A high-precision detection method and system for human bone density value

Abstract

The present disclosure provides a high-precision detection method and system for human bone density value, and relates to the technical field of bone density detection. The method comprises the following steps: image acquisition of a target user is performed by a medical image acquisition device to obtain user image information, wherein the user image information comprises user medical image, acquisition device parameters and user basic information; a bone density database is constructed; a bone density feature mapping library is constructed; image preprocessing is performed; preprocessed images are identified and segmented to determine multiple image regions; matching analysis is performed on the multiple image regions and the bone density feature mapping library to obtain bone density analysis results; and user bone density determination information is obtained by inputting the user basic information and the bone density analysis results into a user evaluation model. The technical problem of insufficient bone density detection accuracy caused by insufficient analysis of bone density characteristics in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of bone density detection, in particular to a high-precision detection method and system for human bone density value. BACKGROUND

[0002] Bone density detection is to measure the bone density value of the human body. Through the detection of bone density, osteoporosis and other diseases can be found in time and effectively treated, which is of great significance for the prevention and treatment of skeletal diseases.

[0003] At present, there is a technical problem of insufficient bone density detection accuracy in the prior art due to insufficient analysis of bone density characteristics. SUMMARY

[0004] The present disclosure provides a high-precision detection method and system for human bone density value to solve the technical problem of insufficient bone density detection accuracy in the prior art due to insufficient analysis of bone density characteristics.

[0005] According to a first aspect of the present disclosure, a high-precision detection method for human bone density value is provided, comprising: acquiring image information of a target user by a medical image acquisition device to obtain user image information, wherein the user image information includes user medical image, acquisition device parameters, and user basic information; constructing a bone density database, wherein the database includes multiple groups of medical images and corresponding bone density values; performing data clustering on the bone density database to construct a bone density feature mapping library; extracting features from the user medical image to obtain image feature information, and performing image preprocessing based on the image feature information and the acquisition device parameters; identifying and segmenting the preprocessed image to determine multiple image regions, and performing matching analysis on the multiple image regions and the bone density feature mapping library to obtain bone density analysis results; inputting the user basic information and the bone density analysis results into a user evaluation model to obtain user bone density determination information.

[0006] According to a second aspect of this disclosure, a high-precision detection system for human bone mineral density (BMD) is provided, comprising: a user image information acquisition module, which acquires images of a target user using a medical imaging acquisition device to obtain user image information, wherein the user image information includes user medical images, acquisition device parameters, and user basic information; a bone mineral density database construction module, which constructs a bone mineral density database, including multiple sets of medical images and corresponding bone mineral density values; a data clustering module, which performs data clustering on the bone mineral density database to construct a bone mineral density feature mapping library; an image preprocessing module, which extracts features from the user medical images to obtain image feature information, and performs image preprocessing based on the image feature information and the acquisition device parameters; a bone mineral density analysis module, which identifies and segments the preprocessed images to determine multiple image regions, performs matching analysis between the multiple image regions and the bone mineral density feature mapping library, and obtains bone mineral density analysis results; and a bone mineral density determination module, which inputs the user basic information and the bone mineral density analysis results into a user assessment model to obtain user bone mineral density determination information.

[0007] According to a third aspect of this disclosure, an electronic device is provided, comprising:

[0008] At least one processor; and

[0009] A memory communicatively connected to the at least one processor; wherein,

[0010] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method described in the first aspect.

[0011] According to the high-precision detection method for human bone mineral density (BMD) adopted in this disclosure, images of a target user are acquired using a medical imaging acquisition device to obtain user image information, which includes user medical images, acquisition device parameters, and user basic information. A bone mineral density database is constructed, which includes multiple sets of medical images and corresponding bone mineral density values. Data clustering is performed on the bone mineral density database to construct a bone mineral density feature mapping library. Feature extraction is performed on the user medical images to obtain image feature information. Image preprocessing is performed based on the image feature information and the acquisition device parameters. The preprocessed image is identified and segmented to determine multiple image regions. The multiple image regions are matched and analyzed with the bone mineral density feature mapping library to obtain bone mineral density analysis results. The user basic information and the bone mineral density analysis results are input into a user assessment model to obtain user bone mineral density determination information. This disclosure involves acquiring user medical images, preprocessing the images, constructing a bone density database containing multiple sets of standard images and corresponding bone density values, clustering the data in the bone density database, constructing a bone density feature mapping library, further identifying and segmenting the preprocessed images to determine multiple image regions, then performing region matching to determine the corresponding bone density values, identifying anomalies, and issuing anomaly warnings. This achieves the technical effect of improving the accuracy of bone density detection, providing timely bone density warnings to users, and assisting users in maintaining bone health.

