfficientNet) to accelerate model training and improve generalization ability. The lesion parameter information includes lesion parameter values ​​for various lesion parameter types. These lesion parameter types include, but are not limited to, inner ear endolymphatic fluid ratio type, vestibular total lymph area area type, and central low-signal endolymphatic area area type. Each lesion feature information is the target lesion parameter value for each lesion parameter type obtained by optimizing the lesion parameter values ​​for each lesion parameter type. For example, the parameter value corresponding to the inner ear endolymphatic fluid ratio type is represented by the area of ​​the endolymphatic fluid. The total area of ​​lymphatic effusion is calculated, and this value can be used to determine and classify the severity of lymphatic effusion in the inner ear. Because single-image recognition is susceptible to interference, inaccurate image information, and blurred image boundaries, the obtained lesion parameter values ​​may be abnormal. Therefore, this solution addresses complex feature scenarios by employing a parameter optimization method derived from innovative experimental research and practical results. This method optimizes the lesion parameter values ​​for each lesion type, thereby avoiding the influence of interference and improving the accuracy and practicality of the obtained lesion parameter values. The specific optimization method will be explained in detail later.

[0064] Step S103: Based on the lesion feature information, generate a lesion feature guidance image for the patient in the lesion range image, and generate a feature guidance report for the patient based on the lesion feature guidance image and the lesion feature information.

[0065] In this embodiment, the terminal generates a lesion feature guidance image for the patient within the lesion area image based on the lesion feature information, and generates a feature guidance report for the patient based on the lesion feature guidance image and the lesion feature information. The specific generation process will be described in detail later.

[0066] Based on the above scheme, the lesion area is first marked, and then the lesion parameter information of the lesion area is identified through an influence recognition model. This transforms complex image data into data content that can be parameterized and analyzed. Then, this scheme combines the transformed lesion parameter information to perform intelligent identification of lesion features, avoiding the limitations, biases, and errors of human identification. This effectively improves the comprehensiveness and accuracy of feature identification of the lesion area. Finally, based on the identified lesion feature information, this scheme generates lesion feature guidance images and feature guidance reports. It can visually show medical staff the image basis for lesion feature identification and includes comprehensive feature analysis results of the lesion area. It can not only show medical staff comprehensive feature analysis results but also visually show users the feature analysis basis. Compared with human feature identification methods, this scheme effectively improves the identification speed and comprehensiveness and avoids identification anomalies caused by complex feature environments, thus comprehensively improving the feature identification efficiency of inner ear lymph effusion.

[0067] Optionally, based on image data, the lesion extent image of the lesion area is marked, including: identifying the lesion extent edge information in the image data through an image edge recognition network; and performing image marking processing on the lesion extent edge information in the image data through a preset marking method to obtain the lesion extent image of the lesion area.

[0068] In this embodiment, the terminal uses an image edge recognition network to identify the edge information of the lesion range in the image data. This lesion range edge information refers to the area where the inner ear lymphatic effusion lesion is located. Then, the terminal performs image labeling processing on the lesion range edge information in the image data using a preset labeling method to obtain an image of the lesion range. The image edge recognition network is a neural network constructed based on the Canny operator image edge recognition algorithm. The preset labeling method is a method of prominently marking the location information of the identified lesion range edge information in the image data.

[0069] Based on the above scheme, by marking the area of ​​the lesion region of the inner ear lymphatic effusion in the image data, not only is the accuracy of subsequent identification of the lesion region improved, but also the interference information of other regions' images on the image feature identification of the region can be effectively reduced. Finally, by optimizing the range, the efficiency of image recognition can also be improved.

[0070] Optionally, based on the lesion range image, the lesion parameter information of the lesion area is identified through an image recognition model, including: identifying sub-range images in the lesion range image, and identifying the image parameter information corresponding to the sub-range images through the image recognition model; based on the image parameter information, identifying the lesion parameter values ​​of each lesion parameter type corresponding to the sub-range images through a lesion parameter identification strategy, and using the lesion parameter values ​​of all lesion parameter types as the lesion parameter information of the lesion area.

