Method, system and storage medium for texture evaluation of ct image gray level variation function for respiratory or skeletal tissue morphology assessment

Abstract

The application discloses a kind of based on medical image three-dimensional gray variation function respiratory or skeletal tissue morphological evaluation method, system, platform and storage medium;The present application scheme is analyzed by constructing three-dimensional gray variation function to respiratory or skeletal tissue in multidirectional, multiscale space variation, and the curve shape parameter with clear physical meaning is extracted in double logarithmic domain, can be quantitatively evaluated on the premise of not needing to finely segmented tissue structure, stable, repeatable three-dimensional texture characteristics of organization, and effectively solve the technical problems of two-dimensional texture analysis structure information loss, scale representation is not clear, direction sensitivity is strong and high dependence on fine segmentation;Meanwhile, the present application scheme also evaluates and outputs the reliability mark by built-in quality control rule to fitting point number, fitting residual and fitting goodness, improves the engineering reliability and clinical usability of batch processing result, can be uniformly applicable to respiratory tissue and skeletal tissue morphological quantitative evaluation.

Description

Technical Field

[0001] This invention belongs to the field of medical image processing technology, specifically relating to a method, system, platform, and storage medium for morphological evaluation of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images. Background Technology

[0002] Chronic respiratory diseases such as chronic obstructive pulmonary disease (COPD) are often accompanied by imaging findings such as lung parenchymal destruction and airway remodeling; some patients may also have skeletal abnormalities such as osteopenia and bone tissue degeneration. Therefore, performing a unified three-dimensional morphological quantitative assessment of respiratory and skeletal tissues based on the same routine CT examination has significant clinical application value.

[0003] In current clinical practice, chest CT scans are primarily used for observing abnormalities in the lung parenchyma; the skeletal system often relies on indicators such as bone mineral density or mean CT value. These indicators mainly reflect the overall tissue attenuation level and are insufficient to express the complex spatial texture and scale changes within the tissue. Although existing research has employed methods such as gray-level co-occurrence matrices and local binary patterns, there are still limitations: two-dimensional texture analysis cannot fully reflect the true three-dimensional structure; some texture features lack clear spatial scale meaning; moreover, traditional features are sensitive to sampling direction and parameter settings, resulting in insufficient stability; their reliance on high-resolution reconstruction or fine segmentation limits their application in routine clinical volume data; and there is a lack of unified quality control and reliability labeling.

[0004] Therefore, in order to address the aforementioned technical deficiencies, there is an urgent need to design and develop a method, system, platform, and storage medium for morphological assessment of respiratory or skeletal tissues based on the three-dimensional grayscale variation function of medical images. Summary of the Invention

[0005] To overcome the shortcomings and difficulties of the existing technology, the purpose of this invention is to provide a method, system, platform and storage medium for morphological evaluation of respiratory or skeletal tissues based on the three-dimensional grayscale variation function of medical images, so as to be able to directly deal with three-dimensional volume data, without relying on fine segmentation, and to stably extract texture change patterns in multiple directions and multiple spatial scales.

[0006] The first objective of this invention is to provide a method for evaluating the morphology of respiratory or skeletal tissues based on the three-dimensional grayscale variation function of medical images; the second objective of this invention is to provide a system for evaluating the morphology of respiratory or skeletal tissues based on the three-dimensional grayscale variation function of medical images; the third objective of this invention is to provide a platform for evaluating the morphology of respiratory or skeletal tissues based on the three-dimensional grayscale variation function of medical images; and the fourth objective of this invention is to provide a computer-readable storage medium.

[0007] The first objective of this invention is achieved by the method comprising the following steps: First data corresponding to the target tissue is generated and acquired, and second data corresponding to the target area of ​​the first data is created based on the first data; wherein, the first data is three-dimensional medical image volume data, preferably three-dimensional CT volume data; the second data is three-dimensional volume of interest, and in this scheme, the second data is specifically three-dimensional volume of interest (VOI) data; A fourth set of data corresponding to the second set of data is constructed, and a fifth set of data corresponding to the fourth set of data is created; wherein, the fourth set of data is a three-dimensional grayscale variation function; and the fifth set of data is a sequence of variation function values. In the double logarithmic domain, the fifth data is fitted and processed, and at least one corresponding sixth data is extracted to represent the spatial variation law of tissue texture; wherein, the sixth data is a curve morphology parameter; Based on the sixth data, corresponding seventh data is created, generated, and output; wherein, the seventh data is the evaluation result data.

[0008] Furthermore, the process of constructing and generating fourth data corresponding to the second data, and creating and generating fifth data corresponding to the fourth data, further includes: Construct and generate a set of directions containing at least one spatial direction and a set of step sizes containing multiple discrete spatial step sizes, respectively; Based on each current step size in the step size set, create corresponding valid voxel pairs in each direction in the direction set, which are located within the second data. The eighth data corresponding to the current step size is calculated based on the sum of the squared gray-level differences of all effective voxel pairs and the total number of effective voxel pairs; wherein, the eighth data is a variation function value; Iterate through all the step distances in the step distance set, collect the eighth data corresponding to each step distance, and construct the corresponding fifth data.

[0009] Furthermore, the curve morphology parameters are selected from one or more of the following parameter groups: The parameters include a slope parameter reflecting the rate of small-scale variation, a sill value parameter reflecting the overall level of variation, a range parameter reflecting the spatially correlated scale, a nugget value parameter reflecting micro-random variation, and an area under the curve parameter within a predetermined scale range. Among these, the mapping from voxel data to physical space mainly maps to R. When the voxel spacing of the three-dimensional medical image volume data is inconsistent in different directions, isotropic resampling is performed on the three-dimensional medical image volume data. Alternatively, the dispersion distance is converted into physical distance based on the voxel spacing, and the variogram results are statistically analyzed or fitted according to the physical distance.

