A method and application of constructing a bone prediction model based on ROI-HU database

Abstract

The present invention provides a method and application for constructing a bone prediction model based on a ROI-HU database, performing image segmentation based on bone CT image data of different age groups, and constructing a corresponding ROI-HU database based on the pixel data of the constructed ROI area and the extracted Hu value. Regression training is performed based on the data features in the database to construct a bone prediction model, which can be used to predict the bone properties of patients' bones in different age groups. Patients can understand the bone density value they should have in the corresponding age group, and can predict the bone properties they should have in the next stage, so that patients can track and understand their own bone development status in a timely manner. The prediction of bone properties is based on two indicators, and patients can also be informed of factors such as whether their bones have lesions, so that they can start treatment in time and improve the accuracy of prediction.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method for constructing a skeleton prediction model based on a ROI-HU database, an application system thereof, an electronic device, and a computer-readable storage medium. Background Art

[0002] The changes in bone properties at different age groups mainly include the following aspects:

[0003] Bone growth: Bone growth is rapid during childhood and adolescence, but growth slows down and eventually stops in adulthood.

[0004] Bone density: As we age, bone density gradually decreases, especially in women after menopause;

[0005] Osteoporosis: The mineral content in the bones of the elderly decreases, the bone tissue structure becomes loose, and fractures are more likely to occur;

[0006] Osteomalacia: Insufficient calcium intake during childhood and adolescence may lead to osteomalacia, affecting normal bone development;

[0007] Fracture healing: With age, fracture healing slows down and the healing effect is poor;

[0008] Bone and joint diseases: The elderly are prone to bone and joint diseases such as osteoarthritis and osteoporosis;

[0009] Bone deformation: Certain diseases or bad living habits may cause bone deformation, such as rickets, malunion of fractures, etc.

[0010] Bone function: As we age, bone function gradually declines, such as decreased joint mobility and weakened muscle strength.

[0011] Bone properties not only include hardness, but also bone density is the main attribute indicator. Bone density changes significantly with age. In clinical bone examinations, especially bone CT angiography, it can be found that the bone imaging data of orthopedic patients at each stage will undergo corresponding bone property changes, showing a certain change pattern. Therefore, if patients of different age groups can understand their bone properties at the current stage in advance, they can continuously check whether the changes in their bone properties, especially bone density, conform to the change patterns of bone big data, so that patients / users can compare and know whether their bones have lesions and other factors, and intervene in treatment in time.

[0012] However, existing bone prediction technologies primarily rely on prediction models constructed from limited clinical data or require the use of clinical monitoring technology to assist in prediction. Therefore, they are unable to concisely, conveniently, and quickly predict bone properties, especially bone density, at various stages of development. For example, CN118538411A discloses an intelligent analysis and height prediction method and system based on X-rays of children's metacarpal and carpal bones. Principal component analysis is used to analyze factors including bone growth rate, bone density, and the interval between decreases in epiphyseal width. A random forest algorithm is then used to analyze the correlations between these factors to identify key factors influencing growth prediction. Based on a micro-biomarker method and monitoring system, corresponding monitoring markers are deployed in key areas of the skeleton and simultaneously monitored as these markers change with the epiphyseal line and cartilage plate to obtain real-time monitoring data of skeletal development. This real-time monitoring data of skeletal development is then processed and analyzed, including using a wavelet transform algorithm to extract features from the changes and using a machine learning algorithm to identify abnormal patterns in skeletal development to detect pathological changes. This method requires the use of clinical biomarker tracking and requires multiple steps of skeletal development prediction calculation and analysis, making the prediction process relatively complex.

[0013] In addition, existing bone prediction cannot integrate two or more bone prediction factors to predict bone attributes. Most of them are based on big data under a single indicator, and the prediction accuracy is not high. Summary of the Invention

[0014] In order to solve the technical problems existing in the prior art, the present invention provides the following technical solutions:

[0015] In one aspect, a method for constructing a skeleton prediction model based on a ROI-HU database is provided. The method is implemented by an electronic device and includes:

[0016] S1. Obtain CT image data of bones of several orthopedic patients at different age groups and perform preprocessing to obtain a bone CT image dataset in a standard format;

[0017] S2, traversing and segmenting the ROI tissue region of each skeletal CT image picture in the skeletal CT image data set, extracting ROI pixel data of the ROI tissue region and binding it to the age value marked on the corresponding picture;

[0018] S3, traversing the HU value of each bone CT image picture in the bone CT image data set, and binding the age value marked on the corresponding picture;

[0019] S4, counting the ROI pixel data and HU values ​​extracted under different age groups, and constructing a ROI-HU database corresponding to the indicators of different age groups;

[0020] S5. Extract feature sets corresponding to indicators of different age groups from the ROI-HU database, perform statistical regression training on the feature sets, and construct a bone prediction model.

[0021] Optionally, S1, obtaining CT image data of bones of a number of orthopedic patients at different age groups and preprocessing them to obtain a bone CT image dataset in a standard format, including:

[0022] Log in to the backend database and filter the CT image data of bones of several orthopedic patients in different age groups;

[0023] Cleaning the screened CT image data and sorting them according to age values;

[0024] Based on the Canny algorithm, the bone region edge in the CT image data is identified, a vector mask is added along the bone region edge, and the bone region portion where the vector mask is located is segmented, and a bone region mask image in a BMP format is outputted and obtained;

[0025] Performing pixel value calibration on the skeleton region mask image and the original CT image data, and calibrating the pixel value of the corresponding skeleton region mask image based on the resolution of the CT image data;

[0026] After calibration, an image set consisting of masked images of bone regions of different age groups is saved to generate the bone CT image dataset.