[0071] In this embodiment, the terminal identifies sub-region images within the lesion area image and uses an image recognition model to identify the image parameter information corresponding to the sub-region images. This image parameter information includes image structure parameters of each structural portion of the target area image surrounded by the lesion area edge information, where each structural portion represents the range of that range within each ear structure of the inner ear.

[0072] Then, based on the image parameter information, the terminal identifies the lesion parameter values ​​of each lesion parameter type corresponding to the sub-range image through a lesion parameter identification strategy. This lesion parameter identification strategy includes the correspondence between the lesion parameter values ​​of different lesion parameter types and the image structural parameter ranges of each structural part. The terminal identifies the lesion parameter values ​​of each lesion parameter type through range adaptation.

[0073] Finally, the terminal uses the lesion parameter values ​​of all lesion parameter types as the lesion parameter information of the lesion region.

[0074] Based on the above scheme, the image recognition model constructed using this scheme can accurately identify image parameters of different structural parts under complex environmental conditions. A Convolutional Neural Network (CNN) is used as the basic model, combined with an attention mechanism and a Feature Pyramid Network (FPN) to enhance the model's ability to identify lesion areas. A multi-task learning framework is designed to simultaneously complete the tasks of lesion area localization and classification. Transfer learning techniques are introduced, utilizing pre-trained models (such as ResNet and EfficientNet) to accelerate model training and improve generalization ability. This comprehensively improves the accuracy of image parameter information recognition and classification.

[0075] Optionally, based on the lesion parameter information of the lesion region, the lesion feature information of each lesion region is identified, including: constructing a two-dimensional model of the lesion based on the image parameter information corresponding to the sub-range image through a two-dimensional modeling strategy; performing model assignment processing on the two-dimensional model of the lesion based on the image parameter information to obtain the lesion structure model of the lesion region; extracting the regional structure features of the lesion region through a two-dimensional feature extraction network based on the lesion structure model, and adjusting the lesion parameter values ​​of each lesion parameter type based on the regional structure features through a lesion parameter identification strategy to obtain the target lesion parameter values ​​of each lesion parameter type; and using the target lesion parameter values ​​of each lesion parameter type as the lesion feature information of each lesion region.

[0076] In this embodiment, the terminal constructs a two-dimensional model of the lesion based on the image parameter information corresponding to the sub-range image and through a two-dimensional modeling strategy. This two-dimensional modeling strategy involves constructing a two-dimensional image model using a two-dimensional modeling program, with the sub-image data enclosed by the edge information of the lesion range as the image reference and the various image parameter information as the model parameters.

[0077] Then, based on the image parameter information, the terminal performs model assignment processing on the two-dimensional lesion model to obtain the lesion structure model of the lesion region. Next, based on the lesion structure model, the terminal extracts the regional structural features of the lesion region through a two-dimensional feature extraction network. This two-dimensional feature extraction network is an image recognition model capable of recognizing image feature information of a two-dimensional image model; that is, it is the recognition model obtained by training the image recognition model designed in this scheme with different training samples.

[0078] Next, based on the regional structural features, the terminal adjusts the lesion parameter values ​​for each lesion parameter type using a lesion parameter identification strategy to obtain the target lesion parameter values ​​for each lesion parameter type. Specifically, the adjustment process is as follows: the terminal identifies new image structural parameters for each structural part based on the regional structural features of the extracted lesion area (the terminal presets the correspondence between different regional structural features and the image structural parameters of different structural parts, and then identifies the new image structural parameters for each structural part through the above correspondence), replaces the image parameter information with the new image structural parameters for each structural part, and returns to execute the step of identifying the lesion parameter values ​​for each lesion parameter type corresponding to the sub-range image based on the image parameter information and the lesion parameter identification strategy, to obtain new lesion parameter values ​​for each lesion parameter type. Then, the terminal uses the new lesion parameter values ​​for each lesion parameter type as the target lesion parameter values ​​for each lesion parameter type.

