Medical image analysis method, apparatus and computer device

Abstract

This application relates to the field of image processing technology, and discloses a medical image analysis method, apparatus, and computer device. The method includes: acquiring image data of a target object at a target skin location to be analyzed; wherein the image data to be analyzed reflects the internal tissue structure characteristics of the target skin location; performing multi-parameter segmentation calculations on the image data to be analyzed based on initial processing parameters to generate initial detection results; wherein the initial processing parameters include initial segmentation parameters, initial region parameters, and initial layering parameters; responding to user editing instructions, performing an update operation based on the initial detection results to obtain target detection results; wherein the target detection results are used to assist in the structural evaluation and analysis of the target skin location. Its beneficial effect is that, while ensuring detection efficiency, it also considers the accuracy and adaptability of the detection results, providing reliable support for the structural evaluation and analysis of the target skin location.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a medical image analysis method, apparatus, and computer device. Background Technology

[0002] In dermatology, ultrasound, as a non-invasive and convenient medical imaging technique, has become an important tool for diagnosing skin diseases and analyzing skin structure. However, most methods for processing skin ultrasound images rely on manual segmentation, which is inefficient and lacks flexibility. Therefore, an interactive, flexible, and precisely segmented medical image analysis method is needed. Summary of the Invention

[0003] This application aims to at least partially solve one of the technical problems in related technologies. To this end, this application proposes a medical image analysis method, apparatus, and computer device. The main technical solutions adopted in this application include: In a first aspect, this application provides a medical image analysis method, which includes: acquiring image data of a target object at a target skin location to be analyzed; wherein the image data to be analyzed is used to reflect the internal tissue structure characteristics of the target skin location; performing multi-parameter segmentation calculations on the image data to be analyzed based on initial processing parameters to generate initial detection results; wherein the initial processing parameters include initial segmentation parameters, initial region parameters, and initial layering parameters; responding to user editing instructions, performing an update operation based on the initial detection results to obtain target detection results; wherein the target detection results are used to assist in structural evaluation and analysis of the target skin location.

[0004] Based on initial processing parameters, multi-parameter segmentation calculations are performed on the image data to be analyzed, quickly generating initial detection results containing the contours of each skin layer and multi-dimensional quantitative parameters, significantly improving the analysis efficiency of skin ultrasound images. Simultaneously, by responding to user editing commands, the initial results are updated and optimized, allowing the automatic analysis results to be flexibly and accurately adjusted according to the user's specific judgment and personalized needs. Ultimately, while ensuring detection efficiency, the accuracy and adaptability of the detection results are also considered, providing reliable support for the structural evaluation and analysis of target skin areas.

[0005] Optionally, in response to a user editing instruction, an update operation is performed based on the initial detection result to obtain a target detection result, including: updating the initial processing parameters using the user editing instruction to obtain updated post-processing parameters; wherein the updated post-processing parameters include at least one of updated segmentation parameters, updated region parameters, and updated layer parameters; and performing a re-detection operation based on the updated post-processing parameters, the initial detection result, and the image data to be analyzed to obtain a target detection result.

[0006] By dynamically generating updated post-processing parameters using user-edited commands, the system accurately translates users' subjective and personalized needs into executable machine instructions, achieving a deep integration of human-computer interaction and enabling highly flexible automated analysis processes. Furthermore, by utilizing initial detection results and the image data to be analyzed for re-detection, the system fully reuses the effective data from the initial detection, improving the overall efficiency of the detection process and effectively solving the problem of balancing automation and accuracy, efficiency and flexibility in traditional methods.

[0007] Optionally, the user editing instructions include slider interaction instructions, outline adjustment instructions, and selection modification instructions; the method further includes: updating the maximum and minimum values ​​and smoothness of the initial segmentation parameters based on the slider interaction instructions to obtain the updated segmentation parameters; updating the contour shape and position of the initial region parameters based on the outline adjustment instructions to obtain the updated region parameters; and updating the quantity of the initial layer parameters based on the selection modification instructions to obtain the updated layer parameters.

[0008] Real-time updates of segmentation parameters via slider-based interactive commands allow users to continuously fine-tune the sensitivity and contour quality of image segmentation, thereby improving the adaptability of segmentation parameters. Updating initial region parameters based on delineation adjustment commands enables users to arbitrarily modify, add, delete, and merge automatically identified regions, achieving refined modification and optimization of region contours and improving the accuracy of region parameters. Updating the number of layer parameters based on selection modification commands allows users to freely set the analysis precision according to actual diagnostic needs, enhancing the personalized adaptability of layer parameters.

[0009] Optionally, the method further includes: using the initial detection result as the target detection result when no user editing instruction is detected.

[0010] By directly using the initial detection results when no user editing instructions are detected, the detection efficiency is greatly improved when the results of the default parameters are ideal, without the need for additional calculation and adjustment steps.

[0011] Optionally, multi-parameter segmentation calculations are performed on the image data to be analyzed based on initial processing parameters, including: performing region segmentation processing based on the image data to be analyzed and initial segmentation parameters to obtain initial segmentation results; performing contour delineation processing based on the initial segmentation results and initial region parameters to obtain initial processed regions; performing layered segmentation and multi-parameter calculations based on the initial processed regions and initial layering parameters to obtain hierarchical analysis results; wherein, the hierarchical analysis results include quantitative parameter data for each sub-layer region; the quantitative parameter data includes depth parameters, area parameters, density parameters, and echo intensity parameters of the sub-layer region; and determining the initial detection results based on the hierarchical analysis results.

[0012] Automated region segmentation based on thresholds and contours rapidly identifies the target area for skin analysis, significantly improving the efficiency and consistency of initial region localization. Subsequent flexible contour delineation ensures a high degree of alignment between the analysis target and user intent or clinical needs. Building upon this foundation, hierarchical segmentation and multi-parameter calculations enable detailed hierarchical analysis and comprehensive quantitative assessment of skin structures. Finally, structured initial detection results are generated based on the hierarchical analysis, transforming complex image information into intuitive analytical conclusions and providing reliable data support for dermatological clinical evaluation and scientific research.

[0013] Optionally, the initial segmentation parameters include a threshold parameter and a smoothness parameter; the region segmentation process based on the image data to be analyzed and the initial segmentation parameters includes: threshold segmentation and contour extraction based on the threshold parameter and the image data to be analyzed to obtain candidate contour data; and contour smoothing optimization processing of the candidate contour data using the smoothness parameter to obtain the initial segmentation result.