[0012] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0014] Figure 1 A flowchart illustrating a high-precision method for detecting human bone mineral density values ​​provided in this embodiment of the disclosure;

[0015] Figure 2 This is a schematic diagram of the structure of a high-precision detection system for human bone mineral density provided in an embodiment of the present disclosure;

[0016] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure.

[0017] Figure labeling: User image information acquisition module 11, bone density database construction module 12, data clustering module 13, image preprocessing module 14, bone density analysis module 15, bone density determination module 16, electronic device 800, processor 801, memory 802, bus 803. Detailed Implementation

[0018] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0019] In order to solve the technical problem that the accuracy of bone density detection is insufficient due to the lack of detailed analysis of bone density characteristics in the prior art, the inventors of this disclosure have creatively obtained a high-precision detection method and system for human bone density values.

[0020] Example 1

[0021] Figure 1 A diagram illustrating a high-precision method for detecting human bone mineral density values ​​provided in this application embodiment is shown below. Figure 1 As shown, the method includes:

[0022] Step S100: Acquire images of the target user using a medical imaging acquisition device to obtain user image information, wherein the user image information includes user medical images, acquisition device parameters, and user basic information;

[0023] Specifically, the aforementioned medical imaging acquisition equipment refers to CT plain scan equipment. Its principle involves using precisely collimated X-ray beams, gamma rays, ultrasound, etc., along with highly sensitive detectors, to perform a series of cross-sectional scans around a specific part of the human body. The target user refers to any user undergoing bone density testing. The medical imaging acquisition equipment acquires images of the target user, obtaining user image information. This user image information includes the user's medical images, acquisition equipment parameters, and basic user information. The user's medical images refer to the target user's CT images; the acquisition equipment parameters refer to the equipment parameters of the CT plain scan equipment used for CT image acquisition, including equipment model and scanning parameters; and the basic user information includes the target user's name, age, and other basic information.

[0024] Step S200: Construct a bone mineral density database, which includes multiple sets of medical images and their corresponding bone mineral density values;

[0025] Specifically, based on big data, multiple sets of medical images and their corresponding bone mineral density values ​​are obtained, with each set of medical images corresponding to a set of bone mineral density values. Based on this, a bone mineral density database is constructed, which includes multiple sets of medical images and their corresponding bone mineral density values.

[0026] Step S300: Perform data clustering on the bone density database to construct a bone density feature mapping library;

[0027] In this embodiment, step S300 further includes:

[0028] Step S310: Extract features from each group of medical images in the bone density database to obtain bone features, and locate and label the bone features;

[0029] Step S320: Analyze the continuity and range of feature changes in each medical image based on the skeletal features to obtain skeletal feature index information, and annotate the skeletal feature index information;

[0030] Step S330: Determine the bone mineral density grade coefficient based on the multiple sets of medical images and the corresponding bone mineral density values;

[0031] Step S340: Based on the mapping relationship between each medical image and the bone mineral density grade coefficient, the skeletal features, and the skeletal feature index information, establish an image feature library;

[0032] Step S350: Perform feature clustering on the image feature library based on the skeletal features, and construct the bone density feature mapping library based on the clustering results and the corresponding mapping relationship between the skeletal features, bone density grade coefficients, and skeletal feature index information.

[0033] Specifically, data clustering of the bone density database involves grouping bone density data with the same characteristics together. Different bone features correspond to different bone density values. Based on this, a bone density feature mapping library is constructed, which facilitates subsequent analysis of the target user's bone density.