[0079] Finally, the terminal uses the target lesion parameter values ​​of each lesion parameter type as the lesion feature information of each lesion region.

[0080] Based on the above scheme, by constructing a two-dimensional image structure model, the lesion parameter values ​​of each lesion parameter type in the image data are identified for a second time, thereby obtaining the target lesion parameter values ​​of each lesion parameter type, which improves the accuracy and practicality of the obtained target lesion parameter values ​​of each lesion parameter type.

[0081] Optionally, based on the lesion feature information, a lesion feature guidance image for the patient is generated in the lesion range image, including: querying the image identification method corresponding to each lesion parameter type and the image identification range corresponding to each lesion parameter type in the lesion identification database; based on the target lesion parameter value of each lesion parameter type, performing image feature marking processing in the image identification range corresponding to each lesion parameter type through the image identification method corresponding to each lesion parameter type to obtain the patient's lesion feature guidance image.

[0082] In this embodiment, the terminal queries the lesion identification database for the image identification method corresponding to each lesion parameter type and the image identification range corresponding to each lesion parameter type. The lesion identification database stores the correspondence between different lesion parameter types, image identification methods, and image identification ranges, where the image identification range refers to the range of each structural part of the ear structure.

[0083] Then, based on the target lesion parameter value for each lesion parameter type, the terminal performs image feature marking processing within the image identifier range corresponding to each lesion parameter type using the image identifier method corresponding to each lesion parameter type, to obtain the patient's lesion feature guidance image.

[0084] Based on the above scheme, different image labeling methods are used to label the target lesion parameter values ​​of different lesion parameter types, thereby improving the generation efficiency and comprehensiveness of the generated lesion feature guidance images.

[0085] Optionally, based on the patient's lesion feature guidance image and various lesion feature information, a patient feature guidance report is generated, including: obtaining a lesion feature report template, and identifying the patient's lesion abnormality information through a lesion abnormality evaluation network based on the target lesion parameter values ​​of each lesion parameter type; filling the lesion feature report template with the target lesion parameter values ​​of each lesion parameter type, the patient's lesion abnormality information, and the patient's lesion feature guidance image to obtain the patient's feature guidance report.

[0086] In this embodiment, the terminal acquires a lesion feature report template and, based on the target lesion parameter values ​​for each lesion parameter type, identifies the patient's lesion abnormality information through a lesion abnormality evaluation network. Then, the terminal fills the lesion feature report template with the target lesion parameter values ​​for each lesion parameter type, the patient's lesion abnormality information, and the patient's lesion feature guidance image to obtain the patient's feature guidance report. The lesion abnormality evaluation network is a classifier neural network, which can identify the patient's corresponding lesion abnormality information based on the range of lesion parameter values ​​for different lesion parameter types. This lesion abnormality information includes the patient's lesion severity and lesion type. The lesion feature report template is a pre-set template on the terminal by the staff. This template includes the lesion feature guidance image, lesion abnormality information, and the filling positions corresponding to the target lesion parameter values ​​for each lesion parameter type. The terminal obtains the patient's feature guidance report by directly filling the data into the corresponding filling positions.

[0087] Based on the above scheme, this scheme generates lesion feature guidance images and feature guidance reports based on the identified lesion feature information. It can not only visually show medical staff the image basis for lesion feature identification, but also include the comprehensive feature analysis results of the lesion area. It can not only show medical staff the comprehensive feature analysis results, but also visually show users the feature analysis basis. Compared with the method of feature identification, this scheme effectively improves the identification speed and identification comprehensiveness, and can avoid the identification abnormality caused by the complexity of the feature environment, thereby comprehensively improving the feature identification efficiency of inner ear lymph effusion.