[0014] By utilizing threshold segmentation and contour extraction based on brightness features, skin-related regions in the image can be automatically and initially identified from the image data to be analyzed, improving the efficiency of region recognition and reducing the subjectivity of manual delineation. Subsequently, the contour is optimized using a smoothness parameter, effectively eliminating jagged edges caused by pixel discretization, making the contour more closely resemble the smooth shape of the real skin tissue, and laying an accurate boundary foundation for subsequent precise layering and parameter calculation.

[0015] Optionally, the initial region parameters include region selection parameters; based on the initial segmentation results and the initial region parameters, contour drawing processing is performed to obtain the initial processed region, including: contour selection and confirmation processing based on the initial segmentation results and the region selection parameters to obtain the initial processed region.

[0016] By using region selection parameters for contour selection and confirmation, candidate regions related to skin structure can be accurately screened, improving the targeting of region selection. It can also flexibly adapt to different scenario requirements, such as automatic batch processing or precise manual specification, enhancing the adaptability of this solution in various application scenarios. Furthermore, the integrity verification processing provided by the preferred embodiment enables the system to promptly detect and mark region anomalies, improving the reliability of candidate regions and thus providing accurate analysis objects for subsequent layering and parameter calculation.

[0017] Optionally, the initial stratification parameters include the number of strata; stratification and multi-parameter calculation are performed based on the initial processing region and the initial stratification parameters, including: performing in-depth analysis and equidistant division processing on the initial processing region based on the number of strata to obtain at least two sub-stratified regions; and extracting and calculating multi-dimensional parameters based on each sub-stratified region to obtain the stratification analysis results.

[0018] By dividing the initial processing area into equidistant layers, the skin region is structured into multiple independent analysis units, enhancing the specificity of the analysis. Subsequently, multi-dimensional parameter extraction and calculation are performed on each sub-layer region, yielding not only basic geometric dimensional information such as depth and area, but also acoustic and statistical parameters reflecting tissue characteristics, such as density and echo intensity distribution. This provides comprehensive data support for the integrated and quantitative assessment of skin condition.

[0019] Secondly, this application provides a medical image analysis device, the device comprising: The data acquisition module is used to acquire the image data to be analyzed of the target object at the target skin; wherein, the image data to be analyzed is used to reflect the internal tissue structure characteristics of the target skin. The multi-parameter segmentation module is used to perform multi-parameter segmentation calculations on the image data to be analyzed based on the initial processing parameters, and generate initial detection results; wherein, the initial processing parameters include initial segmentation parameters, initial region parameters, and initial layer parameters; The detection result update module is used to respond to user editing commands and update the initial detection results to obtain the target detection results; the target detection results are used to assist in the structural evaluation and analysis of the target skin area.

[0020] Thirdly, this application also provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the above methods. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a flowchart of a medical image analysis method according to an embodiment of this application; Figure 2 This is a flowchart of a method for obtaining detection results according to yet another embodiment of this application; Figure 3 This is a flowchart of a detection result updating method provided according to another embodiment of this application; Figure 4 This is a structural block diagram of a medical image analysis device according to an embodiment of this application; Figure 5 This is an internal structural diagram of a computer device provided according to an embodiment of this application. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] Furthermore, in ultrasound examinations, the DICOM format, as a standard format for medical imaging, contains key information such as pixel data, physical dimensions, and equipment parameters of ultrasound images, and is the primary storage and transmission format for skin ultrasound examination data. However, related technologies for DICOM processing of skin ultrasound images have the following shortcomings: (1) Region segmentation relies on manual operation: Traditional methods require doctors to manually outline the boundaries of skin layers, which is inefficient and highly subjective. The segmentation results of different operators vary greatly, affecting the consistency of diagnosis.

[0025] (2) Lack of contour smoothness adjustment: The contours of the regions obtained by manual segmentation or simple automatic segmentation are prone to jaggedness, irregularity and other problems, which cannot accurately reflect the natural shape of skin layering and affect the accuracy of subsequent parameter calculation.

[0026] (3) Insufficient flexibility in layer division: The relevant technologies are difficult to adjust the area range and number of divisions of each skin layer according to actual detection needs, and cannot adapt to the personalized needs of different skin types and different detection sites.

[0027] (4) Limited quantitative analysis parameters: Most methods can only provide basic parameters such as area and thickness, and lack classification statistics on key indicators such as echo intensity and tissue density, which makes it difficult to meet the in-depth needs of clinical diagnosis and scientific research.

[0028] (5) Lack of interactivity and compatibility: Most existing systems have closed processing flow, which do not support users to edit the segmentation results in real time (such as adding, deleting, merging, and segmenting regions), and have poor compatibility with DICOM files from different sources and of different types, which can easily lead to loading failures or processing errors.

[0029] Based on this, according to the embodiments of this application, a medical image analysis method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] This embodiment provides a medical image analysis method, such as Figure 1 As shown, the method includes the following steps: S110. Obtain the image data to be analyzed of the target object on the target skin.

[0031] The image data to be analyzed can refer to image data reflecting the internal tissue structure characteristics of the target skin, such as ultrasound images. The target skin refers to a specific skin area of ​​the target object (such as a dermatology patient) that needs to be analyzed in medical imaging.

[0032] Specifically, the process begins with a medical image scan of the target skin area to acquire raw image data, which can be stored as a standard medical image file such as DICOM. The raw image data is then parsed and metadata extracted to determine the mapping relationship between pixels and actual physical dimensions, thus obtaining physical scale parameters. Next, image type recognition, format conversion, and data standardization are performed to eliminate invalid interference in the image and establish a unified processing benchmark, ultimately yielding the image data to be analyzed.

[0033] For example, targeting the patient's arm skin area, the area is first scanned using an ultrasound scanner to acquire raw image data. Preprocessing is then performed, which involves loading the raw image data and parsing the pixel spacing information from the file header metadata to determine the actual length of each pixel along the X and Y axes (e.g., 0.02 mm per pixel on the X-axis and 0.02 mm per pixel on the Y-axis). This actual length corresponding to a pixel is the physical scale parameter, which can be directly used in subsequent calculations to ensure that the calculation results are consistent with the actual skin structure dimensions.