[0010] Furthermore, the step of creating, generating, and outputting corresponding seventh data based on the sixth data also includes: Generate and acquire corresponding sixth data, and normalize and map the sixth data respectively; A ninth data point corresponding to the sixth data point is created and weighted and fused to generate a tenth data point corresponding to the seventh data point; wherein the ninth data point consists of multiple parameters after normalization; and the tenth data point is a comprehensive texture score.

[0011] Furthermore, the method also includes: Based on at least one of the number of valid data points in the variogram value sequence, the fitting residual, or the goodness of fit, a quality control label is constructed, generated, and output to identify the confidence level of the current evaluation result; wherein, the quality control label includes at least one of the following: pass level, warning level, or invalid level; When the number of valid data points is lower than a preset threshold, the fitting residual is higher than a preset threshold, or the goodness of fit is lower than a preset threshold, a corresponding warning level or invalid level is created and output.

[0012] Furthermore, the target tissue is respiratory tissue; the three-dimensional volume of interest is located in the lung parenchyma or the lesion area within the lung; Alternatively, the target tissue is bone tissue; the three-dimensional volume of interest is located inside the cancellous bone of the vertebral body.

[0013] Furthermore, the three-dimensional volume of interest is generated by removing at least one non-target structure from the cortical bone, endplate edge, or venous groove region.

[0014] The second objective of this invention is achieved as follows: the system is used to implement the aforementioned method for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images, the system comprising: A generation unit is created to generate and acquire first data corresponding to the target tissue, and based on the first data, to create and generate second data corresponding to the target area of ​​the first data; wherein, the first data is three-dimensional medical image volume data, preferably three-dimensional CT volume data; the second data is three-dimensional volume of interest, and in this scheme, the second data is specifically three-dimensional volume of interest (VOI) data; A generation unit is constructed to generate fourth data corresponding to the second data, and to create fifth data corresponding to the fourth data; wherein the fourth data is a three-dimensional grayscale variation function; and the fifth data is a sequence of variation function values. The processing and generation unit is used to fit the fifth data in the double logarithmic domain and extract and generate at least one corresponding sixth data that characterizes the spatial variation law of tissue texture; wherein, the sixth data is a curve morphology parameter; An output unit is created to generate and output corresponding seventh data based on the sixth data; wherein the seventh data is the evaluation result data.

[0015] Furthermore, the construction generation unit also includes: The first generation module is used to construct and generate a set of directions containing at least one spatial direction and a set of step sizes containing multiple discrete spatial step sizes, respectively. The second generation module is used to create corresponding valid voxel pairs located within the second data in each direction of the direction set, based on each current step distance in the step distance set. The third generation module is used to calculate and generate the eighth data corresponding to the current step size based on the sum of the squared gray-level differences of all effective voxel pairs and the total number of effective voxel pairs; wherein, the eighth data is a variation function value; The first processing module is used to traverse and process all step distances in the step distance set, collect the eighth data corresponding to each step distance, and construct and generate the corresponding fifth data.

[0016] The curve morphology parameters are selected from one or more of the following parameter groups: slope parameters reflecting small-scale variation rate, sill value parameters reflecting total variation level, range parameters reflecting spatial correlation scale, nugget value parameters reflecting micro-random variation, and area under the curve parameters within a predetermined scale range; wherein, the mapping of voxel data to physical space mainly maps R. When the voxel spacing of the three-dimensional medical image volume data is inconsistent in different directions, isotropic resampling is performed on the three-dimensional medical image volume data; or the dispersion distance is converted into physical distance based on the voxel spacing, and the variogram results are statistically analyzed or fitted according to the physical distance.

[0017] And / or, the creation output unit further includes: The second processing module is used to generate and acquire multiple sixth data, and to normalize and map the sixth data respectively; The third processing module is used to create and generate the ninth data corresponding to the sixth data, and to perform weighted fusion processing on the ninth data to generate the tenth data corresponding to the seventh data; wherein, the ninth data consists of multiple parameters after normalization processing; and the tenth data is a comprehensive texture score.

[0018] And / or, the system further includes: The fourth generation module is used to construct, generate, and output a quality control label that identifies the credibility of the current evaluation result based on at least one of the number of valid data points in the variogram value sequence, the fitting residual, or the goodness of fit; wherein the quality control label includes at least one of the following: pass level, warning level, or invalid level. The fifth generation module is used to create and output the corresponding warning level or invalid level when the number of valid data points is lower than the preset threshold, the fitting residual is higher than the preset threshold, or the goodness of fit is lower than the preset threshold.

[0019] The target tissue is respiratory tissue, and the three-dimensional volume of interest is located in the lung parenchyma or the lesion area within the lung. Alternatively, the target tissue may be skeletal tissue, and the three-dimensional volume of interest may be located within the cancellous bone of the vertebral body.

[0020] Furthermore, the three-dimensional volume of interest is generated by removing at least one non-target structure from the cortical bone, endplate edge, or venous groove region.

[0021] The third objective of this invention is achieved as follows: It includes a processor, a memory, and a control program for a respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images; wherein, the processor executes the control program for the respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images, and the control program is stored in the memory; the control program for the respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images implements the respiratory or skeletal tissue morphology assessment method based on a three-dimensional grayscale variation function of medical images.

[0022] The fourth objective of this invention is achieved as follows: the computer-readable storage medium stores a control program for a respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images, and the control program for the respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images implements the respiratory or skeletal tissue morphology assessment method based on a three-dimensional grayscale variation function of medical images.

[0023] This invention generates and acquires first data corresponding to a target tissue, and based on the first data, creates and generates second data corresponding to the target area of ​​the first data; wherein the first data is three-dimensional medical image volume data, preferably three-dimensional CT volume data; the second data is a three-dimensional volume of interest; then, it constructs and generates fourth data corresponding to the second data, and creates and generates fifth data corresponding to the fourth data; wherein the fourth data is a three-dimensional grayscale variation function; the fifth data is a sequence of variation function values; in the double logarithmic domain, it fits and processes the fifth data, and extracts and generates at least one corresponding sixth data representing the spatial variation law of tissue texture; wherein the sixth data is a curve morphology parameter; finally, based on the sixth data, it creates, generates and outputs corresponding seventh data; wherein the seventh data is evaluation result data; and the corresponding system, platform and storage medium, which can solve the technical defects of the prior art such as insufficient utilization of three-dimensional structural information, unclear texture spatial scale representation, strong directional sensitivity, high dependence on fine segmentation, and lack of credibility judgment of batch processing results.