[0088] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0089] Based on the same inventive concept, this application also provides a feature guidance generation system for inner ear endolymphatic effusion lesions to implement the feature guidance generation method for inner ear endolymphatic effusion lesions described above. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more feature guidance generation system embodiments for inner ear endolymphatic effusion lesions provided below can be found in the limitations of the feature guidance generation method for inner ear endolymphatic effusion lesions described above, and will not be repeated here.

[0090] Further reference Figure 2 As a response to the above Figure 1 The present application provides an embodiment of a feature guidance generation system 200 for inner ear endolymphatic effusion lesions, which includes an acquisition module 210, an identification module 220, and a generation module 230, wherein:

[0091] The acquisition module 210 is used to acquire image data of the patient in the lesion area, and mark the lesion range image of the lesion area based on the image data;

[0092] The identification module 220 is used to identify lesion parameter information of the lesion area based on the lesion range image and through an image recognition model, and to identify various lesion feature information of the lesion area based on the lesion parameter information of the lesion area.

[0093] The generation module 230 is used to generate a lesion feature guidance image of the patient in the lesion range image based on the lesion feature information, and to generate a feature guidance report of the patient based on the lesion feature guidance image and the lesion feature information.

[0094] Optionally, the acquisition module 210 is specifically used for:

[0095] The image edge recognition network identifies the edge information of the lesion range in the image data;

[0096] In the image data, the edge information of the lesion range is processed by image marking using a preset marking method to obtain the lesion range image of the lesion area.

[0097] Optionally, the identification module 220 is specifically used for:

[0098] Identify sub-region images within the lesion area image, and use an image recognition model to identify the image parameter information corresponding to the sub-region images;

[0099] Based on the image parameter information, the lesion parameter values ​​of each lesion parameter type corresponding to the sub-range image are identified through a lesion parameter identification strategy, and the lesion parameter values ​​of all lesion parameter types are used as the lesion parameter information of the lesion region.

[0100] Optionally, the identification module 220 is specifically used for:

[0101] Based on the image parameter information corresponding to the sub-range image, a two-dimensional model of the lesion is constructed using a two-dimensional modeling strategy;

[0102] Based on the image parameter information, the two-dimensional model of the lesion is subjected to model assignment processing to obtain the lesion structure model of the lesion region;

[0103] Based on the lesion structure model, a two-dimensional feature extraction network is used to extract the regional structure features of the lesion region. Based on the regional structure features, the lesion parameter values ​​of each lesion parameter type are adjusted using the lesion parameter identification strategy to obtain the target lesion parameter values ​​of each lesion parameter type.

[0104] The target lesion parameter values ​​of each of the lesion parameter types are used as the lesion feature information of each lesion region.

[0105] Optionally, the generation module 230 is specifically used for:

[0106] In the lesion identification database, query the image identification method corresponding to each lesion parameter type, and the image identification range corresponding to each lesion parameter type;

[0107] Based on the target lesion parameter value for each lesion parameter type, image feature marking processing is performed within the image marking range corresponding to each lesion parameter type using the image marking method corresponding to each lesion parameter type, to obtain the lesion feature guidance image of the patient.

[0108] Optionally, the generation module 230 is specifically used for:

[0109] Obtain a lesion feature report template, and based on the target lesion parameter values ​​for each of the lesion parameter types, identify the patient's lesion abnormality information through a lesion abnormality evaluation network;

[0110] The target lesion parameter values ​​for each of the lesion parameter types, the patient's lesion abnormality information, and the patient's lesion feature guidance image are filled into the lesion feature report template to obtain the patient's feature guidance report.

[0111] The features of the aforementioned inner ear endolymphatic effusion lesions guide the generation of various modules in the system, which can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0112] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown, the computer device includes a processor, memory, communication interface, display screen, and input system connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for generating characteristic guidelines for inner ear lymphatic effusion lesions. The display screen can be an LCD screen or an e-ink screen. The input system can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0113] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0114] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any one of the first aspects.