[0034] Next, image type analysis is performed, which involves channel feature analysis based on the pixel array of the original image data to determine the image type. Specifically, the proportion of all-zero pixels is first calculated, and invalid blank areas in the image are eliminated based on this proportion. For non-zero valid pixels, the numerical characteristics of the three channels in their valid regions are further analyzed. Subsequently, based on the proportions of these features, the image can be divided into three types: the first type is where most pixels in the valid region have the same values ​​for all three channels, with only a small number differing; the second type is where specific channels in the valid region have mostly fixed values; and the third type is where most pixels in the valid region have different values ​​for all three channels.

[0035] Furthermore, based on the determined image type, the corresponding format conversion logic is selected. If an image is identified as type one, it is determined that the image is essentially close to a grayscale image (or a single-channel image), and it is directly converted to BGR format for subsequent processing, while the original grayscale image is retained as backup data for parameter calculation. If an image is identified as type two or three, it is determined to be a color image, and it is converted to a standard grayscale image according to a preset unified processing standard to ensure consistency in subsequent analysis. After all image conversions are completed, a custom color mapping table is constructed based on the pixel value distribution of the grayscale image to generate a pseudo-color image. This pseudo-color image can significantly enhance the visual recognition of the image and is also used as a backup to allow users to intuitively observe the skin layer structure.

[0036] Finally, the converted standard grayscale image can be standardized to linearly map pixel values ​​to a uniform range, eliminating processing biases caused by differences in device parameters, and ultimately obtaining image data that clearly reflects the internal tissue structure characteristics of the arm skin.

[0037] S120. Based on the initial processing parameters, perform multi-parameter segmentation calculations on the image data to be analyzed to generate initial detection results.

[0038] The initial processing parameters refer to the set of default parameters used in the initial analysis process, which guide the analysis of image data for region segmentation, region selection, and hierarchical calculation. Specifically, the initial processing parameters may include initial segmentation parameters, initial region parameters, and initial hierarchical parameters. The initial segmentation parameters are those used to control the identification and extraction of skin regions from the image, such as the maximum value of the threshold segmentation (v...). max The parameters include contour smoothness parameters, etc. Initial region parameters are used to guide the selection or confirmation of the skin region to be analyzed from the segmentation results, such as contour area screening threshold and region selection method identifier. Initial layering parameters are used to define how to divide the confirmed skin region into layers, such as two layers, three layers, or four layers, corresponding to different skin layer structures.

[0039] Specifically, firstly, region segmentation is performed based on the image data to be analyzed (such as a standardized grayscale image) and initial segmentation parameters. Through threshold filtering, contour extraction, and smoothing optimization, the contour of the skin region is initially identified to obtain the initial segmentation result. Then, the contour of the result is drawn based on the initial region parameters, and the skin region to be analyzed in depth is confirmed according to the preset region selection mode (such as automatic recognition) to form the initial processing region. Next, the region is divided into layers and multi-parameter calculations are performed based on the initial layering parameters. The initial processing region is divided into several sub-layered regions at equal intervals in the depth direction according to the number of layers.

[0040] Subsequently, for each sub-layer region, multi-dimensional quantitative parameters, including depth, area, tissue density, and echo intensity classification ratio based on grayscale statistics, are calculated within its corresponding image pixel range to obtain the hierarchical analysis results. Finally, the hierarchical analysis results are integrated to generate the initial detection results.

[0041] The initial detection result can refer to the analysis results data that includes contour data and parameter data of each skin layer. For example, it can take the form of a structured data report.

[0042] S130, In response to the user's editing command, perform an update operation based on the initial detection result to obtain the target detection result.

[0043] User editing commands refer to operation commands issued by users through an interactive interface to adjust the detection process and results. For example, these may include slider commands to adjust the segmentation threshold or contour smoothness; drawing commands to add, delete, merge, or move analysis regions by clicking or drawing with the mouse; and selection commands to modify the number of layers via drop-down menus or input boxes.

[0044] Specifically, firstly, based on the adjustment intent contained in the user's editing instructions, an initial processing parameter can be updated to obtain a set of updated processing parameters, such as updated segmentation thresholds, updated region contour coordinates, and updated number of layers. Subsequently, based on these updated processing parameters, combined with the original image data to be analyzed and the initial detection results, the re-detection process from region segmentation to layer calculation is retried to obtain updated skin structure analysis results, which are then used as the target detection results. These target detection results can refer to the final analysis result data determined after personalized user interaction adjustments; they are also in the form of a structured data report and are used to assist in the structural evaluation and analysis of the target skin area.

[0045] Specifically, the target detection results can include skin contour data and skin parameter data for the target skin region. The skin contour data can visually display the boundary morphology of each skin layer, while the depth and area in the skin parameter data can reflect the physical size of the layers, and the density and echo intensity ratio can help determine the tissue properties. Combined, these results can effectively assist doctors in making more accurate diagnoses of the target subject's skin health status and disease type.

[0046] Optionally, the method further includes: using the initial detection result as the target detection result when no user editing instruction is detected.

[0047] Specifically, if the user does not issue any editing instructions after the initial detection results are generated, the system determines that there is no need to adjust the initial processing parameters and the initial detection results. In this case, the initial detection results can be directly determined as the final target detection results for subsequent skin structure evaluation and analysis.

[0048] For example, after performing multi-parameter partitioning calculations based on default initial processing parameters, generating initial detection results, and displaying them on the visual interface, the system enters a state of waiting for or listening for user interaction commands. If, within a preset time or before the user explicitly confirms the initial detection results, the system does not capture any valid user editing commands, it determines that the user approves or that no modification to the current detection results is needed. Taking user explicit confirmation of results as an example, if a clinician reviews the results and deems the automatic analysis accurate and reliable, requiring no manual fine-tuning, they can directly click the "Confirm" or "Save Report" button. In this case, the system detects a "Confirm" command rather than an "Edit" command, thus not triggering the parameter update and recalculation process, and directly outputting the currently displayed initial detection results as the final report. By directly using the initial detection results when no user editing commands are detected, and when the results of the default parameters are ideal, no additional calculation and adjustment steps are required, significantly improving detection efficiency.

[0049] In the above implementation, multi-parameter segmentation calculations are performed on the image data to be analyzed based on initial processing parameters, quickly generating initial detection results containing the contours of each skin layer and multi-dimensional quantitative parameters, significantly improving the analysis efficiency of skin ultrasound images. Simultaneously, by responding to user editing commands, the initial results are updated and optimized, allowing the automatic analysis results to be flexibly and accurately adjusted according to the user's specific judgment and personalized needs. Ultimately, while ensuring detection efficiency, the accuracy and adaptability of the detection results are also considered, providing reliable support for the structural evaluation and analysis of the target skin area.