[0034] Specifically, feature extraction is performed on each group of medical images in the bone density database to obtain skeletal features. Skeletal features refer to the changes in bone density corresponding to a segment of bone. Further, based on the skeletal features, the continuity and range of feature changes in each medical image are analyzed. Simply put, bone tissue has a high calcium content, and medical images appear white. Under normal circumstances, the bone density of a segment of bone in the same location is basically the same. However, when a lesion occurs, the bone density may be different. The continuity of change refers to the changes in bone density of a segment of bone. For example, a segment of bone may be normal at one end, but the bone density decreases significantly in the middle. The areas of continuous change are clustered together for analysis. The range of change refers to the interval of continuous change in bone density. The characteristic values ​​of the continuity of change and the range of change are used as skeletal feature indicators, and different skeletal feature indicators are labeled. Based on multiple groups of medical images and their corresponding bone density values, a bone density grade coefficient is determined. The higher the bone density, the higher the corresponding bone density grade coefficient. Each medical image has a one-to-one correspondence with bone mineral density grade coefficient, skeletal features, and skeletal feature index information. Based on the correspondence, an image feature library is established. Based on the skeletal features, the image feature library is clustered. This means that medical images in the image feature library with the same continuity and range of bone mineral density changes are grouped into one category. Based on the clustering results and the corresponding mapping relationship of skeletal features, bone mineral density grade coefficient, and skeletal feature index information, a bone mineral density feature mapping library is constructed.

[0035] Step S400: Extract features from the user's medical images to obtain image feature information, and perform image preprocessing based on the image feature information and the acquisition device parameters;

[0036] In this embodiment, step S400 further includes:

[0037] Step S410: Input the user's medical image into the image feature extraction model to obtain the image feature information;

[0038] Step S420: Based on the image feature information, perform core feature analysis to determine core feature elements;

[0039] Step S430: Locate the user's medical image based on the core feature elements and annotate the feature regions;

[0040] Step S440: Perform local enhancement processing on the feature region to obtain the preprocessed image.

[0041] In this embodiment, step S440 further includes:

[0042] Step S441: Perform standard operations on the feature region according to the preset image display standard to obtain a standardized display of the region;

[0043] Step S442: Determine whether the standardized display of the area meets the preset clarity recognition requirements;

[0044] Step S443: When the condition is not met, the original medical image of the marked feature region is enhanced using a preset enhancement algorithm to obtain the preprocessed image.

[0045] In this embodiment, step S443 further includes:

[0046] Step S4431: Filter and denoise the original medical image using a preset filtering and denoising algorithm;

[0047] Step S4432: Use the Laplacian pyramid algorithm to decompose the original medical image after filtering and denoising to obtain the energy layer and detail layer;

[0048] Step S4433: Enhance the detail layer, and then obtain the preprocessed image through reverse iteration.

[0049] Specifically, different bone densities will result in different images on medical images. For example, medical images may have gaps or edge features. Based on this, feature extraction is performed on the user's medical images to obtain image feature information. Then, image preprocessing is performed based on the image feature information and the parameters of the acquisition device. Simply put, this is to perform local enhancement processing on the feature areas.

[0050] Specifically, user medical images are input into an image feature extraction model to obtain image feature information. This model is a neural network model in machine learning, which analyzes user medical images to identify changes in the images. Image feature information refers to the changes in the images. Based on this information, core feature analysis is performed. Different image features require different feature requirements for identifying bone density. For example, if there are gaps in the image, the size and distribution of these gaps need to be determined to identify bone density; these gap sizes and distribution characteristics are the core feature elements. If bone wear has occurred, the edge features of the medical image are the core feature elements. Locating the user's medical image based on these core feature elements involves identifying the location regions containing these core feature elements and marking them as feature regions. These feature regions are then subjected to local enhancement processing to obtain a preprocessed image.

[0051] Specifically, according to the preset image display standards, which are standard rules for displaying medical images, such as the level of detail in the medical images and the magnification or reduction ratio, the feature regions are subjected to standard operations to display the medical images, thereby obtaining standardized regional displays. Then, it is determined whether the standardized regional displays meet the preset sharpness recognition requirements, which are the expected sharpness of the image display, such as the level at which the details of the skeletal images can be clearly identified. If this is not achieved, the original medical images of the marked feature regions need to be enhanced using preset enhancement algorithms to obtain preprocessed images.

