Medical image acquisition and analysis method for endocrine patient

Abstract

The invention relates to an endocrine patient-oriented medical image acquisition and analysis method, in particular to the technical field of image acquisition and processing, and adopts the scheme that an image acquisition module, a storage module, an image processing and analysis module and a result generation module are included, and the image acquisition module is used for acquiring each area image in the total endocrine area of the current patient subjected to image acquisition in real time; the method comprises the following steps: identifying and extracting images of all areas with annularly closed edge textures, introducing the images into an image processing and analyzing module, sequentially calculating and outputting a titanium alloy texture degree TW, a titanium alloy structure degree C and a current fuzzy degree BH by utilizing the image processing and analyzing module, and determining the texture degree of the titanium alloy on the basis of the results of the titanium alloy texture degree TW, the titanium alloy structure degree C and the current fuzzy degree BH. According to the system and the method provided by the invention, more accurate and reliable image analysis support is provided, and auxiliary image acquisition, processing and analysis on an artifact area generated by an endocrine patient are facilitated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image acquisition processing, and particularly relates to a medical image acquisition and analysis method for endocrine patients. BACKGROUND

[0002] In the technical field of medical image acquisition processing, especially in the diagnosis and treatment process of endocrine patients with titanium alloy implants, the acquisition and analysis of medical images are of great importance. Due to the differences in individual conditions of endocrine patients, different degrees of inflammatory areas may occur between titanium alloy implants and body tissues. These inflammatory areas may form a certain degree of artifacts during the image acquisition process by nuclear magnetic resonance technology, thereby seriously interfering with the accuracy of medical images.

[0003] However, the existing medical image analysis method often fails to accurately extract the feature parameters of the endocrine region, especially when the artifacts overlap with the real tissue features, which may lead to inaccurate extraction of feature parameters, thereby affecting subsequent analysis and diagnosis. Moreover, the existing method lacks comprehensive analysis of the influence of artifacts, especially under the interference of artifacts, which may easily lead to one-sidedness of the diagnosis results.

[0004] Therefore, the present application provides a medical image acquisition and analysis method for endocrine patients to solve the problems raised in the background. SUMMARY

[0005] The present application solves the technical problem of providing a medical image acquisition and analysis method for endocrine patients to improve the accuracy of parameter extraction under the interference of artifacts and the processing and comprehensive analysis of artifacts.

[0006] To solve the above problems, the present application provides the following technical solutions: A medical image acquisition and analysis method for endocrine patients, comprising an image acquisition module, a storage module, an image processing and analysis module, and a result generation module, characterized in that the image processing and analysis module comprises a quantitative implant texture feature unit, a tissue structure difference degree unit, and an internal edge blur condition unit. The specific image acquisition and analysis steps are as follows: Step S1, using the image acquisition module, real-time acquisition of each regional image in the total endocrine region of the current patient being imaged; Step S2, based on the acquired regional images, identifying and extracting the images of all regions with ring-shaped closed edges, and introducing them into the image processing and analysis module; Step S3, using the image processing and analysis module, sequentially calculating the titanium alloy texture degree TW, the titanium alloy tissue degree C, and the current blur degree BH; Step S4, based on the results of the titanium alloy texture degree TW, the titanium alloy tissue degree C and the current blur degree BH, and using the result generation module, output the generated results and analysis; Step S5, the storage module will store all the results collected, processed and generated in steps S1-S4.

[0007] Further: the image acquisition module uses devices including magnetic resonance imaging devices, computed tomography scanners; The storage module uses devices including storage devices; The image processing and analysis module uses devices including image processing workstations; The result generation module uses devices including displays.

[0008] Further: the calculation formula of the quantitative implant texture feature unit is as follows: ; ; Among them: TW is the titanium alloy texture degree; N is the total number of pixels, N reflects the amount of all pixels in the current edge texture ring-like closed region; T i is the titanium alloy texture value of the i-th pixel, T i reflects the titanium alloy texture value of any pixel in the image edge texture ring-like closed region in the total endocrine region of the patient implanted with the implant; T avg is the titanium alloy texture average value, T avg reflects the average degree of the titanium alloy texture value T i of the i-th pixel in all pixels in the current edge texture ring-like closed region; BR is the current radius, BR reflects the radius of the current edge texture ring-like closed region; BR max is the maximum radius, BR max reflects the maximum value of the radius of the total endocrine region; For measuring the dispersion degree of the titanium alloy texture value of the pixel in the current edge texture ring-like closed region; The result range is between 0 and 1, and is used to adjust the value of .