[0024] In other words, the present invention constructs a three-dimensional grayscale variation function to perform multi-directional and multi-scale spatial variation analysis on respiratory or skeletal tissues, and extracts curve morphology parameters with clear physical meaning in the double logarithmic domain. This enables stable and repeatable quantitative evaluation of the three-dimensional texture features of tissues without the need for fine segmentation of the tissue structure, and effectively solves the technical problems of missing structural information, unclear scale representation, strong directional sensitivity, and high dependence on fine segmentation in two-dimensional texture analysis. At the same time, the present invention also evaluates the number of fitting points, fitting residuals, and goodness of fit through built-in quality control rules and outputs a confidence label, thereby improving the engineering reliability and clinical usability of batch processing results. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Figure 1 This is a schematic diagram of the process steps of a method for morphological evaluation of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images according to the present invention. Figure 2 This is a schematic diagram of the process of evaluating the texture of CT images based on the three-dimensional grayscale variation function of medical images, which can be used for the evaluation of respiratory or skeletal tissue morphology. Figure 3This is a schematic diagram of the VOI (Vascular Occlusion) of an embodiment of the present invention, which is a method for morphological assessment of respiratory or skeletal tissue based on a three-dimensional grayscale variation function of medical images. Figure (a) is a schematic diagram of an axial cross-section of a lung CT image; Figure (b) shows the VOI of the lung region; Figure (c) is a schematic diagram of an axial cross-section of a lumbar spine CT image; and Figure (d) shows the VOI of the lumbar cancellous bone region. The VOI is obtained through automatic segmentation, semi-automatic segmentation, or manual assistance. Figure 4 This is a schematic diagram of a three-dimensional direction set of an embodiment of the present invention, which is a method for morphological assessment of respiratory or skeletal tissue based on a three-dimensional grayscale variation function of medical images; wherein, Figure (a) represents a 6-neighborhood, representing 6 axial directions; Figure (b) represents an 18-neighborhood, representing 6 axial directions and diagonal directions; Figure (c) represents a 26-neighborhood, representing diagonal directions in addition to the 18-neighborhood. Figure 5 This is a schematic diagram of the step size set of an embodiment of the present invention, which is a method for evaluating the morphology of respiratory or skeletal tissues based on the three-dimensional grayscale variation function of medical images; wherein, the step size of the pixel point is shown in the figure as 1. Figure 6 This is a schematic diagram of a log-log curve representing an embodiment of the present invention of a method for morphological evaluation of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images; Figure 7 This is a schematic diagram of a scoring report for an embodiment of the present invention, which is a method for evaluating the morphology of respiratory or skeletal tissues based on the three-dimensional grayscale variation function of medical images. Figure 8 This is a schematic diagram of the framework of a respiratory or skeletal tissue morphology assessment system based on a three-dimensional grayscale variation function of medical images according to the present invention. Figure 9 This is a schematic diagram of the architecture of a respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images according to the present invention. Figure 10 This is a schematic diagram of a computer-readable storage medium architecture in an embodiment of a method for evaluating the morphology of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images according to the present invention. Detailed Implementation

[0026] To facilitate a clearer understanding of the objectives, technical solutions, and advantages of this invention, the invention will be further described below in conjunction with the accompanying drawings and specific embodiments. Those skilled in the art can easily understand other advantages and effects of this invention from the content disclosed in this specification.

[0027] This invention can also be implemented or applied through other different specific examples, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of this invention.

[0028] It should be noted that if the embodiments of the present invention involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of the components in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0029] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Secondly, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0030] Preferably, the morphological assessment method for respiratory or skeletal tissues based on the three-dimensional grayscale variation function of medical images of the present invention is applied in one or more terminals or servers. The terminal is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0031] The terminal can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal can interact with the customer via a keyboard, mouse, remote control, touchpad, or voice control device.

[0032] In this embodiment, the method for evaluating respiratory or skeletal tissue morphology based on the three-dimensional grayscale variation function of medical images can be applied to terminals or fixed terminals with display functions. The terminals are not limited to personal computers, smartphones, tablets, desktop computers or all-in-one computers with cameras, etc.

[0033] The method for assessing respiratory or skeletal tissue morphology based on the three-dimensional grayscale variation function of medical images can also be applied to a hardware environment consisting of a terminal and a server connected to the terminal via a network. The network includes, but is not limited to, a wide area network (WAN), a metropolitan area network (MAN), or a local area network (LAN). The method for assessing respiratory or skeletal tissue morphology based on the three-dimensional grayscale variation function of medical images in this embodiment can be executed by the server, by the terminal, or by both the server and the terminal.

[0034] For example, for terminals requiring respiratory or skeletal tissue morphological assessment based on the three-dimensional grayscale variation function of medical images, the respiratory or skeletal tissue morphological assessment function based on the three-dimensional grayscale variation function of medical images provided by the method of this invention can be directly integrated into the terminal, or a client for implementing the method of this invention can be installed. Furthermore, the method provided by this invention can also run on servers or other devices in the form of a Software Development Kit (SDK), providing an interface for the respiratory or skeletal tissue morphological assessment function based on the three-dimensional grayscale variation function of medical images in the form of an SDK. Terminals or other devices can then implement the respiratory or skeletal tissue morphological assessment function based on the three-dimensional grayscale variation function of medical images through the provided interface. The invention will be further described below with reference to the accompanying drawings.