[0052] Specifically, the process of enhancing the original medical image with labeled feature regions using a preset enhancement algorithm is as follows: The original medical image is filtered and denoised using a preset filtering and denoising algorithm, which refers to the Guided Filter algorithm. First, a Guided Filter image pyramid is constructed, using the original medical image as the first layer of the pyramid. Then, the first layer of the pyramid is filtered using the Guided Filter algorithm, and even-numbered rows and columns are removed (downsampling), resulting in the second layer of the Guided Filter pyramid image, with dimensions of 640*480. The third and fourth layers are then obtained using the same method, with sizes of 320*240 and 160*120 respectively. After these steps of constructing the Guided Filter image pyramid, the final Guided Filter pyramid image, as shown in the figure, is obtained. The Laplacian pyramid algorithm is further used to decompose the filtered and denoised original medical image to obtain the energy layer and detail layer. Specifically, the second layer of the Guided Filter pyramid is upsampled and then Gaussian filtered to obtain an image with a dimension of 1280*960. Then, the first layer of the Guided Filter pyramid is subtracted to obtain the first layer of the Laplacian pyramid, which also has a dimension of 1280*960. The third layer of the Guided Filter pyramid is then upsampled and then Gaussian filtered to obtain an image with a dimension of 640*480. The second layer of the Guided Filter pyramid is then subtracted to obtain the second layer of the Laplacian pyramid, which also has a dimension of 640*480. The fourth layer of the Guided Filter pyramid is then upsampled and then Gaussian filtered to obtain an image with a dimension of 320*240. The third layer of the Guided Filter pyramid is then subtracted to obtain the third layer of the Laplacian pyramid, which also has a dimension of 320*240. After the above steps of constructing the Laplacian image pyramid, a Laplacian pyramid image is obtained. This image contains both detail and noise information from the original medical image, namely the detail layer and the energy layer. The detail layer is enhanced, and then through reverse iteration, the preprocessed image is obtained. This involves performing bilateral filtering on the Laplacian image pyramid, and then superimposing the filtered pyramid image back onto the energy layer of the original image to obtain the enhanced detail image, i.e., the preprocessed image. This achieves the technical effect of enhancing medical image details, reducing noise, facilitating bone density analysis, and improving the accuracy of bone density analysis.

[0053] Step S500: The preprocessed image is identified and segmented to determine multiple image regions. The multiple image regions are then matched and analyzed with the bone density feature mapping library to obtain bone density analysis results.

[0054] In this embodiment, step S500 further includes:

[0055] Step S510: Perform continuous region feature change analysis based on the multi-image regions to determine continuous change features and change span features;

[0056] Step S520: Match and analyze the multiple image regions with the bone density feature mapping library to obtain the multi-region bone density analysis results;

[0057] Step S530: Using the continuous change feature and the change span feature, perform matching analysis with the bone density feature mapping library respectively to obtain the feature index bone density analysis results;

[0058] Step S540: Perform a fusion analysis based on the multi-region bone mineral density analysis results and the characteristic index bone mineral density analysis results to obtain the bone mineral density analysis results.

[0059] In this embodiment, step S540 further includes:

[0060] Step S541: Based on the multi-region bone mineral density analysis results and the characteristic index bone mineral density analysis results, perform overlap feature analysis to determine the overlap analysis features;

[0061] Step S542: Perform bone mineral density influence analysis using the aforementioned overlap analysis features to determine the influence adjustment coefficient;

[0062] Step S543: Use the influence adjustment coefficient to adjust the multi-region bone mineral density analysis results and feature index bone mineral density analysis results corresponding to the overlapping analysis features to obtain the corrected bone mineral density analysis results;

[0063] Step S544: Based on the multi-region bone mineral density analysis results, the characteristic index bone mineral density analysis results, and the modified bone mineral density analysis results, perform a fully connected bone region analysis to determine the fully connected bone mineral density analysis results;

[0064] Step S545: Perform differential analysis based on the fully connected bone mineral density analysis results, determine the differential analysis results, and obtain the bone mineral density analysis results based on the differential analysis results.

[0065] Specifically, the preprocessed image is segmented and identified to determine multiple image regions. In simple terms, different features of bone density are divided and segmented for analysis. Then, the multiple image regions are matched with a bone density feature mapping library. Image features that match the multiple image regions are found from the bone density feature mapping library. Based on the mapping relationship between the data in the bone density feature mapping library, the bone density analysis results are obtained. The bone density analysis results include bone density features of different regions.