[0009] Further: the calculation formula of the tissue structure difference degree unit is as follows: ; wherein: C is the titanium alloy texture degree; ZF diff is the texture orientation distribution difference degree, ZF diff reflects the difference degree of the texture orientation distribution in the current edge texture ring-closed region and the surrounding edge texture ring-closed region; ZF avg is the texture orientation distribution average difference degree, ZF avg reflects the average degree of the texture orientation distribution difference in the current edge texture ring-closed region and all surrounding edge texture ring-closed regions; TW0 is the titanium alloy texture initial degree, TW0 reflects the titanium alloy texture degree TW of the current edge texture ring-closed region in the initial state; TW norm is the titanium alloy texture normal degree, TW norm reflects the titanium alloy texture degree TW of the same edge texture ring-closed region in the total endocrine region and known as the normal texture degree.

[0010] Further: the titanium alloy texture normal degree TW norm serves as a reference for analysis after image acquisition, when the titanium alloy texture degree TW of the current edge texture ring-closed region is higher than the titanium alloy texture normal degree TW norm , the titanium alloy texture degree TW, texture orientation and difference of the current edge texture ring-closed region are introduced into the tissue structure difference degree unit for processing, and similarly, when the titanium alloy texture degree TW of the current edge texture ring-closed region is lower than and equal to the titanium alloy texture normal degree TW norm , the titanium alloy texture degree TW, texture orientation and difference of the current edge texture ring-closed region do not need to be introduced into the tissue structure difference degree unit for processing.

[0011] Further: the texture orientation distribution difference degree ZF diff and the texture orientation distribution average difference degree ZF avg are calculated as follows, respectively: ; ; wherein: M is the total number of features; f A,i is the i-th feature dimension value of region A; f B,i is the i-th feature dimension value of region B; K is the total number of regions; Texture orientation distribution difference degree ZF diff Before calculation, for each edge texture ring-closed region, according to the direction histogram, gradient direction, extract its texture orientation distribution characteristics, and for each pair of adjacent regions, use Calculate the difference.

[0012] Further: the internal edge blur condition unit calculation formula is as follows: ; Among them: BH is the current blur degree; H0 is the initial blur degree, H0 reflects the blur degree of the initial state of the current edge texture ring-closed region when the implant is just implanted; H max is the maximum blur degree, H max reflects the maximum blur degree known to exist in the total endocrine region; TW min is the minimum titanium alloy texture degree, TW min reflects the minimum titanium alloy texture degree TW known to exist in the total endocrine region; TW max is the maximum titanium alloy texture degree, TW max reflects the maximum titanium alloy texture degree TW known to exist in the total endocrine region.

[0013] Further: the storage module extracts the titanium alloy texture degree TW, the titanium alloy organization degree C and the current blur degree BH respectively The determination threshold of the titanium alloy texture degree TW reflects the complexity and uniformity of the titanium alloy texture of the current edge texture ring-closed region, and the titanium alloy texture degree TW is greater than the determination threshold of the titanium alloy texture degree TW, which indicates that the region has pathological changes and abnormal structure, and the titanium alloy texture degree TW is less than the determination threshold of the titanium alloy texture degree TW, which indicates that the region is normal and stable; The titanium alloy organization degree C reflects the difference degree of the organization structure of the current edge texture ring-closed region and other surrounding regions, and the titanium alloy organization degree C is greater than the determination threshold of the titanium alloy organization degree C, which indicates that the organization structure of the region has abnormal changes, and the titanium alloy organization degree C is less than the determination threshold of the titanium alloy organization degree C, which indicates that the region is consistent with other regions in the organization structure; The current blur degree BH reflects the blur degree of the internal edge texture ring-closed region, and the current blur degree BH is greater than the determination threshold of the current blur degree BH, which indicates that the edge texture information of the region is not clear, and the current blur degree BH is less than the determination threshold of the current blur degree BH, which indicates that the edge texture information of the region is clear.