[0027] Optionally, S2, traversing and segmenting the ROI tissue region of each skeletal CT image picture in the skeletal CT image dataset, extracting ROI pixel data of the ROI tissue region and binding it to the age value marked on the corresponding picture, includes:

[0028] Traversing the bone CT image pictures in the bone CT image dataset in order;

[0029] In the bone CT image, a corresponding ROI mark is added to the bone tissue region of interest to obtain the ROI tissue region;

[0030] Using an image segmentation algorithm to perform image segmentation on the ROI tissue region, obtain a segmented image of the ROI tissue region and convert it into a corresponding image matrix;

[0031] Extracting pixel data of the position of the ROI mark from the image matrix according to the relative position of the ROI mark to obtain ROI pixel data of the ROI tissue area;

[0032] The age value corresponding to the bone CT image is bound to the ROI pixel data of the ROI tissue region extracted from the bone CT image.

[0033] Optionally, S3, traversing the HU value of each bone CT image picture in the bone CT image dataset and binding the corresponding age value marked on the picture, including:

[0034] Reading metadata of the bone CT image pictures in the bone CT image dataset in order;

[0035] Parsing the metadata to obtain an adjustment slope k and an adjustment intercept l of the bone CT image;

[0036] Calculate the voxel HU value of the bone CT image:

[0037] HU=C*k+l, C is the pixel value;

[0038] The HU value corresponding to the bone CT image is bound to the age value of the patient corresponding to the bone CT image.

[0039] Optionally, S4, counting the ROI pixel data and HU values ​​extracted under different age groups, and constructing an ROI-HU database corresponding to indicators of different age groups, including:

[0040] Reading pixel data in the bone CT image data set, and counting the ROI pixel data bound to different age values ​​in each age group;

[0041] Reading the HU values ​​in the bone CT image data set, and counting the HU values ​​bound to different age values ​​in each age group;

[0042] Fusing the ROI pixel data and the HU value bound to different age values ​​in each age group to generate ROI-HU data under different age group indicators;

[0043] The ROI-HU data under different age group indicators are written into a preset structure database in order to obtain the ROI-HU database.

[0044] Optionally, S5, extracting feature sets corresponding to indicators of different age groups from the ROI-HU database, and performing statistical regression training on the feature sets to construct a bone prediction model, including:

[0045] Read the ROI-HU database to obtain the ROI pixel data and the HU value corresponding to different age group indicators, and import them into a preset Histogram table to generate the corresponding ROI histogram and HU histogram under each age group indicator;

[0046] Fitting the ROI histogram and HU histogram under each age group index to obtain the HU-Histogram curve under each age group index;

[0047] Extracting the histogram curve features of the HU-histogram curve under each age group indicator and performing ordered statistics to obtain a feature set consisting of the histogram curve features under each age group indicator; the histogram curve features include bone attributes under the corresponding age group indicator, and the bone attributes include bone density;

[0048] Dividing the feature set into a training set and a validation set according to a preset ratio;

[0049] Importing the training set into a preset statistical regression model, performing learning and training on the histogram curve characteristics under the indicators of each age group, and generating the bone prediction model for predicting the bone properties of bones in different age groups;

[0050] Validating the skeleton prediction model using the validation set:

[0051] If the verification is passed, the skeleton prediction model is deployed and applied;

[0052] Otherwise, repeat S1-S5.

[0053] On the other hand, an application system is provided, the application system being used to implement the method for constructing a skeleton prediction model based on the ROI-HU database, the application system comprising:

[0054] Backend server, on which are deployed:

[0055] An electronic medical record system to record orthopedic patients’ medical history information, including age;

[0056] The bone density prediction module is used to calculate the age group Ta to which the orthopedic patient's current age belongs, call the pre-deployed bone prediction model, predict the bone attributes matching the current age group Ta, and output the corresponding bone density;

[0057] A bone attribute reasoning module is used to record the bone attributes in the next age group Tb predicted and output by the bone density prediction module, and generate a corresponding bone change prediction report;

[0058] Backend database, used for background data storage;

[0059] The patient-side APP is used to log in to the backend server, view the bone density corresponding to the current age group Ta, and view the bone change prediction report;

[0060] The medical terminal is used to log in to the backend server, check the patient's current bone density, and determine whether to issue corresponding bone care instructions;

[0061] The patient-side APP and the medical-side APP are respectively connected to the background server for communication.

[0062] On the other hand, an electronic device is provided, comprising: a processor; a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, any one of the methods for constructing a skeleton prediction model based on the ROI-HU database is implemented.

[0063] On the other hand, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement any of the above-mentioned methods for constructing a bone prediction model based on the ROI-HU database.