[0066] Specifically, based on the analysis of continuous regional feature changes in multiple imaging regions, the continuous change characteristics and change span characteristics are determined. This involves dividing and segmenting the different change characteristics of bone density. Continuous change characteristics refer to the situation where bone density changes continuously in a segment of bone. For example, gaps appear continuously in multiple imaging regions, and the distribution of the gaps is also changing. Change span refers to the range of change intervals. Multiple image regions are matched and analyzed against a bone density feature mapping library to obtain multi-region bone density analysis results. Continuous change features and change span features are used to match and analyze against the bone density feature mapping library to obtain corresponding feature index bone density analysis results. Then, the multi-region bone density analysis results and feature index bone density analysis results are fused to obtain the final bone density analysis result. This method identifies abnormalities in bone density in different regions based on their bone density change characteristics, helping users understand their bone density status in a timely manner. It can also consider the change characteristics of each segment and analyze them in conjunction with the user's specific situation. For example, if the bone density of a region is normal, but the individual's age and exercise status result in a higher bone standard, there is a difference compared to other connected regions. In this case, the bone density of this region is lower than that of other regions. Although the bone density is within the normal range, extra attention is needed.

[0067] Specifically, based on the multi-region bone density analysis results and the characteristic index bone density analysis results, overlap feature analysis is performed to determine the overlap analysis features. This means that if a certain region is analyzed repeatedly during the bone density analysis process, there will be duplicate regions. Overlap analysis features refer to the bone density features corresponding to two or more overlapping regions. It is necessary to determine which region's analysis result should be used to determine the user's bone density status. It is necessary to use the overlap analysis features to perform bone density impact analysis to analyze the degree of influence of the overlap analysis features on the bone density results, thereby determining the influence adjustment coefficient. The greater the degree of influence, the larger the influence adjustment coefficient. The influence adjustment coefficient is used to perform a comprehensive analysis of the overlapping regions. For example, the weight ratio of the overlapping regions is set according to the influence adjustment coefficient. The multi-region bone density analysis results and characteristic index bone density analysis results corresponding to the overlap analysis features are adjusted according to the weight ratio to obtain more accurate corrected bone density analysis results. Further, based on the results of multi-regional bone mineral density analysis, characteristic index bone mineral density analysis, and corrected bone mineral density analysis, a fully connected component analysis of the skeletal region is performed. This fully connected component analysis examines the integrity of the bone, ensuring no areas are missed, unanalyzed, or over-analyzed. This ensures the determination of the fully connected component analysis results, which include the bone mineral density analysis results for each region. Differential analysis is then performed based on these results to determine the differential analysis outcome. Based on this differential analysis, the final bone mineral density analysis result is obtained. The complete bone is analyzed to determine if any differentials exist. If there are no differences or the differences are very small, the fully connected component analysis result is output as a single result. If the differences are significant, a tiered output is performed. Based on the bone mineral density differences, the bone mineral density analysis results for each region are divided into multiple different levels from low to high, and the bone mineral density analysis results for each level and their corresponding regions are output in a tiered manner.

[0068] Step S600: Input the user's basic information and the bone density analysis results into the user assessment model to obtain the user's bone density determination information.

[0069] Specifically, user basic information and bone density analysis results are input into the user assessment model to obtain user bone density determination information. The user assessment model determines whether a user's bone density is within the normal range, slightly below normal, or higher than normal. Normal or higher values ​​indicate healthy bones, while values ​​below normal suggest poor bone health, prompting timely warnings so users can take appropriate measures, such as calcium supplementation, to maintain bone health. The user assessment model is a neural network model in machine learning. Its input data consists of user basic information and bone density analysis results, while its output data is the user bone density determination information. Specifically, the user assessment model can be trained and tested using a sample training dataset. This dataset includes multiple sets of sample user basic information, sample bone density analysis results, and corresponding sample user bone density determination information. The user assessment model is trained until convergence using this dataset, and its accuracy is tested to obtain a user assessment model that meets the expected requirements.

[0070] Based on the above analysis, this disclosure provides a high-precision method for detecting human bone mineral density (BMD). In this embodiment, by acquiring user medical images, preprocessing the user images, and then constructing a bone mineral density database containing multiple sets of standard images and corresponding BMD values, the BMD database is clustered to construct a bone mineral density feature mapping library. The preprocessed images are further identified and segmented to determine multiple image regions, and then region matching is performed to determine the corresponding BMD values. Anomalies are identified and anomaly warnings are issued, thereby improving the accuracy of bone mineral density detection, providing timely bone mineral density warnings to users, and assisting users in ensuring bone health.