[0035] like Figures 1-7 As shown, this invention provides a method for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images. The method includes the following steps: S101. Generate and acquire first data corresponding to the target tissue, and based on the first data, create and generate second data corresponding to the target area of ​​the first data; wherein, the first data is three-dimensional medical image volume data, preferably three-dimensional CT volume data; the second data is a three-dimensional volume of interest; S102. Construct and generate fourth data corresponding to the second data, and create and generate fifth data corresponding to the fourth data; wherein, the fourth data is a three-dimensional grayscale variation function; and the fifth data is a sequence of variation function values; S103. In the double logarithmic domain, the fifth data is fitted and processed, and at least one sixth data corresponding to it is extracted to represent the spatial variation law of tissue texture; wherein, the sixth data is a curve morphology parameter; S104. Based on the sixth data, create, generate, and output the corresponding seventh data; wherein, the seventh data is the evaluation result data.

[0036] The process of constructing and generating fourth data corresponding to the second data, and creating and generating fifth data corresponding to the fourth data, further includes: S1021. Construct and generate a set of directions containing at least one spatial direction and a set of step sizes containing multiple discrete spatial step sizes, respectively. S1022. Based on each current step distance in the step distance set, create a corresponding valid voxel pair located within the second data in each direction in the direction set; S1023. Based on the sum of the squared grayscale differences of all effective voxel pairs and the total number of effective voxel pairs, calculate and generate the eighth data corresponding to the current step size; wherein, the eighth data is a variation function value; S1024. Traverse and process all step distances in the step distance set, collect the eighth data corresponding to each step distance, and construct the corresponding fifth data.

[0037] The curve shape parameters are selected from one or more of the following parameter groups: The parameters include a slope parameter reflecting the rate of small-scale variation, a sill value parameter reflecting the overall level of variation, a range parameter reflecting the spatially correlated scale, a nugget value parameter reflecting micro-random variation, and an area under the curve parameter within a predetermined scale range. Among these, the mapping from voxel data to physical space mainly maps to R. When the voxel spacing of the three-dimensional medical image volume data is inconsistent in different directions, isotropic resampling is performed on the three-dimensional medical image volume data. Alternatively, the dispersion distance is converted into physical distance based on the voxel spacing, and the variogram results are statistically analyzed or fitted according to the physical distance.

[0038] The process of creating, generating, and outputting corresponding seventh data based on the sixth data also includes: S1041. Generate and obtain a plurality of corresponding sixth data, and normalize and map the sixth data respectively; S1042. Create and generate a ninth data corresponding to the sixth data, and perform weighted fusion processing on the ninth data to generate a tenth data corresponding to the seventh data; wherein, the ninth data consists of multiple parameters after normalization processing; and the tenth data is a comprehensive texture score.

[0039] The method further includes: S1051. Based on at least one of the number of valid data points in the variogram value sequence, the fitting residual, or the goodness of fit, construct, generate, and output a quality control label that identifies the credibility of the current evaluation result; wherein, the quality control label includes at least one of a pass level, a warning level, or an invalid level; S1052. When the number of valid data points is lower than the preset threshold, the fitting residual is higher than the preset threshold, or the goodness of fit is lower than the preset threshold, create, generate, and output the corresponding warning level or invalid level.

[0040] The target tissue is respiratory tissue; the three-dimensional volume of interest is located in the lung parenchyma or a lesion area within the lung. Alternatively, the target tissue is bone tissue; the three-dimensional volume of interest is located inside the cancellous bone of the vertebral body.

[0041] The three-dimensional volume of interest is generated by removing at least one non-target structure from the cortical bone, endplate margin, or venous groove region.

[0042] Specifically, in this embodiment of the invention, a texture evaluation method based on a three-dimensional CT grayscale variation function is provided. This method constructs a three-dimensional variation function curve by statistically analyzing grayscale differences at different spatial step sizes within the target tissue's VOI (Voice of Intent). The curve's morphological parameters are extracted in the logarithmic domain, and quality control and scoring are performed to achieve quantitative evaluation of the three-dimensional texture of respiratory or skeletal tissues. The core technical problems this invention aims to solve include insufficient utilization of three-dimensional structural information, unclear texture spatial scale representation, strong directional sensitivity, high dependence on fine segmentation, and lack of reliability assessment for batch processing results.

[0043] In other words, this invention provides a method for evaluating the texture of a medical image (e.g., CT image) using grayscale variation function. The method takes three-dimensional CT volume data as input, automatically, semi-automatically, or manually assisted in generating a volume of interest (VOI) within the target tissue region. It performs multi-directional, multi-step spatial statistics on the voxel grayscale values ​​within the VOI, calculates the three-dimensional grayscale variation function V(r), and performs curve fitting, parameter extraction, normalized score fusion, and quality control in the log-log domain. Finally, it outputs a comprehensive texture score, fitting parameters, anisotropic auxiliary indicators, a curve, and quality control markers.

[0044] The target tissues may include respiratory tissues and skeletal tissues. For example, respiratory tissues may be the lung parenchyma, areas related to local lesions in the lungs, or other target lung areas; skeletal tissues may be the cancellous bone areas of the vertebral bodies, the internal areas of the thoracic and lumbar vertebral bodies, or other bone tissue areas. The method of this invention does not require segmentation of microstructures such as alveoli, small blood vessels, bronchioles, and trabeculae; instead, it directly performs three-dimensional texture statistics based on conventional CT grayscale data. The overall flowchart of the specific evaluation method is shown below. Figure 2 As shown.

[0045] For the input and preprocessing of 3D CT volumetric data, the 3D CT volumetric data of the target anatomical region is input, and information such as voxel spacing, slice thickness, reconstruction kernel, and scan area is read. 3D CT grayscale volumetric data can be represented as: (1) In the formula, Ω represents the three-dimensional volumetric data space, and I(p) represents the CT gray value or HU value at voxel point p. Preprocessing may include operations such as gray-level normalization, noise suppression, voxel homogenization, resampling, outlier clipping, and coordinate orientation unification.

[0046] For target tissue region identification and VOI generation, the target tissue region is identified in the 3D CT volumetric data, and a volume of interest (VOI) is generated within the target region for texture calculation. VOIs can be obtained through automatic segmentation, semi-automatic segmentation, regular geometry generation, manual assisted bounding box selection, or localization based on anatomical landmarks.