[0064] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0065] Based on the implementation of the present invention, the present invention carries out image segmentation based on the skeletal CT imaging data of different age groups, and builds corresponding ROI-HU database based on the pixel data of constructed ROI region and the Hu value extracted. Carry out regression training based on the data features in the database, build and obtain a skeleton prediction model, which can be used to predict the skeletal properties of patient skeletons under different age groups. Allow the patient to understand the bone density value that should be possessed under the corresponding age group, and can predict the skeletal properties that should be possessed in the next stage, so that the patient can track and understand the skeletal development state of the patient in time.

[0066] The present invention can predict bone properties based on two indicators, allowing patients to identify factors such as whether their bones are pathological, allowing for timely treatment and improved prediction accuracy. The present invention eliminates the need for biomarkers, simplifying prediction. It also allows patients to understand the changing patterns of bone density at different ages, facilitating timely tracking of bone density. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0068] Figure 1This is a flow chart of a method for constructing a skeleton prediction model based on a ROI-HU database provided by an embodiment of the present invention;

[0069] Figure 2 1 is a schematic diagram of marking for edge detection of a skeleton region provided by an embodiment of the present invention;

[0070] Figure 3 is a schematic diagram of adding a vector mask to a detected marked area provided by an embodiment of the present invention;

[0071] Figure 4 This is a histogram (Histogram) of ROI pixel data and HU values ​​extracted under different age groups provided by an embodiment of the present invention;

[0072] Figure 5 Schematic diagram of a HU-Histogram curve fitting provided by an embodiment of the present invention;

[0073] Figure 6 This is a block diagram of an application system provided by an embodiment of the present invention;

[0074] Figure 7 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0075] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0076] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0077] In the embodiments of the present invention, the terms "image" and "picture" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same. The terms "of," "corresponding," and "corresponding" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same.

[0078] In the embodiments of the present invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0079] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0080] In the present invention, pixel value refers to the grayscale value or color value of each smallest unit in the image, which is usually used to describe the brightness or color information of the image. HU value refers to a texture feature value used in image processing, which is calculated based on the grayscale co-occurrence matrix of the image (HU value is generally the unit of CT value, just like the unit of weight is g or kg, HU value is also the unit of CT value. CT is a black and white image of different brain tissues due to their different X-ray absorption conditions. The higher the CT value, the higher the value before HU (which can also be used to characterize tissue density), indicating that the CT value is also higher, which appears as a whitish signal on CT. For example, if there is calcification in the bones, the CT value of calcification may be 90-100HU, and the CT value is usually relatively high, which appears as a white high-density change on CT. Similarly, for cerebrospinal fluid and water, the CT value is relatively low, which appears as a darker black appearance on CT. Specifically, the HU value on the CT image can be identified and read in combination with the clinical radiological CT examination system), which is used to characterize the thickness, direction and other characteristics of the image texture.

[0081] The embodiment of the present invention provides a method for constructing a skeleton prediction model based on a ROI-HU database, which can be implemented by an electronic device, which can be a terminal or a server. Figure 1 The flowchart of the method for building a skeleton prediction model based on the ROI-HU database is shown. The processing flow of the method may include the following steps:

[0082] S1. Obtain CT image data of bones of several orthopedic patients at different age groups and perform preprocessing to obtain a bone CT image dataset in a standard format;

[0083] S2, traversing and segmenting the ROI tissue region of each skeletal CT image picture in the skeletal CT image data set, extracting ROI pixel data of the ROI tissue region and binding it to the age value marked on the corresponding picture;

[0084] S3, traversing the HU value of each bone CT image picture in the bone CT image data set, and binding the age value marked on the corresponding picture;

[0085] S4, counting the ROI pixel data and HU values ​​extracted under different age groups, and constructing a ROI-HU database corresponding to the indicators of different age groups;

[0086] S5. Extract feature sets corresponding to indicators of different age groups from the ROI-HU database, perform statistical regression training on the feature sets, and construct a bone prediction model.

[0087] On bone CT images, patients of different age groups exhibit different Hu values ​​and pixel values. Therefore, this invention uses the big data features of the fusion of two CT dimensions to identify bone properties and bone density in different age groups. The bone tissue ROI region of interest is divided by the bone CT image, and its pixel values ​​are extracted. At the same time, the Hu value of the image is read, and data statistics are performed in the corresponding CT dimension.

[0088] Because pixel values ​​reflect changes in corresponding bone properties, including bone density, and Hu values ​​can also be used to characterize bone density, this paper combines data from these two different formats to construct an ROI-Hu database for different age groups. This allows statistical analysis of both pixel value data and Hu values ​​for patients' bones across different age groups.

[0089] During regression statistical training, pixel values ​​and HU features under different age group indicators are extracted to train and learn the patient's bone properties, especially the performance characteristics of bone density, so that the model can recognize bone properties in different age groups. This is used to predict bone properties in different age groups, and to judge and identify whether the bone density of patients in the current age group meets the characteristics recognized by the model, which is used for judging diseased bone lesions and for prediction.