[0071] Example 2

[0072] Based on the same inventive concept as the high-precision detection method for human bone mineral density values ​​described in the foregoing embodiments, such as Figure 2 As shown, this application also provides a high-precision detection system for human bone mineral density values, the system comprising:

[0073] User image information acquisition module 11 is used to acquire images of a target user through a medical image acquisition device to obtain user image information, wherein the user image information includes user medical images, acquisition device parameters, and user basic information.

[0074] Bone density database construction module 12, which is used to construct a bone density database, including multiple sets of medical images and corresponding bone density values;

[0075] Data clustering module 13 is used to perform data clustering on the bone density database and construct a bone density feature mapping library;

[0076] Image preprocessing module 14 is used to extract features from the user's medical image to obtain image feature information, and to perform image preprocessing based on the image feature information and the acquisition device parameters.

[0077] Bone density analysis module 15 is used to identify and segment the preprocessed image, determine multiple image regions, and perform matching analysis between the multiple image regions and the bone density feature mapping library to obtain bone density analysis results.

[0078] Bone density determination module 16 is used to obtain user bone density determination information by inputting the user basic information and the bone density analysis results into the user assessment model.

[0079] Furthermore, the system also includes:

[0080] A skeletal feature extraction module is used to extract features from each group of medical images in the bone density database to obtain skeletal features, and to locate and label the skeletal features.

[0081] The feature change analysis module is used to perform feature change continuity and change span analysis on each medical image based on the skeletal features, obtain skeletal feature index information, and annotate the skeletal feature index information.

[0082] A bone mineral density grade coefficient determination module is used to determine the bone mineral density grade coefficient based on the multiple sets of medical images and the corresponding bone mineral density values.

[0083] An image feature library establishment module is used to establish an image feature library based on the mapping relationship between each medical image and the bone density grade coefficient, the skeletal features, and the skeletal feature index information.

[0084] A bone density feature mapping library construction module is used to perform feature clustering on an image feature library based on the skeletal features, and construct the bone density feature mapping library based on the clustering results and the corresponding mapping relationship between the skeletal features, bone density grade coefficients, and skeletal feature index information.

[0085] Furthermore, the system also includes:

[0086] An image feature information acquisition module is used to input the user's medical image into an image feature extraction model to obtain the image feature information.

[0087] The core feature analysis module is used to perform core feature analysis based on the image feature information to determine core feature elements.

[0088] The feature region annotation module is used to locate the user's medical image based on the core feature elements and annotate the feature regions.

[0089] A local enhancement processing module is used to perform local enhancement processing on the feature region to obtain the preprocessed image.

[0090] Furthermore, the system also includes:

[0091] A region-standardized display acquisition module is used to perform standard operations on the feature region according to a preset image display standard to obtain a region-standardized display.

[0092] A region standardization display judgment module is used to determine whether the region standardization display meets the preset clarity recognition requirements.

[0093] A preprocessed image acquisition module is used to enhance the original medical image with marked feature regions using a preset enhancement algorithm when the feature region is not reached, thereby obtaining the preprocessed image.

[0094] Furthermore, the system also includes:

[0095] A filtering and denoising module, wherein the filtering and denoising module is used to filter and denoise the original medical image using a preset filtering and denoising algorithm;

[0096] The original medical image decomposition module is used to decompose the filtered and denoised original medical image using the Laplacian pyramid algorithm to obtain the energy layer and detail layer.

[0097] A reverse iteration module is used to enhance the detail layer and then obtain the preprocessed image through reverse iteration.

[0098] Furthermore, the system also includes:

[0099] A continuous region feature analysis module is used to perform continuous region feature change analysis based on the multi-image region to determine continuous change features and change span features.

[0100] A multi-region bone mineral density analysis result acquisition module is used to match and analyze the multiple image regions with the bone mineral density feature mapping library to obtain multi-region bone mineral density analysis results.

[0101] The feature index bone matching module is used to perform matching analysis with the bone density feature mapping library using the continuous change feature and the change span feature, respectively, to obtain the feature index bone density analysis result.

[0102] The fusion analysis module is used to perform fusion analysis based on the multi-region bone density analysis results and the characteristic index bone density analysis results to obtain the bone density analysis results.

[0103] Furthermore, the system also includes:

[0104] The overlapping feature analysis module is used to perform overlapping feature analysis based on the multi-region bone density analysis results and the feature index bone density analysis results to determine the overlapping analysis features.