[0115] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0116] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0117] It should be noted that the patient information (including but not limited to patient device information, patient personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the patient or fully authorized by all parties.

[0118] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), data processing logic devices based on quantum computing, and the like.

[0119] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0120] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for generating characteristic indicators for endolymphatic effusion lesions of the inner ear, characterized in that, The method includes: Acquire image data of the patient in the lesion area, and based on the image data, mark the lesion extent in the lesion area; Based on the lesion range image, the lesion parameter information of the lesion area is identified by the image recognition model, and based on the lesion parameter information of the lesion area, the lesion feature information of each lesion area is identified; Based on the lesion feature information, a lesion feature guidance image of the patient is generated in the lesion range image, and a feature guidance report of the patient is generated based on the lesion feature guidance image of the patient and the lesion feature information.

2. The method according to claim 1, characterized in that, The step of marking the lesion extent image based on the image data includes: The image edge recognition network identifies the edge information of the lesion range in the image data; In the image data, the edge information of the lesion range is processed by image marking using a preset marking method to obtain the lesion range image of the lesion area.

3. The method according to claim 1, characterized in that, The step of identifying lesion parameter information of the lesion area based on the lesion range image using an image recognition model includes: Identify sub-region images within the lesion area image, and use an image recognition model to identify the image parameter information corresponding to the sub-region images; Based on the image parameter information, the lesion parameter values ​​of each lesion parameter type corresponding to the sub-range image are identified through a lesion parameter identification strategy, and the lesion parameter values ​​of all lesion parameter types are used as the lesion parameter information of the lesion region.

4. The method according to claim 3, characterized in that, The process of identifying various lesion features in the lesion region based on lesion parameter information includes: Based on the image parameter information corresponding to the sub-range image, a two-dimensional model of the lesion is constructed using a two-dimensional modeling strategy; Based on the image parameter information, the two-dimensional model of the lesion is subjected to model assignment processing to obtain the lesion structure model of the lesion region; Based on the lesion structure model, a two-dimensional feature extraction network is used to extract the regional structure features of the lesion region. Based on the regional structure features, the lesion parameter values ​​of each lesion parameter type are adjusted using the lesion parameter identification strategy to obtain the target lesion parameter values ​​of each lesion parameter type. The target lesion parameter values ​​of each of the lesion parameter types are used as the lesion feature information of each lesion region.

5. The method according to claim 1, characterized in that, The step of generating a lesion feature guidance image for the patient based on the lesion feature information includes: In the lesion identification database, query the image identification method corresponding to each lesion parameter type, and the image identification range corresponding to each lesion parameter type; Based on the target lesion parameter value for each lesion parameter type, image feature marking processing is performed within the image marking range corresponding to each lesion parameter type using the image marking method corresponding to each lesion parameter type, to obtain the lesion feature guidance image of the patient.

6. The method according to claim 4, characterized in that, The process of generating a feature guidance report for the patient based on the patient's lesion feature guidance image and the lesion feature information includes: Obtain the lesion feature report template, and based on the target lesion parameter values ​​of each lesion parameter type, identify the patient's lesion abnormality information through the lesion abnormality evaluation network; The target lesion parameter values ​​for each of the lesion parameter types, the patient's lesion abnormality information, and the patient's lesion feature guidance image are filled into the lesion feature report template to obtain the patient's feature guidance report.

7. A feature-guided generation system for inner ear endolymphatic effusion lesions, characterized in that, The system includes: The acquisition module is used to acquire image data of the patient in the lesion area, and mark the lesion range image of the lesion area based on the image data; The identification module is used to identify lesion parameter information of the lesion area based on the lesion range image and through an image recognition model, and to identify various lesion feature information of the lesion area based on the lesion parameter information of the lesion area. The generation module is used to generate a lesion feature guidance image of the patient in the lesion range image based on the lesion feature information, and to generate a feature guidance report of the patient based on the lesion feature guidance image and the lesion feature information.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

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

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images