[0090] like Figure 2 and Figure 3 As shown, optionally, S1, obtains skeletal CT image data of several orthopedic patients at different age groups and preprocesses them to obtain a skeletal CT image dataset in a standard format, including:

[0091] Log in to the backend database and filter the CT image data of bones of several orthopedic patients in different age groups;

[0092] Cleaning the screened CT image data and sorting them according to age values;

[0093] Based on the Canny algorithm, the bone region edge in the CT image data is identified, a vector mask is added along the bone region edge, and the bone region portion where the vector mask is located is segmented, and a bone region mask image in a BMP format is outputted and obtained;

[0094] Performing pixel value calibration on the skeleton region mask image and the original CT image data, and calibrating the pixel value of the corresponding skeleton region mask image based on the resolution of the CT image data;

[0095] After calibration, an image set consisting of masked images of bone regions of different age groups is saved to generate the bone CT image dataset.

[0096] The orthopedic examination backend server database contains skeletal CT image data from several orthopedic patients. Therefore, administrators can filter and clean skeletal CT image data from the backend database for orthopedic patients of different age groups. The administrator is responsible for the cleaning process, removing images that do not meet the requirements.

[0097] Bone CT images of different ages can be sorted in order according to the patient file information recorded on the image. This solution sorts the collected CT images according to age values, making it easier for subsequent systems to identify them.

[0098] However, bone tissue only occupies a portion of the CT image. Therefore, edge detection and identification of the bone region are necessary. This solution utilizes the Canny algorithm for edge detection and marks the edges of the bone region. Mask segmentation is also used to segment the bone region and save it as a separate bone CT film for subsequent image recognition processing.

[0099] like Figure 2 As shown, the edges of the bone regions identified in the CT image data are marked.

[0100] like Figure 3 As shown, a vector mask is added along the edge of the bone area.

[0101] The mask segmentation component can be downloaded from the system, and specifically, the mask tool component in Photoshop can be used. Here, after detecting the corresponding edge area, the corresponding vector mask is added, and the bone area of ​​that part is segmented to obtain the corresponding bone area mask image. In order to ensure pixel consistency between the mask image and the CT image, the pixel values ​​of the segmented mask image need to be calibrated. At this time, the segmented template image is corrected based on the pixel values ​​of the original CT image to ensure clarity.

[0102] Some steps for mask segmentation can be referred to:

[0103] 1. Data preprocessing:

[0104] Convert DICOM format to an image format that can be used for processing, such as NIfTI or PNG.

[0105] Normalize the image to ensure that the pixel values ​​are between 0 and 255.

[0106] 2. Mask segmentation:

[0107] CT images are segmented using image segmentation algorithms, such as threshold segmentation, region growing, edge detection, etc.

[0108] The segmentation results are optimized to remove noise and small isolated areas.

[0109] Generate BMP mask data:

[0110] Convert the segmented mask to BMP format.

[0111] Ensure that the resolution of the BMP mask data matches the original CT image.

[0112] 3. Save data:

[0113] Save the processed CT image data (such as NIfTI or PNG format) and the generated BMP mask data separately.

[0114] When saving files, ensure that they are named in a standardized manner to facilitate subsequent processing and use.

[0115] 4. Post-processing:

[0116] Check the saved data to ensure that the segmentation quality meets the requirements.

[0117] If necessary, further image enhancement or segmentation adjustments were performed.

[0118] Optionally, S2, traversing and segmenting the ROI tissue region of each skeletal CT image picture in the skeletal CT image dataset, extracting ROI pixel data of the ROI tissue region and binding it to the age value marked on the corresponding picture, includes:

[0119] Traversing the bone CT image pictures in the bone CT image dataset in order;

[0120] In the bone CT image, a corresponding ROI mark is added to the bone tissue region of interest to obtain the ROI tissue region;

[0121] Using an image segmentation algorithm to perform image segmentation on the ROI tissue region, obtain a segmented image of the ROI tissue region and convert it into a corresponding image matrix;

[0122] Extracting pixel data of the position of the ROI mark from the image matrix according to the relative position of the ROI mark to obtain ROI pixel data of the ROI tissue area;

[0123] The age value corresponding to the bone CT image is bound to the ROI pixel data of the ROI tissue region extracted from the bone CT image.

[0124] The above segmentation obtains a bone CT image dataset consisting of bone region mask images. Here, a ROI region is created and segmented for the bone tissue region of interest in each bone region mask image.

[0125] Administrators can manually add corresponding ROI marks to the bone tissue area of ​​interest.

[0126] Image segmentation algorithms can use convolutional neural networks (CNN), deep learning (such as U-Net, FCN, etc.), graph cutting algorithms (such as CRF), region growing algorithms, region competition algorithms, etc.

[0127] The step of performing image segmentation on the ROI tissue region using an image segmentation algorithm may be:

[0128] Preprocessing images: Preprocess the original images, such as denoising, grayscale, binarization, etc., to enhance image quality and improve segmentation effect;

[0129] Select a segmentation algorithm: Choose an appropriate segmentation algorithm based on actual needs, such as threshold-based segmentation, region-based segmentation, edge-based segmentation (edges marked by ROI), graph-cut-based segmentation, etc.

[0130] Initialize the marked area: mark the area to be segmented so that the algorithm can focus on this area in subsequent processing;

[0131] Train the segmentation model: Based on the data of the labeled regions, train the segmentation model so that it can identify and segment the target objects in the image;

[0132] Image segmentation: Use the trained model to segment the original image to obtain the segmented image;

[0133] Post-processing: Perform post-processing on the segmented image, such as removing noise, repairing holes, smoothing edges, etc., to obtain high-quality segmentation results;

[0134] Evaluate segmentation effect: Evaluate the segmentation results, such as calculating segmentation accuracy, recall rate, F1 value and other indicators to determine whether the segmentation effect meets the requirements.