[0014] Further: the tissue structure difference degree unit processes and obtains the titanium alloy tissue degree C area, which all need to be introduced into the internal edge blur condition unit, to calculate and generate the current blur degree BH of the current edge texture in the ring-like closed area.

[0015] The effects of the above scheme are as follows: 1、The present application can realize accurate extraction of image features in the endocrine region by designing a special algorithm formula and considering the artifact characteristics generated by the titanium alloy implant, especially the calculation of the titanium alloy texture degree TW in the quantitative implant texture feature unit, wherein the introduction of artifact correction and texture feature enhancement effectively improves the extraction accuracy of the texture features under the interference of artifacts.

[0016] 2、The algorithm formula proposed in the present application not only considers the texture features of the image, but also combines the difference degree of the texture direction distribution and the internal edge blur degree, and especially considers the influence of artifacts on these features, realizes comprehensive analysis of the image features in the endocrine region by introducing an artifact identification and suppression algorithm, and effectively reduces the interference of artifacts on the diagnosis results, in addition, according to the comparison processing of the titanium alloy texture degree TW and the normal titanium alloy texture degree TW norm , the image acquisition image with obvious abnormalities, i.e. the inflammation area, can be screened out in time and effectively, so as to further reduce the interference of the edge texture in the ring-like closed normal area. BRIEF DESCRIPTION OF DRAWINGS

[0017] Fig. 1 It is a method flow diagram of the medical image acquisition and analysis method; Fig. 2 It is a structure diagram of the image processing and analysis module in the medical image acquisition and analysis method; Fig. 3 It is a region diagram of the total endocrine region in the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be clearly and completely introduced below with reference to the drawings in the embodiments of the present application.

[0019] Embodiment one, please refer to Figs. 1-3 A medical image acquisition and analysis method for endocrine patients, comprising an image acquisition module, a storage module, an image processing and analysis module, and a result generation module, characterized in that the image processing and analysis module comprises a quantitative implant texture feature unit, a tissue structure difference degree unit, and an internal edge blur condition unit. The specific image acquisition and analysis steps are as follows: Step S1, using the image acquisition module, real-time acquisition of each region image in the total endocrine region of the current patient being imaged; Step S2, based on the collected images of each region, identify and extract the images of all regions with ring-like closed edge texture, and introduce into the image processing analysis module; Step S3, using the image processing analysis module, sequentially calculate the output titanium alloy texture degree TW, titanium alloy organization degree C and current blur degree BH; Step S4, based on the results of titanium alloy texture degree TW, titanium alloy organization degree C and current blur degree BH, and using the result generation module to output the generated results and analysis; Step S5, the storage module will store all the results collected, processed and generated in steps S1-S4; The devices used by the image acquisition module include magnetic resonance imaging equipment, computed tomography scanner; The devices used by the storage module include storage devices; The devices used by the image processing analysis module include image processing workstations; The devices used by the result generation module include displays.

[0020] The embodiment relates to multiple steps, modules and units, and the use of multiple devices, which can realize accurate collection, processing and analysis of endocrine patient image data by comprehensively using these methods and technologies, wherein the quantified implant texture feature unit, the organization structure difference degree unit and the internal edge blur condition unit each bear different calculation purposes in the medical image collection and analysis method for endocrine patients, and collectively constitute a complete image processing and analysis process; Among them, step S2 adopts a deep learning segmentation algorithm based on U-Net to identify ring-closed regions, the specific network structure includes an encoder (4 layers of convolution + pooling) and a decoder (4 layers of upsampling + jump connection), and the training data is MRI images (ring region labeled) containing 1000 titanium alloy implant patients; For the MRI titanium alloy artifact, a multi-echo acquisition technology (ME-GRE sequence) combined with an adaptive threshold artifact removal algorithm is adopted, the specific parameters are echo time TE=5ms, 10ms and 15ms, and the signal intensity difference is used to eliminate artifact interference.