[0050] In some implementations, multi-parameter partitioning calculations are performed on the image data to be analyzed based on initial processing parameters. Please refer to the appendix. Figure 2 ,include: S210. Based on the image data to be analyzed and the initial segmentation parameters, perform region segmentation processing to obtain the initial segmentation result.

[0051] The initial segmentation parameters include threshold parameters and smoothness parameters.

[0052] The threshold parameter can refer to the maximum and minimum brightness values ​​used to distinguish the target area from the background in an image. These two values ​​can form a brightness parameter range, such as [0, 100]. Setting the threshold parameter to this range will retain the areas corresponding to pixels with gray values ​​in the range of 0-100 as potential skin areas. The smoothness parameter can refer to the precision parameter that controls the degree of simplification of the segmentation contour to eliminate jagged edges. For example, a smoothness parameter of 2 can be set to represent simplification of the contour with medium precision, optimizing edge smoothness while preserving the core structure.

[0053] Specifically, region segmentation based on the image data to be analyzed and initial segmentation parameters may include the following steps: first, threshold segmentation and contour extraction are performed based on threshold parameters and the image data to be analyzed to obtain candidate contour data; then, contour smoothing optimization is performed on the candidate contour data using smoothness parameters to obtain the initial segmentation result.

[0054] Candidate contour data can refer to a set of contour data that can initially identify potential skin regions. That is, after the target region is screened out by threshold segmentation, a set of contour lines that may belong to the skin structure are drawn from the image data to be analyzed.

[0055] The initial segmentation result can refer to the set of contour data obtained after further smoothing and optimization processing based on the candidate contour data, which can be used to describe the potential skin region in the image data to be analyzed.

[0056] For example, the image data to be analyzed is first converted from a standardized grayscale image to the HSV color space so that the luminance channel can be processed separately. Simultaneously, invalid areas at the image corners are masked to reduce irrelevant interference. Then, a preset threshold parameter (e.g., a minimum value v) is used... min =20, highest value v max A binary mask image is generated by marking regions with grayscale values ​​within the range of 100 in the luminance channel as foreground (potential skin regions) and the remaining regions as background. Then, a contour tracking algorithm is used to traverse the connected regions in the binary mask, extracting the boundary coordinates of each foreground region to obtain the original contour set of all potential skin regions. Optionally, to further eliminate noise, a contour area filtering threshold (e.g., 1000 pixels) can be set to filter out noise points and irrelevant small regions with areas smaller than this threshold; the remaining contour coordinate set is the candidate contour data.

[0057] After obtaining the candidate contour data, the simplification accuracy threshold for each candidate contour is first calculated by combining a preset smoothness parameter (e.g., a smoothness value of 2, corresponding to a medium precision threshold) and the perimeter of each candidate contour in all candidate contour data. Then, a polygon approximation algorithm (such as the Douglas-Peucker algorithm) is used, which removes redundant detail vertices based on the smoothness parameter and approximates the original contour of each candidate contour before smoothing with a small number of key points. Next, the simplified contours are further checked for strong collinearity or excessively close points; if any are found, they are merged and optimized to ensure smooth and natural edges. Finally, the resulting smoothed and optimized contour set is used as the initial segmentation result.

[0058] By utilizing threshold segmentation and contour extraction based on brightness features, skin-related regions in the image can be automatically and initially identified from the image data to be analyzed, improving the efficiency of region recognition and reducing the subjectivity of manual delineation. Subsequently, the contour is optimized using a smoothness parameter, effectively eliminating jagged edges caused by pixel discretization, making the contour more closely resemble the smooth shape of the real skin tissue, and laying an accurate boundary foundation for subsequent precise layering and parameter calculation.

[0059] S220. Based on the initial segmentation results and initial region parameters, perform contour drawing to obtain the initial processed region.

[0060] The initial region parameters include region selection parameters, which can refer to the region selection mode and related judgment criteria for processing the skin region to be analyzed in detail. For example, the region selection mode can include automatic recognition or manual selection; the related judgment criteria can be sorting from top to bottom according to vertical position, selecting the first N contours, or subjective selection.

[0061] The initial processing region can refer to a set of one or more image regions that clearly describe the layered structures of the skin region to be analyzed in depth.

[0062] Specifically, contour selection and confirmation can be performed based on the initial segmentation results and region selection parameters to obtain the initial processing region.

[0063] For example, since ultrasound images of the skin can vertically display the four main skin-related structures—the epidermis, dermis, subcutaneous tissue, and muscle layer—if the region selection mode set in the region selection parameters is automatic recognition, and the relevant judgment criterion is to select the first four main contours by vertical position, then firstly, all contours in the initial segmentation result can be sorted from top to bottom according to their vertical position (Y-axis direction). Next, the first four contours are selected as the undetermined skin regions to be analyzed further. For each undetermined skin region, effective center points are found within it by calculating the geometric center coordinates of the contour, and these center points are used as markers for the region of interest (RoI) to identify the core location of each undetermined skin region. Finally, these four selected and marked undetermined skin regions constitute the initial processing area.

[0064] Optionally, taking the region selection mode set in the region selection parameters as manual recognition and the relevant judgment criteria as subjective selection as an example, the system can recognize the user-specified region of interest in response to user mouse clicks or multi-point drawing. For example, if the user clicks a region with a single point, the system automatically identifies the contour where that point is located and selects that contour as the pending skin region. If the user draws a closed polygon with multiple points, the system directly uses the contour corresponding to that polygon as the pending skin region. If the user draws non-closed line segments with multiple points, an contour is generated through automatic line segment feature processing (similarity recognition or automatic supplementation) as the pending skin region. Then, the region of interest is marked and integrated to form the initial processing area.

[0065] Preferably, after contour selection and confirmation, integrity verification can also be performed. This involves first performing contour selection and confirmation based on the initial segmentation results and region selection parameters to obtain candidate region data. Then, integrity verification is performed based on the candidate region data to obtain the delineation association results. Finally, the initial processing region is determined based on the delineation association results and the candidate region data.

[0066] Candidate region data refers to the set of contour data initially selected from the initial segmentation results that can enter the subsequent analysis process; that is, the candidate contour data most likely representing the various layers of skin structure selected from all smoothed and optimized contours. Similarly, the selection methods and judgment criteria included in the region selection parameters can be used to sort, select, and mark points of interest in all contours in the initial segmentation results, thereby obtaining candidate region data.