[0047] (2) When evaluating respiratory tissue, the VOI (Void of Infection) can be located within the lung parenchyma or in areas related to localized lung lesions; when evaluating bone tissue, the VOI can be located within the vertebral cancellous bone or other target bone tissue. To improve stability, the influence of non-target structures on bone tissue VOIs can be reduced through methods such as boundary indentation, morphological erosion, endplate / cortical bone exclusion, and venous groove region exclusion. Specifically, for example... Figure 3 As shown.

[0048] In the definition of the direction set and step size set, in order to describe the gray-scale variation characteristics of the target tissue at different spatial scales, a three-dimensional direction set D and a step size set R are defined. The direction set D is used to define the spatial direction of voxel pairing, and the step size set R is used to define the discrete distance between voxels.

[0049] (3) (4) (5) In the formula, D6 corresponds to a 6-neighborhood, which includes only the axial direction; D 18 Add the diagonal direction to the axial direction; D 26 Further adding diagonal directions can cover all neighborhood directions around a 3D voxel, such as... Figure 4 As shown.

[0050] The step size set can be set as R = {1, 2, ..., r} max The unit is voxels. The distance step can also be converted to physical distance h(r) by combining the voxel spacing, for example, h(r) = r·s, where s represents the voxel physical spacing or equivalent spatial step size in the corresponding direction. Figure 5 As shown.

[0051] In the calculation of the 3D grayscale variation function, for any step distance r and direction d, effective voxel pairs are generated within the VOI. If both the central voxel point p and the displaced voxel point p+r·d are located inside the VOI, then this combination forms an effective voxel pair. The set of effective voxel pairs can be represented as: (6) The three-dimensional gray-level variation function value at step size r is obtained by summing and averaging the squared gray-level differences between all directions and all effective voxel pairs. Preferably, the following calculation formula can be used: (7) This yields the discrete sequence: (8) The core computational idea is: if the grayscale difference is large at a certain spatial distance, then V(r) is large, indicating stronger texture change at that scale; if the grayscale difference is small, then V(r) is small, indicating weaker texture change at that scale. By traversing multiple step sizes, the discrete variation function sequence {V(r)} can be obtained.

[0052] To construct and fit the log-log curve, in order to extract the overall pattern of the variation function with spatial scale, a logarithmic transformation is performed on the horizontal axis distance and the vertical axis variogram values ​​to construct the log-log curve: (9) In the formula, h(r) can be the offset distance r, or the physical distance calculated based on the voxel spacing. When h(r)≤0 or V(r)≤0, the corresponding data points are discarded as invalid points. This yields the point set {(x(r), y(r))} used for fitting.

[0053] This invention fits the variation function curve in log-log coordinates to extract stable curve shape parameters, which are used to characterize the spatial variation law and scale characteristics of the target tissue texture. The curve after log transformation is shown below. Figure 6 As shown.

[0054] The curve is fitted using a pre-defined fitting model within the log-log domain. The fitting model can be a linear model, a piecewise linear model, an exponentially approaching model, a saturation model, or any other model that can describe the curve's growth and plateau characteristics. After fitting, the curve's morphological parameters are extracted to characterize the spatial variation of the target tissue texture. The fitted curve, such as... Figure 7 As shown.

[0055] For texture parameter extraction, one or more texture parameters can be extracted based on the fitted curve and the discrete variogram sequence. Preferred parameters include: ; In quality control and anomaly handling, to improve result stability and engineering usability, this invention establishes quality control rules. These rules may include: When V(r) shows a non-positive value, a significant abnormal jump, or an abnormal decline with increasing distance, the fitting interval is truncated or outliers are removed; 1) When the number of effective fitted points after log-log transformation is less than the preset minimum number of points, the result is deemed invalid and an invalid flag is output; 2) When the fitting residual is too large, the goodness of fit R² is lower than the preset threshold, or the parameters exceed the reasonable range, a warning flag is output and a review is prompted; 3) Record the number of effective voxel pairs, the number of effective scale points, the fitting residual, the goodness of fit, and the final quality control status for each case; 4) The quality control results can be divided into "pass," "warning," and "invalid" levels, which facilitates batch processing, clinical review, and subsequent algorithm iteration.

[0056] For the normalized and scored fusion output 3D-TBS, since the dimensions and numerical ranges of the fitted parameters are different, in order to avoid scale imbalance caused by direct addition, this invention preferably performs normalization mapping on the parameters involved in the scoring first: (10) Five combinations of parameters are preferred: (11) Finally, the result was obtained by equal-weight fusion: (12) The comprehensive texture score can be output as a three-dimensional texture evaluation result of the target organization.

[0057] When the evaluation object is respiratory tissue, the score can be output as a lung tissue texture score; when the evaluation object is skeletal tissue, the score can be output as a bone tissue texture score.

[0058] It also outputs the corresponding fitting parameters, goodness of fit, and quality control markers.

[0059] In this embodiment of the solution, the preferred method flow includes the following steps: Input the three-dimensional CT volume data of the target anatomical region and read the voxel spacing information; The three-dimensional CT volume data is preprocessed, including but not limited to grayscale normalization, noise suppression, and voxel unification; further, the preprocessing includes but is not limited to CT value correction, outlier clipping, noise suppression, resampling, or grayscale range limitation. Identify the target region of respiratory or skeletal tissue, and select the volume of interest (VOI) within the target region; Calculate the three-dimensional variation function curve V(r) within the VOI; The curve is fitted in the log-log domain to obtain parameters such as slope, sill, range, nugget, and area, and a comprehensive texture score is calculated. Output quality control markers according to quality control rules, and output comprehensive texture score and related parameters.