[0105] A bone mineral density impact analysis module is used to perform bone mineral density impact analysis using the overlap analysis features to determine the impact adjustment coefficient;

[0106] The analysis result adjustment module is used to adjust the multi-region bone density analysis results and feature index bone density analysis results corresponding to the overlapping analysis features using the influence adjustment coefficient, so as to obtain the corrected bone density analysis results;

[0107] A fully connected analysis module is used to perform a fully connected analysis of the skeletal region based on the multi-regional bone mineral density analysis results, the feature index bone mineral density analysis results, and the modified bone mineral density analysis results, and to determine the fully connected bone mineral density analysis results.

[0108] A differential analysis module is used to perform differential analysis based on the fully connected bone mineral density analysis results, determine the differential analysis results, and obtain the bone mineral density analysis results based on the differential analysis results.

[0109] The specific example of the high-precision detection method for human bone mineral density values ​​in Embodiment 1 described above is also applicable to the high-precision detection system for human bone mineral density values ​​in this embodiment. Through the foregoing detailed description of the high-precision detection method for human bone mineral density values, those skilled in the art can clearly understand the high-precision detection system for human bone mineral density values ​​in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here. As for the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant details can be found in the method section.

[0110] Example 3

[0111] Figure 3 This is a schematic diagram based on the third embodiment of the present disclosure, as shown below. Figure 3 As shown, the electronic device 800 in this disclosure may include a processor 801 and a memory 802.

[0112] Memory 802 is used to store programs. Memory 802 may include volatile memory, such as random-access memory (RAM), such as static random-access memory (SRAM), double data rate synchronous dynamic random-access memory (DDR SDRAM), etc.; memory may also include non-volatile memory, such as flash memory. Memory 802 is used to store computer programs (such as application programs, functional modules, etc. that implement the above methods), computer instructions, etc. The computer programs, computer instructions, etc., can be partitioned and stored in one or more memories 802. Furthermore, the computer programs, computer instructions, data, etc., can be accessed by processor 801.

[0113] The aforementioned computer programs and instructions can be stored in one or more partitions of memory 802. Furthermore, the aforementioned computer programs and instructions can be invoked by processor 801.

[0114] The processor 801 is configured to execute the computer program stored in the memory 802 to implement the various steps in the methods described in the above embodiments.

[0115] For details, please refer to the relevant descriptions in the preceding method embodiments.

[0116] The processor 801 and the memory 802 can be independent structures or integrated structures. When the processor 801 and the memory 802 are independent structures, the memory 802 and the processor 801 can be coupled together via bus 803.

[0117] The electronic device in this embodiment can execute the technical solution in the above method. Its specific implementation process and technical principle are the same, and will not be repeated here.

[0118] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0119] According to embodiments of this disclosure, this disclosure also provides a computer program product comprising: a computer program stored in a readable storage medium, at least one processor of an electronic device being able to read the computer program from the readable storage medium, and the at least one processor executing the computer program causing the electronic device to perform the scheme provided in any of the above embodiments.

[0120] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders.

[0121] This document does not impose any restrictions as long as the desired results of the disclosed technical solution can be achieved.

[0122] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A high-precision method for detecting human bone mineral density, characterized in that, The method includes: The target user's image information is obtained by acquiring images of the target user through medical imaging acquisition equipment. The user image information includes the user's medical images, acquisition equipment parameters, and user basic information. Construct a bone mineral density database, which includes multiple sets of medical images and their corresponding bone mineral density values; Data clustering is performed on the bone density database to construct a bone density feature mapping library; Feature extraction is performed on the user's medical images to obtain image feature information, and image preprocessing is performed based on the image feature information and the parameters of the acquisition device. The preprocessed image is identified and segmented to determine multiple image regions. The multiple image regions are then matched and analyzed with the bone density feature mapping library to obtain bone density analysis results. Based on the user's basic information and the bone density analysis results, the user assessment model is input to obtain the user's bone density determination information. Based on the identification and segmentation of the preprocessed image, multiple image regions are determined. These regions are then matched and analyzed with the bone density feature mapping library to obtain bone density analysis results, including: Based on the analysis of continuous regional feature changes in the multiple image regions, the continuous change features and the change span features are determined. The multiple image regions are matched and analyzed with the bone density feature mapping library to obtain the bone density analysis results of multiple regions; By using the continuous change feature and the change span feature, respectively, matching analysis is performed with the bone density feature mapping library to obtain the feature index bone density analysis results; The bone density analysis results are obtained by performing a fusion analysis based on the multi-region bone density analysis results and the characteristic index bone density analysis results. The bone mineral density analysis results are obtained by fusing the multi-region bone mineral density analysis results and the characteristic index bone mineral density analysis results, including: Based on the multi-region bone density analysis results and the characteristic index bone density analysis results, an overlap feature analysis is performed to determine the overlap analysis features; The influence of the aforementioned overlap analysis features was used to perform bone mineral density impact analysis and determine the influence adjustment coefficient; The influence adjustment coefficient is used to adjust the multi-region bone mineral density analysis results and feature index bone mineral density analysis results corresponding to the overlapping analysis features to obtain the corrected bone mineral density analysis results; Based on the multi-regional bone mineral density analysis results, the characteristic index bone mineral density analysis results, and the modified bone mineral density analysis results, a full connectivity analysis of the skeletal region is performed to determine the full connectivity bone mineral density analysis results. Differential analysis is performed based on the results of the fully connected bone mineral density analysis to determine the differential analysis results, and the bone mineral density analysis results are obtained based on the differential analysis results.