[0067] Subsequently, a delineation and association result can be obtained based on the candidate region data. This delineation and association result can be a verification result formed after confirming the logical and topological relationships of the candidate region data, indicating whether the outline in the candidate region data is complete and whether there are any omissions.

[0068] For example, integrity checks may include contour integrity checks, coverage integrity checks, and region correlation checks.

[0069] Specifically, contour integrity verification checks whether the contour is a closed curve and whether there are obvious morphological abnormalities such as breakpoints, severe contour distortion, or discontinuities. Coverage integrity verification combines the conventional structure of skin ultrasound images to determine whether the contours in the candidate region data correspond to these conventional structures and whether there are obvious omissions. Region correlation verification first verifies whether the arrangement of each candidate region conforms to the natural hierarchical relationship of the skin structure. For example, taking the vertical relationship as an example, it checks whether it conforms to the progressive relationship of the epidermis above and the muscle layer below. Secondly, it detects whether there is a reasonable spacing between each contour and whether there are abnormalities such as overlap or excessive spacing. Finally, the results of the above multi-dimensional verification are integrated to form the delineation correlation result.

[0070] The initial processing area can then be determined using the delineation results and candidate region data.

[0071] For example, if the delineation results show that the contours in the candidate region data simultaneously meet the requirements of contour integrity compliance, coverage integrity compliance, and region correlation compliance, it indicates that the candidate region data is complete and without anomalies. In this case, the contours in the candidate region data can be directly determined as the initial processing region, and the region corresponding to each contour is treated as a sub-region within the initial processing region. If non-compliance exists, a prompt is given so that the user can choose automatic repair (e.g., supplementing breakpoints through curve interpolation) or manual repair (re-delineating and supplementing). After the repair is completed, integrity verification is performed again until all are compliant. The last repaired candidate region data is then determined as the initial processing region.

[0072] By using region selection parameters for contour selection and confirmation, candidate regions related to skin structure can be accurately screened, improving the targeting of region selection. It can also flexibly adapt to different scenario requirements, such as automatic batch processing or precise manual specification, enhancing the adaptability of this solution in various application scenarios. Furthermore, the integrity verification processing provided by the preferred embodiment enables the system to promptly detect and mark region anomalies, improving the reliability of candidate regions and thus providing accurate analysis objects for subsequent layering and parameter calculation.

[0073] S230. Based on the initial processing region and initial stratification parameters, perform stratification and multi-parameter calculation to obtain the stratification analysis results.

[0074] Understandably, the skin is a complex tissue composed of multiple structural layers, including the epidermis, dermis, subcutaneous tissue, and muscle layer, with significant differences in physiological structure and pathological characteristics between these layers. Analyzing only the overall area of ​​the target skin cannot accurately reflect the specific conditions of each skin layer. Therefore, it is necessary to divide the initial processing area into multiple sub-layers corresponding to the actual skin structure. Then, multi-parameter calculations are performed on each sub-layer to obtain specific quantitative indicators for each layer, ultimately generating a hierarchical analysis result that comprehensively describes the condition of each skin layer.

[0075] The hierarchical analysis result can refer to a comprehensive dataset containing the contour information of each sublayer and its corresponding quantitative parameters. Specifically, the hierarchical analysis result includes quantitative parameter data for each sublayer region, where a sublayer region can refer to a region with independent structural characteristics obtained after decomposing the actual skin structure layers. The quantitative parameter data includes depth parameters, area parameters, density parameters, and echo intensity parameters for each sublayer region. For example, taking forearm skin analysis as an example, if the initial processing area is divided into four layers, four sublayer regions will be generated, corresponding to the epidermis, dermis, subcutaneous tissue, and muscle layer, respectively. The quantitative parameter data is a set of numerical indicators for each sublayer. The depth parameter reflects the physical depth of the sublayer in the vertical direction, including average depth, maximum depth, and minimum depth; the area parameter reflects the actual physical area of ​​the sublayer; the density parameter reflects the density of the tissue within the sublayer; and the echo intensity parameter reflects the ultrasound echo characteristics within the sublayer, which can be characterized by the proportion of pixels with different brightness levels within that tissue layer.

[0076] Furthermore, the initial layering parameters include the number of layers, which can refer to the number of sub-layers that the user sets or the system presets, and the expected number of sub-layers into which the initial processing area is divided.

[0077] Specifically, the hierarchical division and multi-parameter calculation based on the initial processing region and initial hierarchical parameters may include the following steps: First, perform in-depth analysis and equidistant division of the initial processing region based on the number of layers to obtain at least two sub-hierarchical regions; then, extract and calculate multi-dimensional parameters based on each sub-hierarchical region to obtain the hierarchical analysis results.

[0078] For example, taking the initial processing area as the contour region corresponding to the arm skin and the number of layers in the initial layering parameters as four, firstly, a mask corresponding to the contour of the initial processing area is created. The coordinate information of all pixels within this mask is extracted to determine the distribution range of pixels in the vertical direction (Y-axis), i.e., the maximum value (Y). max ) and minimum value (Y) min Then, the difference between the maximum and minimum pixel coordinates in the vertical direction is calculated to obtain the total depth of the initial processing region.

[0079] However, it should be noted that the total depth calculated here is in pixel coordinates. Therefore, it needs to be converted into the actual physical total depth by combining it with pre-calculated physical scale parameters (e.g., 1 pixel = 0.02 mm). Next, based on the number of layers, the total depth is divided into four equal parts to determine the theoretical depth interval for each layer. Then, using the top boundary of the initial processing area as a reference, the system generates three layer boundary curves that follow the natural curvature of the edge of the initial processing area downwards according to this depth interval, thus dividing the initial processing area into four layers. Finally, the boundary curve of each layer is made to close with the left and right boundaries of the initial processing area, forming four independent connected regions, which serve as sub-layer regions corresponding to the epidermis, dermis, subcutaneous tissue, and muscle layer, respectively.

[0080] After obtaining the sub-layered regions, multi-dimensional parameters can be extracted and calculated for each region.

[0081] For example, all pixel coordinates and corresponding grayscale values ​​within the contour of each sub-layer region can be extracted. Then, based on the depth range of each region and the physical scale parameter, the median value of the depth range is taken as the average depth of the layer, the maximum value is taken as the maximum depth, and the minimum value is taken as the minimum depth.

[0082] Then, the number of pixels within the contour of each sub-layer region is counted, and the actual physical area corresponding to each pixel is calculated based on the physical scale parameters. Finally, the actual physical area of ​​the sub-layer region is obtained by multiplying the number of pixels by the actual physical area.