[0135] The steps of separating the pixel data of ROI from the image can be:

[0136] Determine the ROI area: manually mark the region of interest (ROI) that needs to be separated in the image or automatically identify it through an algorithm;

[0137] Read image data: Use image processing libraries (such as OpenCV) to load images;

[0138] Get ROI pixel data: According to the coordinates of ROI, extract the corresponding pixel data from the image matrix;

[0139] Store ROI data: Save the extracted ROI pixel data as a new image file or store it in memory;

[0140] Post-processing: Perform post-processing operations such as filtering, cropping, and scaling on ROI data as needed;

[0141] Output results: Output the processed ROI data or use it for further analysis.

[0142] Using the above segmentation method, the bone ROI area in the bone area mask image can be segmented, and the pixel data of the corresponding ROI tissue area can be extracted. This part of the pixel data can be used to characterize and represent the patient's bone properties and bone density. The bone density of the bones under the current age group can be converted from the ROI pixel data. The pixel values ​​of the ROI tissue area are different under different age groups. Therefore, by extracting the pixel data of the bone tissue ROI area in the bone area mask image, the ROI pixel data of the ROI area in the bone CT image under the corresponding age can be obtained. The pixel data in this area characterizes the bone properties under this age, especially the bone density.

[0143] Subsequently, the pixel values ​​of the bone image area in each mask image are extracted and bound to the corresponding age value, and the pixel data interval and range under different age groups are obtained by statistics. Through the above statistics, the pixel interval or the maximum value of the pixel of the ROI pixel data under different age groups can be obtained to represent the maximum value of bone density. In the histogram, the pixel value corresponding to the histogram vertex under different age groups also represents the maximum bone density value corresponding to the age group ( Figure 4 and Figure 5 The center point of the pixel value of the vertical coordinate in is not 0, but the minimum value).

[0144] Optionally, S3, traversing the HU value of each bone CT image picture in the bone CT image dataset and binding the corresponding age value marked on the picture, including:

[0145] Reading metadata of the bone CT image pictures in the bone CT image dataset in order;

[0146] Parsing the metadata to obtain an adjustment slope k and an adjustment intercept l of the bone CT image;

[0147] Calculate the voxel HU value of the bone CT image:

[0148] HU=C*k+l, C is the pixel value;

[0149] The HU value corresponding to the bone CT image is bound to the age value of the patient corresponding to the bone CT image.

[0150] According to the traversal reading method, the HU value of each bone CT image in each mask image can be obtained and bound to the corresponding age value, so as to obtain the maximum and minimum HU values ​​under different age groups. The larger the HU value, the greater the bone density. As the age group increases, the HU value gradually decreases ( Figure 4 and Figure 5 The center point of the pixel value of the horizontal coordinate in is not 0, but the minimum HU value).

[0151] like Figure 4 As shown, optionally, S4, statistics the ROI pixel data and HU values ​​extracted under different age groups, and construct a ROI-HU database corresponding to the indicators of different age groups, including:

[0152] Reading pixel data in the bone CT image data set, and counting the ROI pixel data bound to different age values ​​in each age group;

[0153] Reading the HU values ​​in the bone CT image data set, and counting the HU values ​​bound to different age values ​​in each age group;

[0154] Fusing the ROI pixel data and the HU value bound to different age values ​​in each age group to generate ROI-HU data under different age group indicators;

[0155] The ROI-HU data under different age group indicators are written into a preset structure database in order to obtain the ROI-HU database.

[0156] Based on the above method, the ROI area pixels and HU values ​​of the CT images of the mask images obtained by segmentation under patients of different ages are extracted, and the ROI pixel data (one segment) and HU values ​​(one segment) under different age group indicators are obtained by statistics. In fact, the ROI pixel segment or HU value segment of each age group indicator is composed of specific values ​​on the images of patients of different ages. Therefore, in terms of age group, after statistics, the following can be obtained Figure 4 The range of values ​​or range data segments represented by the histogram shown.

[0157] By binding the age values, the ROI pixel data and the HU values ​​bound to different age values ​​can be obtained, and the structured data under the two dimensions under these different age group indicators are associated together through the binding relationship, so they are merged into two-dimensional data under one age group indicator as ROI-HU data.

[0158] The steps of writing the structured data in different dimensions (the ROI pixel data and the HU value under each age group indicator) into the structured database may be as follows:

[0159] 1. Determine the database structure: Design the database table structure based on the data dimensions, including field type, length, constraints, etc.

[0160] 2. Create database tables: Create corresponding tables in the database according to the designed table structure.

[0161] 3. Prepare data: Organize the different dimensional data that need to be bound and ensure that the data format matches the database table structure.

[0162] 4. Data import: Use the tools or commands provided by the database to import the prepared data into the corresponding table.

[0163] 5. Data association: If you need to associate data of different dimensions, create an association table or use foreign key constraints to achieve data association.

[0164] 6. Data verification: After importing the data, check the completeness and accuracy of the data to ensure that the data is correct.

[0165] 7. Data indexing: Create indexes for commonly used fields based on query requirements to improve query efficiency.