[0021] Please refer to Figs. 1-3 The calculation formula of the quantified implant texture feature unit is as follows: ; ; Among them: TW is the titanium alloy texture degree; N is the total number of pixels, which reflects the amount of all pixels in the current edge texture with ring-closed region; Ti Ti alloy texture value of the i-th pixel, T i reflects the Ti alloy texture value of any pixel of the image edge texture in the total endocrine region where the patient is implanted with the implant, wherein the edge texture is in the form of a ring-like closed region; T avg Ti alloy texture average value, T avg reflects the average degree of the Ti alloy texture value of the i-th pixel T i in the ring-like closed region of the current edge texture; BR is the current radius, BR BR max is the maximum radius, BR max reflects the maximum value of the radius of the total endocrine region; for measuring the dispersion degree of the Ti alloy texture value of the pixel in the ring-like closed region of the current edge texture; the result ranges between 0 and 1, and is used to adjust the value of T .

[0022] The embodiment first calculates the Ti alloy texture value of the i-th pixel T i is a texture feature value extracted by a Gabor filter, and the specific steps are as follows: after pre-processing (such as noise reduction and normalization) of the CT / MRI image, a 5x5 Gabor filter (direction 0°, 45°, 90°, 135°, wavelength 5px) is used to extract the texture feature of the pixel, and the Ti alloy texture value of the i-th pixel T i is the normalized result (range 0-255) of the feature value; The standard deviation of the calculation part is a commonly used measure in statistics, which is used to describe the dispersion degree of a group of data. In the formula, the standard deviation is obtained by calculating the average of the square of the difference between the Ti alloy texture value of the i-th pixel T i and the Ti alloy texture average value T avg , and then taking the square root. This standard deviation reflects the fluctuation of the Ti alloy texture value, i.e. the complexity and non-uniformity of the texture. The larger the standard deviation, the higher the dispersion degree of the Ti alloy texture, i.e. the more complex the texture; The calculation part calculates the ratio of the current radius BR to the maximum radius BR maxThe adjustment factor is designed to consider the influence of the size of the current edge texture ring-closed region on the titanium alloy texture degree TW. When the closed ring region is small, i.e., the current radius BR is close to 0, the adjustment factor is close to 1, and the influence on the standard deviation is also small. When the closed ring region is large, i.e., the current radius BR is close to the maximum radius BR max , the adjustment factor decreases, and the influence on the standard deviation increases, which reflects that the larger the closed ring region, the smaller its contribution to the overall titanium alloy texture degree TW. In addition, the current radius BR is obtained by the minimum circumscribed circle algorithm to calculate the ring region boundary, and the maximum radius BR max is the maximum circumscribed circle radius of the total endocrine region in three-dimensional space, which is converted by the layer thickness and pixel spacing in the DICOM file header information of the CT / MRI image; This algorithm unit can accurately quantify the texture characteristics of titanium alloy in the endocrine region by calculating the standard deviation and multiplying it by the adjustment factor related to the radius of the closed ring region. This quantification method not only improves the accuracy of image analysis but also provides reliable data support for subsequent analysis and diagnosis. Due to the differences in individual conditions of endocrine patients, different degrees of inflammatory regions may occur between titanium alloy implants and body tissues. These inflammatory regions will form artifacts in MRI images, thereby interfering with image accuracy. The quantification of implant texture characteristics unit can indirectly reflect the influence of the inflammatory region on titanium alloy texture by calculating the titanium alloy texture degree TW to provide clues about the patient's inflammatory condition. The titanium alloy texture degree TW calculated by the quantification of implant texture characteristics unit is the basis for the subsequent calculation of tissue structure difference degree unit and internal edge blur condition unit. Through the accurate calculation of this formula, accurate data input can be provided for the subsequent calculation of tissue degree and edge blur degree, thereby improving the accuracy and reliability of the entire image processing and analysis process.