[0083] Next, the average grayscale value of all pixels in each sub-layer region is calculated, and this average value is linearly mapped to the range of 0-1 to obtain the normalized density parameter.

[0084] Next, calculate the echo intensity parameters: First, set a standard echo intensity range based on user needs or industry standards. For example, a grayscale value of 0-80 is considered low echo, 80-160 is considered iso-echo, and 160-255 is considered high echo. Then, count the number of pixels within each sub-layer region whose grayscale values ​​fall within each echo intensity range, and calculate the ratio of the number of pixels in each range to the total number of pixels in that layer to obtain the echo intensity parameters. For example, if a sub-layer region has a total of 50,000 pixels and 32,500 low echo pixels, then the low echo percentage is 32,500 / 50,000 = 65%. Finally, integrate the contour data (coordinate information) of each sub-layer region with the calculated depth, area, density, and echo intensity percentage parameters to form the hierarchical analysis results.

[0085] By dividing the initial processing area into equidistant layers, the skin region is structured into multiple independent analysis units, enhancing the specificity of the analysis. Subsequently, multi-dimensional parameter extraction and calculation are performed on each sub-layer region, yielding not only basic geometric dimensional information such as depth and area, but also acoustic and statistical parameters reflecting tissue characteristics, such as density and echo intensity distribution. This provides comprehensive data support for the integrated and quantitative assessment of skin condition.

[0086] S240. Determine the initial detection results based on the analytic hierarchy process (AHP) results.

[0087] Specifically, the hierarchical analysis results can be organized and standardized. For example, the contour data of each sub-layer region can be stored in a unified format to ensure data integrity and readability. Alternatively, based on the core requirements of skin detection, the contour data can be correlated and matched with corresponding parameter data to form a complete analysis record for each sub-layer region. Finally, the complete analysis records of all sub-layer regions are integrated to form an initial detection result that comprehensively reflects the structural characteristics of each layer of the target skin. For example, the initial detection results can be organized according to a predefined data structure (such as JSON format) to generate an initial detection result report that can be displayed, saved, or exported. Furthermore, the results can be displayed on the user's device through a visual interface for viewing and subsequent editing.

[0088] In the above implementation, automated region segmentation based on thresholds and contours quickly identifies the target area for skin analysis, significantly improving the efficiency and consistency of initial region localization. Subsequently, flexible contour delineation ensures a high degree of alignment between the analysis target and user intent or clinical needs. Building upon this, hierarchical segmentation and multi-parameter calculations enable detailed hierarchical analysis and comprehensive quantitative assessment of skin structures. Finally, structured initial detection results are generated based on the hierarchical analysis results, transforming complex image information into intuitive analytical conclusions and providing reliable data support for dermatological clinical evaluation and scientific research.

[0089] In some implementations, in response to user editing instructions, an update operation is performed based on the initial detection results to obtain the target detection results. Please refer to the appendix. Figure 3 ,include: S310. Update the initial processing parameters using user editing commands to obtain the updated processing parameters.

[0090] The updated processing parameters can refer to a set of parameters modified according to the user's interaction intent, including at least one of the updated segmentation parameters, updated region parameters, and updated layering parameters.

[0091] It should be noted that the initial segmentation parameters, initial region parameters, and initial stratification parameters are the original parameter sets used to generate the initial detection results. The updated segmentation parameters, updated region parameters, and updated stratification parameters, on the other hand, are new parameter sets obtained after user intervention, through targeted adjustments to the aforementioned original parameters. These new parameters can replace or partially replace the original parameters to drive a new round of analysis and calculation, thereby generating target detection results.

[0092] Optionally, user editing instructions may include slider interaction instructions, outline adjustment instructions, and selection modification instructions. Specifically, the method further includes: updating the initial segmentation parameters for maximum and minimum values ​​and smoothness based on slider interaction instructions to obtain updated segmentation parameters; updating the initial region parameters for contour shape and position based on outline adjustment instructions to obtain updated region parameters; and updating the initial layer parameters for quantity based on selection modification instructions to obtain updated layer parameters.

[0093] Here, the slider interaction command refers to the instruction that the user continuously adjusts the processing parameters by manipulating the slider in the interactive interface. For example, the user can manipulate v... max The slider adjusts the maximum value of the threshold parameter from 80 to 100, or the smoothness slider adjusts the parameter from 2 to 3, thus obtaining the updated segmentation parameters.

[0094] Drawing and adjustment commands refer to instructions that allow users to modify the shape, position, or topological relationship of a region's outline through mouse clicks or drawing. For example, users can manually draw closed curves to add missing skin areas, drag outline key points to adjust the region's position, or click to delete / select a region to obtain updated region parameters.

[0095] A selected modification command refers to an instruction that allows a user to modify a specific option or numerical parameter by selecting an option in the interactive interface or entering a value. For example, a user can change the number of layers from four to three through a drop-down menu, or change the number of layers from two to five through an input box, thereby obtaining the updated layer parameters.

[0096] Real-time updates of segmentation parameters via slider-based interactive commands allow users to continuously fine-tune the sensitivity and contour quality of image segmentation, thereby improving the adaptability of segmentation parameters. Updating initial region parameters based on delineation adjustment commands enables users to arbitrarily modify, add, delete, and merge automatically identified regions, achieving refined modification and optimization of region contours and improving the accuracy of region parameters. Updating the number of layer parameters based on selection modification commands allows users to freely set the analysis precision according to actual diagnostic needs, enhancing the personalized adaptability of layer parameters.

[0097] S320. Based on the updated post-processing parameters, initial detection results, and image data to be analyzed, a re-detection operation is performed to obtain the target detection result.

[0098] It should be noted that responding to user editing commands to update the corresponding parameters does not simply change the specific values ​​in the initial detection results, but triggers a complete re-detection process.

[0099] Specifically, because this detection scheme has interconnected progressive steps, if the user updates the parameters used in the first step, the first step needs to be re-executed using the original input and the updated parameters. However, if the user updates the parameters used in the second step, they can either execute the process from scratch (execute the first step using the original input and initial parameters, and then re-execute the second step using the output of the first step and the updated parameters), or directly call the updated parameters from the first step included in the initial detection result to re-execute the second step, thereby reducing the computational load.