[0166] 8. Data backup: Back up the database regularly to prevent data loss or damage.

[0167] 9.Data maintenance: Regularly check database performance, optimize query statements, adjust indexes, etc. to ensure stable operation of the database.

[0168] The structured database is prepared by the administrator. Storage space corresponding to the indicators can be prepared in it to store the written data and obtain the ROI-HU database.

[0169] Optionally, S5, extracting feature sets corresponding to indicators of different age groups from the ROI-HU database, and performing statistical regression training on the feature sets to construct a bone prediction model, including:

[0170] Read the ROI-HU database to obtain the ROI pixel data and the HU value corresponding to different age group indicators, and import them into a preset Histogram table to generate the corresponding ROI histogram and HU histogram under each age group indicator;

[0171] Fitting the ROI histogram and HU histogram under each age group index to obtain the HU-Histogram curve under each age group index;

[0172] Extracting the histogram curve features of the HU-histogram curve under each age group indicator and performing ordered statistics to obtain a feature set consisting of the histogram curve features under each age group indicator; the histogram curve features include bone properties under the corresponding age group indicator, and the bone properties include bone density;

[0173] Dividing the feature set into a training set and a validation set according to a preset ratio;

[0174] Importing the training set into a preset statistical regression model, performing learning and training on the histogram curve characteristics under the indicators of each age group, and generating the bone prediction model for predicting the bone properties of bones in different age groups;

[0175] Validating the skeleton prediction model using the validation set:

[0176] If the verification is passed, the skeleton prediction model is deployed and applied;

[0177] Otherwise, repeat S1-S5.

[0178] The present invention defines the bone properties (including bone density values) under the corresponding age group indicators by finding the data features of the two dimensional data displayed by the bone images of the bone patients under the different age group indicators. By fusing the data features of the two dimensional data, the corresponding histogram is generated and the corresponding curve is fitted (the peak values ​​of the histogram under the two dimensions are calculated and connected to form the corresponding change curve (in the form of a broken line, if the peak values ​​on the histograms of different age groups are averaged, the histogram fixed point of each age group can be defined as a point, and the ROI pixel value or HU value corresponding to the point is taken)), and the variation pattern of the patient's bone properties or bone density is described from the curve characteristics.

[0179] Therefore, by reading the data values ​​under two dimensions in the database: ROI pixel value or HU value, and generating histograms under different age groups, and connecting the peak values ​​under different age groups of the histogram, the data variation curves (broken lines) under each dimension are generated to characterize the variation patterns of bone properties and bone density under different age groups. Finally, by fitting the data curves under two dimensions, the following is generated: Figure 5 HU-Histogram curve shown.

[0180] The histogram curve is a fused variation curve that reflects the changing patterns of bone properties, particularly bone density, across different age groups. By extracting curve features from this histogram curve, we can identify the curve characteristics for different age groups. These curve features reflect the changing patterns of bone properties and bone density across age groups, allowing predictions of bone properties for patients of different age groups.

[0181] Subsequently, a training set is obtained by collecting several groups of curve features, and training is performed based on a statistical regression model to obtain a bone prediction model that can predict bone properties, especially bone density, in different age groups. The patient enters his or her age, the model matches the age group, and outputs the curve features of the age group. From the curve features, the maximum bone density reflected by the curve and other curve features can be further analyzed and obtained (curve features). Therefore, the curve features of the current age group, as well as the changing patterns of different bone properties and the bone density of the age group can be viewed from different angles.

[0182] After inputting the patient's current age value, the model can also predict the bone properties of the next age group. The system automatically identifies the age group of the current age, and allows the model to predict and output the bone properties of the next age group. This allows patients to predict and view their bone change patterns in the next stage, allowing patients to know their bone prediction results in advance, which is convenient for bone care and protection.

[0183] The HU-Histogram curve is a graphical representation method used to describe the distribution of edge directions in an image. The steps to extract and count the Histogram features are as follows:

[0184] 1. Perform edge detection on the image. Common methods include Sobel operator, Canny operator, etc.

[0185] 2. Divide the edge image into several small blocks;

[0186] 3. Calculate the gradient direction of the edge points in each small block;

[0187] 4. Divide the gradient direction into several intervals, usually 0 to 180 degrees or 0 to 360 degrees;

[0188] 5. Count the number of edge points in each interval to obtain each column value of the histogram;

[0189] 6. Normalize the histogram to eliminate the influence of image size;

[0190] 7. Calculate the moment features based on the histogram, including mean, variance, skewness, kurtosis, etc., as the feature vector of the image.

[0191] Different curve features such as mean, variance, skewness, kurtosis, etc. represent corresponding bone properties. For example, the mean of the curve corresponding to a certain age group (the average pixel value of the upper and lower endpoints of the curve) represents the average bone density value of the age group; the skewness, that is, the slope of the curve, can represent the osteoporosis change process of the age group, etc. The specific definition can be matched by the administrator in combination with the clinical bone change definition. After the system finally outputs the prediction result, it can directly output the corresponding bone medicine prediction result for the patient.

[0192] Therefore, the present invention can predict the changing patterns of bone properties, especially bone density, in different age groups, and can improve the prediction accuracy by combining big data in two dimensions.