2. The method as described in claim 1, characterized in that, The bone density database is subjected to data clustering to construct a bone density feature mapping library, including: Feature extraction is performed on each group of medical images in the bone density database to obtain skeletal features, and the skeletal features are then located and labeled. Based on the skeletal features, the continuity and range of feature changes in each medical image are analyzed to obtain skeletal feature index information, and the skeletal feature index information is labeled. Based on the multiple sets of medical images and their corresponding bone mineral density values, a bone mineral density grade coefficient is determined. Based on the mapping relationship between each medical image and the bone mineral density grade coefficient, the skeletal features, and the skeletal feature index information, an image feature library is established. Based on the skeletal features, the image feature library is clustered. Based on the clustering results and the corresponding mapping relationship between the skeletal features, bone density grade coefficients, and skeletal feature index information, the bone density feature mapping library is constructed.

3. The method as described in claim 1, characterized in that, Feature extraction is performed on the user's medical images to obtain image feature information. Image preprocessing is then performed based on the image feature information and the parameters of the acquisition device, including: The user's medical images are input into the image feature extraction model to obtain the image feature information; Based on the image feature information, core feature analysis is performed to determine the core feature elements; Based on the core feature elements, the user's medical image is located and the feature regions are marked; The preprocessed image is obtained by performing local enhancement processing on the feature region.

4. The method as described in claim 3, characterized in that, The preprocessed image is obtained by performing local enhancement processing on the feature regions, including: According to the preset image display standard, the feature area is subjected to standard operation to obtain a standardized display of the area; Determine whether the standardized display of the area meets the preset clarity recognition requirements; If the target is not met, the original medical image with the marked feature region is enhanced using a preset enhancement algorithm to obtain the preprocessed image.

5. The method as described in claim 4, characterized in that, The preprocessed image is obtained by enhancing the original medical image with annotated feature regions using a preset enhancement algorithm, including: The original medical image is filtered and denoised using a preset filtering and denoising algorithm. The Laplacian pyramid algorithm is used to decompose the filtered and denoised original medical image to obtain the energy layer and detail layer; The detail layer is enhanced, and then the preprocessed image is obtained through reverse iteration.

6. A high-precision detection system for human bone mineral density values, said system being used to implement the high-precision detection method for human bone mineral density values ​​according to any one of claims 1-5, characterized in that, The system includes: The user image information acquisition module is used to acquire images of a target user through a medical image acquisition device to obtain user image information, wherein the user image information includes user medical images, acquisition device parameters, and user basic information. A bone density database construction module is used to construct a bone density database, which includes multiple sets of medical images and corresponding bone density values. A data clustering module is used to cluster data from the bone density database and construct a bone density feature mapping library. An image preprocessing module is used to extract features from the user's medical images to obtain image feature information, and to perform image preprocessing based on the image feature information and the parameters of the acquisition device. A bone density analysis module is used to identify and segment the preprocessed image, determine multiple image regions, and perform matching analysis between the multiple image regions and the bone density feature mapping library to obtain bone density analysis results. A bone density determination module is used to obtain user bone density determination information by inputting the user's basic information and the bone density analysis results into the user assessment model.

7. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-5.

Images