[0100] For example, if the user only updates the initial segmentation parameters to their maximum and minimum values ​​and smoothness, then the image data to be analyzed needs to be called, and the region segmentation process needs to be re-executed in conjunction with the updated segmentation parameters. Specifically, threshold segmentation and contour extraction are performed using the updated threshold parameters and the image data to be analyzed to obtain updated candidate contour data; then, contour smoothing optimization is performed on the updated candidate contour data using the updated smoothness parameters to obtain the target segmentation result. Subsequently, contour delineation is performed based on the target segmentation result as the base data to obtain the target processing region. Then, hierarchical division and multi-parameter calculation are performed based on the target processing region to obtain the updated hierarchical analysis result. Finally, the target detection result is determined based on the updated hierarchical analysis result.

[0101] If the user only updates the contour shape and position of the initial region parameters, then on one hand, the image data to be analyzed can be retrieved again, and region segmentation processing can be performed in conjunction with the initial segmentation parameters to obtain the initial segmentation result. Then, based on the initial segmentation result, contour delineation processing can be performed using the updated region parameters to obtain the target processing region. On the other hand, the initial segmentation result can also be retrieved through reverse analysis based on the initial detection result, and the updated region parameters can be directly used to perform contour delineation processing on the initial segmentation result to obtain the target processing region. Furthermore, the target detection result can be determined based on the target processing region.

[0102] Furthermore, if a user updates multiple initial parameters simultaneously, the system needs to follow the step order of the detection scheme (region segmentation, region selection, and hierarchical calculation), jump back to the first step in the sequence, and perform a re-detection operation to update the parameters, thereby obtaining the updated target detection result.

[0103] In the above implementation, user-edited instructions are used to dynamically generate updated post-processing parameters, accurately translating the user's subjective and personalized needs into executable machine instructions. This achieves a deep integration of human-computer interaction and makes the automated analysis process highly flexible. Furthermore, re-detection operations are performed using the initial detection results and the image data to be analyzed, fully reusing the effective data from the initial detection and improving the overall efficiency of the detection process. This effectively solves the problem of traditional methods struggling to achieve both automation and accuracy, as well as efficiency and flexibility.

[0104] It should be understood that although the steps in the flowchart above are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart above may include multiple steps or stages, which are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps.

[0105] This specification also provides a medical image analysis device 400, such as... Figure 4 As shown, it includes: a data acquisition module 410, a multi-parameter partitioning module 420, and a detection result update module 430, wherein: The data acquisition module 410 is used to acquire the image data to be analyzed of the target object on the target skin; wherein, the image data to be analyzed refers to the image data obtained by scanning the area of ​​the target skin of the target object.

[0106] The multi-parameter segmentation module 420 is used to perform multi-parameter segmentation calculations on the image data to be analyzed based on the initial processing parameters, and generate initial detection results; wherein, the initial processing parameters include initial segmentation parameters, initial region parameters and initial layering parameters.

[0107] The detection result update module 430 is used to respond to user editing instructions and perform update operations based on the initial detection results to obtain the target detection results; wherein, the target detection results are used to assist in the structural evaluation and analysis of the target skin area.

[0108] In some implementations, the detection result update module 430 is further configured to update the initial processing parameters using user editing instructions to obtain updated post-processing parameters; wherein, the updated post-processing parameters include at least one of updated segmentation parameters, updated region parameters, and updated layer parameters; and perform a re-detection operation based on the updated post-processing parameters, the initial detection result, and the image data to be analyzed to obtain the target detection result.

[0109] In some implementations, user editing instructions include slider interaction instructions, outline adjustment instructions, and selection modification instructions; the detection result update module 430 is also used to update the maximum and minimum values ​​and smoothness of the initial segmentation parameters based on the slider interaction instructions to obtain updated segmentation parameters; to update the contour shape and position of the initial region parameters based on the outline adjustment instructions to obtain updated region parameters; and to update the quantity of the initial layer parameters based on the selection modification instructions to obtain updated layer parameters.

[0110] In some implementations, the detection result update module 430 is also used to use the initial detection result as the target detection result when no user editing instruction is detected.

[0111] In some implementations, the multi-parameter segmentation module 420 is further used to perform region segmentation processing based on the image data to be analyzed and initial segmentation parameters to obtain an initial segmentation result; perform contour delineation processing based on the initial segmentation result and initial region parameters to obtain an initial processing region; perform layer division and multi-parameter calculation based on the initial processing region and initial layering parameters to obtain a hierarchical analysis result; wherein, the hierarchical analysis result includes quantitative parameter data for each sub-layer region; the quantitative parameter data includes depth parameters, area parameters, density parameters, and echo intensity parameters of the sub-layer region; and determine the initial detection result based on the hierarchical analysis result.

[0112] In some implementations, the multi-parameter segmentation module 420 is also used to perform threshold segmentation and contour extraction processing based on threshold parameters and image data to be analyzed to obtain candidate contour data; and to perform contour smoothing optimization processing on the candidate contour data using smoothness parameters to obtain initial segmentation results.

[0113] In some implementations, the multi-parameter segmentation module 420 is also used to perform contour selection and confirmation processing based on the initial segmentation result and region selection parameters to obtain the initial processing region.

[0114] In some implementations, the initial layering parameters include the number of layers; the multi-parameter partitioning module 420 is also used to perform in-depth analysis and equidistant partitioning of the initial processing region based on the number of layers to obtain at least two sub-layered regions; and to extract and calculate multi-dimensional parameters based on each sub-layered region to obtain the hierarchical analysis results.

[0115] For specific limitations regarding a medical image analysis device, please refer to the limitations regarding a medical image analysis method above, which will not be repeated here. Each module in the aforementioned medical image analysis device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0116] In this embodiment, a medical image analysis device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above-mentioned functions.

[0117] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application, such as... Figure 5 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processor can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations. Figure 5 Take a processor 10 as an example.

[0118] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0119] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0120] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0121] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0122] The computer device also includes an input device 30 and an output device 40. The processor 10, memory 20, input device 30, and output device 40 can be connected via a bus or other means. Figure 5 Taking the example of a connection between China and Israel via a bus.

[0123] Input device 30 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the computer device, such as a touchscreen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. Output device 40 may include display devices, auxiliary lighting devices (e.g., LEDs), and haptic feedback devices (e.g., vibration motors). The aforementioned display devices include, but are not limited to, liquid crystal displays, light-emitting diodes, displays, and plasma displays. In some alternative embodiments, the display device may be a touchscreen.

[0124] This application also provides a computer-readable storage medium. The methods described in this application can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded over a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the methods shown in the above embodiments are implemented.

[0125] This application provides a computer program product including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method of any embodiment of this application.