[0060] Example 1: Texture Evaluation Based on 3D CT Gray-Scale Variation Function of Respiratory Tissue This embodiment illustrates the application of the present invention in respiratory tissues. First, three-dimensional volumetric data of the subject's chest CT scan is acquired, voxel spacing information is read, and the original CT volumetric data undergoes grayscale normalization, noise suppression, and voxel homogenization processing. Subsequently, VOIs are generated in the lung parenchyma region or in areas related to local lesions within the lung. VOIs can be automatically generated within the lung region obtained through a deep learning segmentation model, or they can be obtained through manual point selection, local region expansion, fixed volume cropping, or anatomical landmark localization.

[0061] After obtaining the VOI, a set of directions D and a set of step distances R are defined. For each direction and step distance, effective voxel pairs are generated, and the corresponding three-dimensional gray-level variation function V(r) is calculated. The V(r) sequence can describe the variation of gray level within lung tissue with spatial distance, thereby reflecting the spatial heterogeneity of lung texture at different scales.

[0062] Subsequently, a log-log transformation was performed on V(r) and a curve was fitted to extract parameters such as slope, sill, range, nugget, and area. These parameters were then normalized and weighted to obtain a three-dimensional texture score for lung tissue. The system simultaneously outputs goodness-of-fit, residual analysis results, anisotropy auxiliary indicators, and quality control markers.

[0063] In some implementations, the set of directions can be divided into different directional subsets, and the variation function curves and fitting parameters in each direction can be calculated. Then, directional texture indices can be constructed based on the differences, ratios, standard deviations, or combination functions between the parameters in different directions to help characterize the directional non-uniformity of lung texture.

[0064] Example 2: Texture Evaluation Based on 3D CT Gray-Level Variation Function of Bone Tissue This embodiment illustrates the application of the present invention in bone tissue. First, spinal CT, thoracic and lumbar CT, or three-dimensional CT volume data containing the target bone region is acquired, and the corresponding voxel spacing information is read. Subsequently, the CT volume data is preprocessed, and a VOI (voxel-in-place) is constructed within the target bone tissue. Preferably, the VOI is located in the cancellous bone region within the vertebral body.

[0065] In a preferred embodiment, a mask of the target vertebral body or bone tissue region is first obtained through a deep learning segmentation model, threshold segmentation, anatomical template localization, or manual coarse segmentation. Then, based on boundary distance constraints, morphological erosion, fixed ratio shrinkage, or anatomical level limitations, cortical bone, endplate edges, venous groove regions, and other regions unsuitable for stability analysis are removed, leaving only the relatively stable cancellous bone VOI.

[0066] Multi-directional voxel pairs are generated within the VOI using 18- or 26-directional sets, and V(r) is calculated at multiple step sizes. This variogram sequence is used to characterize the degree of grayscale variation of bone tissue at different spatial scales, reflecting the spatial heterogeneity, scale variation, and directional characteristics of the internal texture structure of bone tissue.

[0067] After fitting the variogram curve to log-log, parameters such as slope, sill, range, nugget, and area are extracted and then normalized and fused to obtain a three-dimensional texture score of bone tissue. For cancellous bone regions, parameters can be calculated separately in the vertical, horizontal, anterior-posterior, or oblique directions to construct anisotropy indices for bone tissue, which can help characterize the differences in bone texture in different directions.

[0068] To achieve the above objectives, such as Figure 8 As shown, the present invention also provides a respiratory or skeletal tissue morphological assessment system based on a three-dimensional grayscale variation function of medical images. The system is used to implement the aforementioned respiratory or skeletal tissue morphological assessment method based on a three-dimensional grayscale variation function of medical images. The system includes: A generation unit is constructed to generate fourth data corresponding to the second data, and to create fifth data corresponding to the fourth data; wherein the fourth data is a three-dimensional grayscale variation function; and the fifth data is a sequence of variation function values. The processing and generation unit is used to fit the fifth data in the double logarithmic domain and extract and generate at least one corresponding sixth data that characterizes the spatial variation law of tissue texture; wherein the sixth data is a curve morphology parameter; An output unit is created to generate and output corresponding seventh data based on the sixth data; wherein the seventh data is the evaluation result data.

[0069] And / or, the construction generation unit further includes: The first generation module is used to construct and generate a set of directions containing at least one spatial direction and a set of step sizes containing multiple discrete spatial step sizes, respectively. The second generation module is used to create corresponding valid voxel pairs located within the second data in each direction of the direction set, based on each current step distance in the step distance set. The third generation module is used to calculate and generate the eighth data corresponding to the current step size based on the sum of the squared gray-level differences of all effective voxel pairs and the total number of effective voxel pairs; wherein, the eighth data is a variation function value; The first processing module is used to traverse and process all step distances in the step distance set, collect the eighth data corresponding to each step distance, and construct and generate the corresponding fifth data.

[0070] The curve morphology parameters are selected from one or more of the following parameter groups: slope parameters reflecting small-scale variation rate, sill value parameters reflecting total variation level, range parameters reflecting spatial correlation scale, nugget value parameters reflecting micro-random variation, and area under the curve parameters within a predetermined scale range; wherein, the mapping of voxel data to physical space mainly maps R. When the voxel spacing of the three-dimensional medical image volume data is inconsistent in different directions, isotropic resampling is performed on the three-dimensional medical image volume data; or the dispersion distance is converted into physical distance based on the voxel spacing, and the variogram results are statistically analyzed or fitted according to the physical distance.

[0071] And / or, the creation output unit further includes: The second processing module is used to generate and acquire multiple sixth data, and to normalize and map the sixth data respectively; The third processing module is used to create and generate the ninth data corresponding to the sixth data, and to perform weighted fusion processing on the ninth data to generate the tenth data corresponding to the seventh data; wherein, the ninth data consists of multiple parameters after normalization processing; and the tenth data is a comprehensive texture score.

[0072] And / or, the system further includes: The fourth generation module is used to construct, generate, and output a quality control label that identifies the credibility of the current evaluation result based on at least one of the number of valid data points in the variogram value sequence, the fitting residual, or the goodness of fit; wherein the quality control label includes at least one of the following: pass level, warning level, or invalid level. The fifth generation module is used to create and output the corresponding warning level or invalid level when the number of valid data points is lower than the preset threshold, the fitting residual is higher than the preset threshold, or the goodness of fit is lower than the preset threshold.