[0193] The present invention utilizes a statistical regression model to learn and train the histogram curve characteristics under various age group indicators to generate the bone prediction model for predicting bone attributes of different age groups. The steps may be:

[0194] Data collection and preprocessing: Collect several sets of histogram curve features, including histogram curve features under age group indicators, and perform preprocessing operations such as data cleaning, missing value processing, and outlier processing.

[0195] Feature engineering: Based on the relationship between learning features and indicators, operations such as feature selection, feature conversion, and feature extraction are performed to improve the predictive ability of the model.

[0196] Divide the dataset: Divide the processed dataset into training set and validation set for model training, tuning, and evaluation.

[0197] Select a statistical regression model: Based on the data characteristics and learning objectives, select an appropriate statistical regression model, such as linear regression, logistic regression, support vector machine, etc.

[0198] Model training: Use the training set to train the selected model and adjust the model parameters so that the model can effectively predict the learning features.

[0199] Model tuning: Tune the model using the validation set, including adjusting model parameters and regularization terms to improve model performance.

[0200] Model evaluation: Use the validation set to evaluate the trained model and calculate the prediction error and related indicators, such as mean square error (MSE) and root mean square error (RMSE).

[0201] Model deployment: Apply the trained model to actual scenarios for predictive analysis.

[0202] Model monitoring and updating: Monitor the model regularly and update or retrain the model when performance deteriorates to ensure the accuracy of prediction results.

[0203] The administrator can set the corresponding training conditions for the verification and optimization training of specific models.

[0204] Figure 6 1 is a block diagram of an application system according to an exemplary embodiment, wherein the application system is used to construct a skeleton prediction model based on the ROI-HU database. Figure 6 The application system includes a backend server 310, a patient-side APP 320, and a medical terminal 330.

[0205] Backend server, on which are deployed:

[0206] An electronic medical record system to record orthopedic patients’ medical history information, including age;

[0207] The bone density prediction module is used to calculate the age group Ta to which the orthopedic patient's current age belongs, call the pre-deployed bone prediction model, predict the bone attributes matching the current age group Ta, and output the corresponding bone density;

[0208] A bone attribute reasoning module is used to record the bone attributes in the next age group Tb predicted and output by the bone density prediction module, and generate a corresponding bone change prediction report;

[0209] Backend database, used for background data storage;

[0210] The patient-side APP is used to log in to the backend server, view the bone density corresponding to the current age group Ta, and view the bone change prediction report;

[0211] The medical terminal is used to log in to the backend server, check the patient's current bone density, and determine whether to issue corresponding bone care instructions;

[0212] The patient-side APP and the medical-side APP are respectively connected to the background server for communication.

[0213] Figure 7 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention, such as Figure 7 As shown, the electronic device may include the above Figure 3 Optionally, the electronic device 410 may include a first processor 2001 .

[0214] Optionally, the electronic device 410 may further include a memory 2002 and a transceiver 2003 .

[0215] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0216] The following combination Figure 7 The components of the electronic device 410 are described in detail.

[0217] The first processor 2001 is the control center of the electronic device 410 and can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement an embodiment of the present invention, such as one or more digital signal processors (DSPs) or one or more field programmable gate arrays (FPGAs).

[0218] Optionally, the first processor 2001 can execute various functions of the electronic device 410 by running or executing software programs stored in the memory 2002 and calling data stored in the memory 2002.

[0219] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 7 CPU0 and CPU1 are shown in FIG.

[0220] In a specific implementation, as an embodiment, the electronic device 410 may also include multiple processors, such as Figure 7 1 and 2. The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). A processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0221] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled by the first processor 2001 for execution. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0222] Alternatively, the memory 2002 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently and be accessed through the interface circuit ( Figure 7 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0223] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0224] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 7 (not shown separately in the figure). The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0225] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 7 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0226] It should be noted that Figure 7 The structure of the electronic device 410 shown in the figure does not constitute a limitation on the router. The actual knowledge structure recognition device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0227] In addition, the technical effects of the electronic device 410 can refer to the technical effects of the method for constructing a skeleton prediction model based on the ROI-HU database described in the above method embodiment, and will not be repeated here.

[0228] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0229] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0230] The above embodiments can be implemented in whole or in part through software, hardware (such as circuits), firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable application system. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired method (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, or magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0231] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0232] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0233] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0234] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0235] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, application systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0236] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, application systems, and methods can be implemented in other ways. For example, the application system embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the application system or unit can be electrical, mechanical or other forms.