[0126] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and all such modifications and variations fall within the scope defined by the appended claims.

[0127] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0128] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware, such as the electronic device, application, server, or storage medium performing the operations of this disclosed technical solution, based on the prompt message.

[0129] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0130] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant regulations may also be applied to the implementation of this disclosure.

[0131] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws and regulations.

[0132] It is understood that in the specific embodiments of this application, data such as user information, location information, and navigation data are involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws and standards of the relevant countries and regions.

[0133] The apparatus, module, or unit described in the above embodiments can be implemented by a computer chip or entity, or by a product having a certain function. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0134] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0135] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0136] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0137] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0138] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0139] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0140] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0141] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0142] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A medical image analysis method, characterized in that, The method includes: Acquire image data of the target object at the target skin location to be analyzed; wherein, the image data to be analyzed is used to reflect the internal tissue structure characteristics at the target skin location; The image data to be analyzed is subjected to multi-parameter segmentation calculations based on initial processing parameters to generate initial detection results. The initial processing parameters include initial segmentation parameters, initial region parameters, and initial layering parameters. The multi-parameter segmentation calculations include: performing region segmentation processing based on the image data to be analyzed and the initial segmentation parameters to obtain initial segmentation results; performing contour delineation processing based on the initial segmentation results and initial region parameters to obtain initial processed regions; performing depth analysis and equidistant segmentation processing on the initial processed regions based on the number of layers in the initial layering parameters to obtain at least two sub-layered regions; and performing multi-dimensional parameter extraction and calculation based on each sub-layered region to obtain hierarchical analysis results. The hierarchical analysis results include quantitative parameter data for each sub-layered region. The quantitative parameter data includes depth parameters, area parameters, density parameters, and echo intensity parameters of the sub-layered region; the depth parameters include average depth, maximum depth, and minimum depth; the depth parameters are calculated based on pixel coordinates, grayscale values, and physical scale parameters within the contour of the sub-layered region; the area parameters reflect the actual physical area of ​​the sub-layered region and are calculated using the number of pixels within the contour of the sub-layered region and physical scale parameters; the density parameters reflect the density of the tissue within the sub-layered region and are obtained by mapping the average grayscale value of all pixels within the sub-layered region; the echo intensity parameters are characterized by the proportion of pixels with different brightness within the sub-layered region; the initial detection results are determined based on the hierarchical analysis results. In response to a user editing command, an update operation is performed based on the initial detection result to obtain a target detection result; wherein, the target detection result is used to assist in structural evaluation and analysis of the target skin area; the update operation includes: updating the initial processing parameters using the user editing command to obtain updated post-processing parameters; wherein, the updated post-processing parameters include at least one of updated segmentation parameters, updated region parameters, and updated layering parameters; and performing a re-detection operation based on the updated post-processing parameters, the initial detection result, and the image data to be analyzed to obtain the target detection result.

2. The method according to claim 1, characterized in that, The user editing commands include slider interaction commands, outline adjustment commands, and selection modification commands; the method further includes: Based on the slider interaction command, the initial segmentation parameters are updated with maximum and minimum values ​​and smoothness to obtain the updated segmentation parameters; Based on the outline adjustment command, the initial region parameters are updated in terms of outline shape and position to obtain the updated region parameters; The initial layering parameters are updated based on the selected modification instruction to obtain the updated layering parameters.

3. The method according to claim 1, characterized in that, The method further includes: If the user's editing instruction is not detected, the initial detection result is taken as the target detection result.

4. The method according to claim 1, characterized in that, The initial segmentation parameters include a threshold parameter and a smoothness parameter; the region segmentation process based on the image data to be analyzed and the initial segmentation parameters includes: Based on the threshold parameter and the image data to be analyzed, threshold segmentation and contour extraction are performed to obtain candidate contour data. Using the smoothness parameter, contour smoothing optimization processing is performed on the candidate contour data to obtain the initial segmentation result.

5. The method according to claim 1, characterized in that, The initial region parameters include region selection parameters; the contour drawing process based on the initial segmentation result and the initial region parameters to obtain the initial processed region includes: Based on the initial segmentation result and the region selection parameters, contour selection and confirmation processing are performed to obtain the initial processed region.

6. A medical image analysis device, characterized in that, The device includes: The data acquisition module is used to acquire image data of the target object at the target skin to be analyzed; wherein, the image data to be analyzed is used to reflect the internal tissue structure characteristics of the target skin. A multi-parameter partitioning module is used to perform multi-parameter partitioning calculations on the image data to be analyzed based on initial processing parameters to generate initial detection results. The initial processing parameters include initial segmentation parameters, initial region parameters, and initial layering parameters. The multi-parameter partitioning calculation includes: performing region segmentation processing based on the image data to be analyzed and the initial segmentation parameters to obtain initial segmentation results; performing contour delineation processing based on the initial segmentation results and the initial region parameters to obtain initial processed regions; performing depth analysis and equidistant partitioning processing on the initial processed regions based on the number of layers in the initial layering parameters to obtain at least two sub-layered regions; and performing multi-dimensional parameter extraction and calculation based on each sub-layered region to obtain hierarchical analysis results. The hierarchical analysis results include the definition of each sub-layered region. The quantitative parameter data includes depth, area, density, and echo intensity parameters of the sub-layered region. The depth parameters include average depth, maximum depth, and minimum depth. These depth parameters are calculated based on pixel coordinates, grayscale values, and physical scale parameters within the sub-layered region's outline. The area parameter reflects the actual physical area of ​​the sub-layered region and is calculated using the number of pixels within the sub-layered region's outline and physical scale parameters. The density parameter reflects the density of the tissue within the sub-layered region and is obtained by mapping the average grayscale value of all pixels within the sub-layered region. The echo intensity parameter is characterized by the proportion of pixels with different brightness levels within the sub-layered region. The initial detection result is determined based on the hierarchical analysis results. The detection result update module is used to respond to user editing instructions and perform update operations based on the initial detection results to obtain target detection results; wherein, the target detection results are used to assist in structural evaluation and analysis of the target skin area; the update operation includes: updating the initial processing parameters using the user editing instructions to obtain updated post-processing parameters; wherein, the updated post-processing parameters include at least one of updated segmentation parameters, updated region parameters, and updated layering parameters; and performing a re-detection operation based on the updated post-processing parameters, the initial detection results, and the image data to be analyzed to obtain the target detection results.

7. A computer device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of any one of claims 1 to 5.

Images