[0073] The target tissue is respiratory tissue, and the three-dimensional volume of interest is located in the lung parenchyma or the lesion area within the lung. Alternatively, the target tissue may be skeletal tissue, and the three-dimensional volume of interest may be located within the cancellous bone of the vertebral body.

[0074] Furthermore, the three-dimensional volume of interest is generated by removing at least one non-target structure from the cortical bone, endplate edge, or venous groove region.

[0075] In the system embodiment of the present invention, the specific details of the method steps involved in the respiratory or skeletal tissue morphology assessment system based on the three-dimensional grayscale variation function of medical images have been described above. That is to say, the steps in the system are applied to the functional units or functional modules in the above method embodiment, and will not be repeated here.

[0076] To achieve the above objectives, the present invention also provides a platform for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images, such as... Figure 9 The system includes a processor, a memory, and a control program for a respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images. The processor executes the control program, which is stored in the memory. This control program implements the steps of the respiratory or skeletal tissue morphology assessment method based on the three-dimensional grayscale variation function of medical images, for example: S101. Generate and acquire first data corresponding to the target tissue, and based on the first data, create and generate second data corresponding to the target area of ​​the first data; wherein, the first data is three-dimensional medical image volume data, preferably three-dimensional CT volume data; the second data is a three-dimensional volume of interest; S102. Construct and generate fourth data corresponding to the second data, and create and generate fifth data corresponding to the fourth data; wherein, the fourth data is a three-dimensional grayscale variation function; and the fifth data is a sequence of variation function values; S103. In the double logarithmic domain, the fifth data is fitted and processed, and at least one sixth data corresponding to it is extracted to represent the spatial variation law of tissue texture; wherein, the sixth data is a curve morphology parameter; S104. Based on the sixth data, create, generate, and output the corresponding seventh data; wherein, the seventh data is the evaluation result data.

[0077] The specific details of the steps have been explained above and will not be repeated here.

[0078] In this embodiment of the invention, the built-in processor of the respiratory or skeletal tissue morphology assessment platform based on the three-dimensional grayscale variation function of medical images can be composed of integrated circuits. For example, it can be composed of a single packaged integrated circuit, or it can be composed of multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor connects to various components through various interfaces and lines, and executes programs or units stored in memory, and calls data stored in memory to perform various functions of respiratory or skeletal tissue morphology assessment based on the three-dimensional grayscale variation function of medical images and process data. The memory is used to store program code and various data. It is installed in a respiratory or skeletal tissue morphology evaluation platform based on the three-dimensional grayscale variation function of medical images and enables high-speed and automatic access to programs or data during operation. The memory includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0079] To achieve the above objectives, the present invention also provides a computer-readable storage medium, such as... Figure 10 As shown, the computer-readable storage medium stores a control program for a respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images. This control program implements the steps of the respiratory or skeletal tissue morphology assessment method based on a three-dimensional grayscale variation function of medical images; for example: S101. Generate and acquire first data corresponding to the target tissue, and based on the first data, create and generate second data corresponding to the target area of ​​the first data; wherein, the first data is three-dimensional medical image volume data; the second data is a three-dimensional volume of interest; S102. Construct and generate fourth data corresponding to the second data, and create and generate fifth data corresponding to the fourth data; wherein, the fourth data is a three-dimensional grayscale variation function; and the fifth data is a sequence of variation function values; S103. In the double logarithmic domain, the fifth data is fitted and processed, and at least one sixth data corresponding to it is extracted to represent the spatial variation law of tissue texture; wherein, the sixth data is a curve morphology parameter; S104. Based on the sixth data, create, generate, and output the corresponding seventh data; wherein, the seventh data is the evaluation result data.

[0080] The specific details of the steps have been explained above and will not be repeated here.

[0081] In the description of embodiments of the present invention, it should be noted that any process or method description in the flowcharts or otherwise described herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the present invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order according to the functions involved, as should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0082] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processing module, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, a “computer-readable medium” can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, the computer-readable medium can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0083] This invention generates and acquires first data corresponding to a target tissue, and based on the first data, creates and generates second data corresponding to the target region of the first data; wherein the first data is three-dimensional medical image volume data; the second data is a three-dimensional volume of interest; then, it constructs and generates fourth data corresponding to the second data, and creates and generates fifth data corresponding to the fourth data; wherein the fourth data is a three-dimensional grayscale variation function; the fifth data is a sequence of variation function values; in the double logarithmic domain, it fits and processes the fifth data, and extracts and generates at least one corresponding sixth data representing the spatial variation law of tissue texture; wherein the sixth data is a curve morphology parameter; finally, based on the sixth data, it creates, generates and outputs corresponding seventh data; wherein the seventh data is evaluation result data; and the corresponding system, platform and storage medium, which can solve the technical defects of the prior art such as insufficient utilization of three-dimensional structural information, unclear texture spatial scale representation, strong directional sensitivity, high dependence on fine segmentation, and lack of credibility judgment of batch processing results.

[0084] In other words, the present invention constructs a three-dimensional grayscale variation function to perform multi-directional and multi-scale spatial variation analysis on respiratory or skeletal tissues, and extracts curve morphology parameters with clear physical meaning in the double logarithmic domain. It can stably and repeatedly quantify and evaluate the three-dimensional texture features of tissues without the need for fine segmentation of tissue structures, and effectively solves the technical problems of missing structural information, unclear scale representation, strong directional sensitivity, and high dependence on fine segmentation in two-dimensional texture analysis. At the same time, the present invention also evaluates the number of fitting points, fitting residuals, and goodness of fit through built-in quality control rules and outputs a confidence label, which improves the engineering reliability and clinical usability of batch processing results and can be uniformly applied to the morphological quantitative evaluation of respiratory and skeletal tissues.