[0237] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0238] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0239] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0240] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for constructing a skeleton prediction model based on ROI-HU database, characterized in that: The method comprises: S1. Obtain CT image data of bones of several orthopedic patients at different age groups and perform preprocessing to obtain a bone CT image dataset in a standard format; S2, traversing and segmenting the ROI tissue region of each skeletal CT image picture in the skeletal CT image data set, extracting ROI pixel data of the ROI tissue region and binding it to the age value marked on the corresponding picture; S3, traversing the HU value of each bone CT image picture in the bone CT image data set, and binding the age value marked on the corresponding picture; S4, counting the ROI pixel data and HU values ​​extracted under different age groups, and constructing a ROI-HU database corresponding to the indicators of different age groups; S5, extracting feature sets corresponding to indicators of different age groups from the ROI-HU database, and performing statistical regression training on the feature sets to construct a bone prediction model; The step of extracting feature sets corresponding to indicators of different age groups from the ROI-HU database and performing statistical regression training on the feature sets to construct a bone prediction model includes: Read the ROI-HU database to obtain the ROI pixel data and the HU value corresponding to different age group indicators, and import them into a preset Histogram table to generate the corresponding ROI histogram and HU histogram under each age group indicator; Fitting the ROI histogram and HU histogram under each age group index to obtain the HU-Histogram curve under each age group index; Extracting the histogram curve features of the HU-histogram curve under each age group indicator and performing ordered statistics to obtain a feature set consisting of the histogram curve features under each age group indicator; the histogram curve features include bone properties under the corresponding age group indicator, and the bone properties include bone density; Dividing the feature set into a training set and a validation set according to a preset ratio; Importing the training set into a preset statistical regression model, performing learning and training on the histogram curve characteristics under the indicators of each age group, and generating a bone prediction model for predicting the bone properties of bones in different age groups; Validating the skeleton prediction model using the validation set: If the verification is passed, the skeleton prediction model is deployed and applied; Otherwise, repeat S1-S5.

2. The method for constructing a skeleton prediction model based on the ROI-HU database according to claim 1, wherein S1. Obtain and preprocess skeletal CT image data of several orthopedic patients at different ages to obtain a standard format skeletal CT image dataset, including: Log in to the backend database and filter the CT image data of bones of several orthopedic patients in different age groups; Cleaning the screened CT image data and sorting them according to age values; Based on the Canny algorithm, the bone region edge in the CT image data is identified, a vector mask is added along the bone region edge, and the bone region portion where the vector mask is located is segmented, and a bone region mask image in a BMP format is outputted and obtained; Performing pixel value calibration on the skeleton region mask image and the original CT image data, and calibrating the pixel value of the corresponding skeleton region mask image based on the resolution of the CT image data; After calibration, an image set consisting of masked images of bone regions of different age groups is saved to generate the bone CT image dataset.

3. The method for constructing a skeleton prediction model based on the ROI-HU database according to claim 1, wherein: S2, traversing and segmenting the ROI tissue region of each skeletal CT image picture in the skeletal CT image data set, extracting ROI pixel data of the ROI tissue region and binding it to the age value marked on the corresponding picture, including: Traversing the bone CT image pictures in the bone CT image dataset in order; In the bone CT image, a corresponding ROI mark is added to the bone tissue region of interest to obtain the ROI tissue region; Using an image segmentation algorithm to perform image segmentation on the ROI tissue region, obtain a segmented image of the ROI tissue region and convert it into a corresponding image matrix; Extracting pixel data of the position of the ROI mark from the image matrix according to the relative position of the ROI mark to obtain ROI pixel data of the ROI tissue area; The age value corresponding to the bone CT image is bound to the ROI pixel data of the ROI tissue region extracted from the bone CT image.

4. The method for constructing a skeleton prediction model based on the ROI-HU database according to claim 1, characterized in that: S3. Traverse the HU value of each bone CT image in the bone CT image dataset and bind the corresponding age value marked on the image, including: Reading metadata of the bone CT image pictures in the bone CT image dataset in order; Parsing the metadata to obtain an adjustment slope k and an adjustment intercept l of the bone CT image; Calculate the voxel HU value of the bone CT image: HU=C*k+l, C is the pixel value; The HU value corresponding to the bone CT image is bound to the age value of the patient corresponding to the bone CT image.

5. The method for constructing a skeleton prediction model based on the ROI-HU database according to claim 1, characterized in that: S4. Count the ROI pixel data and HU values ​​extracted under different age groups, and construct an ROI-HU database corresponding to the indicators of different age groups, including: Reading pixel data in the bone CT image data set, and counting the ROI pixel data bound to different age values ​​in each age group; Reading the HU values ​​in the bone CT image data set, and counting the HU values ​​bound to different age values ​​in each age group; Fusing the ROI pixel data and the HU value bound to different age values ​​in each age group to generate ROI-HU data under different age group indicators; The ROI-HU data under different age group indicators are written into a preset structure database in order to obtain the ROI-HU database.

6. An application system for implementing the method for constructing a skeleton prediction model based on the ROI-HU database according to any one of claims 1 to 5, characterized in that: The application system includes: Backend server, on which are deployed: An electronic medical record system to record orthopedic patients’ medical history information, including age; The bone density prediction module is used to calculate the age group Ta to which the orthopedic patient's current age belongs, call the pre-deployed bone prediction model, predict the bone attributes matching the current age group Ta, and output the corresponding bone density; A bone attribute reasoning module is used to record the bone attributes in the next age group Tb predicted and output by the bone density prediction module, and generate a corresponding bone change prediction report; Backend database, used for background data storage; The patient-side APP is used to log in to the backend server, view the bone density corresponding to the current age group Ta, and view the bone change prediction report; The medical terminal is used to log in to the backend server, check the patient's current bone density, and determine whether to issue corresponding bone care instructions; The patient-side APP and the medical-side APP are respectively connected to the background server for communication.

7. An electronic device, characterized in that: The electronic device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program code, which can be called by a processor to execute the method according to any one of claims 1 to 5.

Images