[0085] In other words, the technical solution provided by this invention constructs a three-dimensional grayscale variation function within the VOI based on three-dimensional CT volume data. Compared with two-dimensional texture methods, it can make fuller use of the three-dimensional spatial information of tissues and is more conducive to characterizing the real tissue structure features. Through multi-step grayscale difference statistics and log-log curve fitting, parameters such as slope, sill, range, nugget, and area are extracted, giving the evaluation results clear spatial scale meaning and better interpretability. Furthermore, the use of multi-directional voxel pairing statistics can reduce the impact of local noise, abnormal voxels, and single-directional sampling on the results, improving stability and repeatability. Moreover, it does not rely on fine segmentation of the internal microstructure of tissues and can directly process routine clinical CT volume data, reducing the threshold for data acquisition and algorithm deployment. This invention can improve the reliability of batch processing by setting fitting interval truncation, outlier removal, effective point quantity determination, goodness-of-fit evaluation, and quality control markers. It can also be uniformly used for the texture evaluation of respiratory and skeletal tissues, facilitating the expansion of morphological evaluation of multiple tissues and indicators in a single CT examination.

[0086] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images, characterized in that, The method includes the following steps: First data corresponding to the target tissue is generated and acquired, and second data corresponding to the target area of ​​the first data is created based on the first data; wherein, the first data is three-dimensional medical image volume data; and the second data is a three-dimensional volume of interest. A fourth set of data corresponding to the second set of data is constructed, and a fifth set of data corresponding to the fourth set of data is created; wherein, the fourth set of data is a three-dimensional grayscale variation function; and the fifth set of data is a sequence of variation function values. The fifth data is fitted and processed, and at least one corresponding sixth data is extracted to represent the spatial variation law of tissue texture; wherein, the sixth data is a curve morphology parameter; Based on the sixth data, corresponding seventh data is created, generated, and output; wherein, the seventh data is the evaluation result data.

2. The method for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images according to claim 1, characterized in that, The process of constructing and generating fourth data corresponding to the second data, and creating and generating fifth data corresponding to the fourth data, further includes: Construct and generate a set of directions containing at least one spatial direction and a set of step sizes containing multiple discrete spatial step sizes, respectively; Based on each current step size in the step size set, create corresponding valid voxel pairs in each direction in the direction set, which are located within the second data. The eighth data corresponding to the current step size is calculated based on the sum of the squared gray-level differences of all effective voxel pairs and the total number of effective voxel pairs; wherein, the eighth data is a variation function value; Iterate through all the step distances in the step distance set, collect the eighth data corresponding to each step distance, and construct the corresponding fifth data.

3. A method for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images, as described in claim 1 or 2, characterized in that... The curve shape parameters are selected from one or more of the following parameter groups: The parameters include the slope parameter reflecting the small-scale variation rate, the sill value parameter reflecting the total variation level, the range parameter reflecting the spatial correlation scale, the nugget value parameter reflecting micro-random variation, and the area under the curve parameter within a predetermined scale range.

4. The method for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images according to claim 1, characterized in that, The process of creating, generating, and outputting corresponding seventh data based on the sixth data also includes: Generate and acquire corresponding sixth data, and normalize and map the sixth data respectively; A ninth data point corresponding to the sixth data point is created and weighted and fused to generate a tenth data point corresponding to the seventh data point; wherein, the ninth data point consists of multiple parameters after normalization; and the tenth data point is a comprehensive texture score.

5. A method for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images, as described in claim 1, 2, or 4, characterized in that... The method further includes: Based on at least one of the number of valid data points in the variogram value sequence, the fitting residual, or the goodness of fit, a quality control label is constructed, generated, and output to identify the confidence level of the current evaluation result; wherein, the quality control label includes at least one of a pass level, a warning level, or an invalid level; When the number of valid data points is lower than a preset threshold, the fitting residual is higher than a preset threshold, or the goodness of fit is lower than a preset threshold, a corresponding warning level or invalid level is created and output.

6. The method for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images according to claim 1, characterized in that, The target tissue is respiratory tissue; the three-dimensional volume of interest is located in the lung parenchyma or a lesion area within the lung. Alternatively, the target tissue is bone tissue; the three-dimensional volume of interest is located inside the cancellous bone of the vertebral body.

7. A method for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images, as described in claim 1 or 6, characterized in that... The three-dimensional volume of interest is generated by removing at least one non-target structure from the cortical bone, endplate margin, or venous groove region.

8. A system for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images, characterized in that, The system is used to implement the respiratory or skeletal tissue morphological assessment method based on the three-dimensional grayscale variation function of medical images as described in any one of claims 1 to 7, the system comprising: A generation unit is created to generate and acquire first data corresponding to the target tissue, and based on the first data, to create and generate second data corresponding to the target area of ​​the first data; wherein, the first data is three-dimensional medical image volume data; and the second data is a three-dimensional volume of interest. A generation unit is constructed to generate fourth data corresponding to the second data, and to create fifth data corresponding to the fourth data; wherein the fourth data is a three-dimensional grayscale variation function; and the fifth data is a sequence of variation function values. The processing and generation unit is used to fit and process the fifth data and extract and generate at least one corresponding sixth data that characterizes the spatial variation law of tissue texture; wherein, the sixth data is a curve morphology parameter; An output unit is created to generate and output corresponding seventh data based on the sixth data; wherein the seventh data is the evaluation result data.

9. A platform for morphological assessment of respiratory or skeletal tissues based on a three-dimensional grayscale variation function of medical images, characterized in that, The system includes a processor, a memory, and a control program for a respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images. The processor executes the control program, which is stored in the memory. The control program implements the respiratory or skeletal tissue morphology assessment method based on a three-dimensional grayscale variation function of medical images as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a control program for a respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images. The control program for the respiratory or skeletal tissue morphology assessment platform based on a three-dimensional grayscale variation function of medical images implements the respiratory or skeletal tissue morphology assessment method based on a three-dimensional grayscale variation function of medical images as described in any one of claims 1 to 7.

Images