[0023] Please refer to Figs. 1-3 , the calculation formula of the tissue structure difference degree unit is as follows: ; Where: C is the titanium alloy tissue degree; ZF diff is the texture orientation distribution difference degree, ZF diff reflects the difference degree of the texture orientation distribution in the current edge texture ring-closed region and the surrounding edge texture ring-closed region; ZF avg is the average texture orientation distribution difference degree, ZF avgreflects the average degree of difference between the texture direction distribution of the current edge texture ring-closed region and the texture direction distribution of all surrounding edge texture ring-closed regions; TW0 is the initial degree of titanium alloy texture, and TW0 reflects the titanium alloy texture degree TW of the current edge texture ring-closed region in the initial state, specifically, after the patient is implanted with the titanium alloy implant, the image acquisition module is used to acquire images of the total endocrine region, and then the quantification implant texture feature unit in the image processing analysis module is used to calculate and obtain the initial degree of titanium alloy texture TW0 according to the calculation formula: TW norm is the normal degree of titanium alloy texture, and TW norm reflects the titanium alloy texture degree TW of the same edge texture ring-closed region in the total endocrine region and known as the normal texture degree. texture direction distribution difference degree ZF diff and texture direction distribution average difference degree ZF avg The respective calculation formulas are as follows: ; ; Wherein: M is the total number of features; f A,i is the i-th feature dimension value of region A; f B,i is the i-th feature dimension value of region B; K is the total number of regions; texture direction distribution difference degree ZF diff Before calculation, for each edge texture ring-closed region, the texture direction distribution features are extracted according to the direction histogram and gradient direction, and for each pair of adjacent regions, the difference is calculated using .

[0024] The embodiment first calculates the ratio of the texture direction distribution difference degree ZF diff and the texture direction distribution average difference degree ZF avg , and then multiplies the titanium alloy texture degree TW after adding 1 to reflect the uniqueness of the texture direction of the initial titanium alloy endocrine region. The greater the ratio, the greater the difference between the texture direction of the current edge texture ring-closed region and the surrounding regions. Wherein, for the texture direction distribution difference degree ZF diff ​The calculation is based on the direction histogram, gradient direction, and extraction of the texture orientation distribution characteristics. Therefore, the selection of the surrounding edge texture ring-closed region is based on the spatial position adjacent relationship, that is, the region that is spatially adjacent to the edge texture ring-closed region being analyzed is selected as the "surrounding edge texture ring-closed region". Then, for each pair of adjacent regions, the texture orientation distribution difference ZF is calculated using the following formula to measure the difference in texture orientation distribution between them. diff The calculation part reflects the difference between the two by calculating the square root of the square of the difference between the initial degree of titanium alloy texture TW0 and the normal degree of titanium alloy texture TW norm . The greater the difference, the greater the difference between the texture degree of the edge texture ring-closed region and the normal region. The initial degree of titanium alloy texture TW0 is calculated by MRI image collected immediately after the operation (within 24 hours). The specific process is as follows: after the patient is implanted with a titanium alloy implant, a 3.0T MRI device (Siemens Prisma) is used to scan using a T1 weighted sequence (TR=500ms, TE=20ms). The initial degree of titanium alloy texture TW0 is calculated according to the calculation formula of the quantitative implant texture feature unit. The normal degree of titanium alloy texture TW norm is obtained by multi-center clinical data statistics: 500 cases of titanium alloy implant patients without inflammatory reaction are collected, and the ring region TW value is calculated. The upper limit of the 95% confidence interval is taken as TW norm threshold (such as TW norm =15.6±2.3). This algorithm unit not only considers the texture characteristics of titanium alloy, but also combines the texture orientation distribution difference, so as to more comprehensively evaluate the tissue state of titanium alloy in the endocrine region. This comprehensive evaluation method helps to understand the relationship between the patient's titanium alloy implant and the surrounding tissue. By comparing the texture degree difference between the current edge texture ring-closed region and the surrounding other endocrine regions and the normal endocrine region, the tissue structure difference degree unit can identify abnormal tissue changes, which has important significance for timely detection and treatment of potential complications.

[0025] ​It should be noted that the titanium alloy texture degree TW reflects the complexity and uniformity of the titanium alloy texture in the current edge texture ring-closed region, and the titanium alloy organization degree C reflects the difference in the organization structure between the current edge texture ring-closed region and the surrounding other regions. The titanium alloy organization degree C is calculated by the titanium alloy texture degree TW, which indicates that there is an inherent relationship between the two in the organization characteristics reflected by the medical image, because the texture characteristics are often the external manifestation of the organization structure characteristics. TW and C are multiplied, which is to further consider the difference in the organization structure on the basis of the texture characteristics reflected by TW, and This item is used to modify the result calculated by the titanium alloy texture degree TW combined with the texture trend distribution difference, so that the titanium alloy organization degree C more accurately reflects the difference between the current region and the normal organization structure.

[0026] Please refer to Figs. 1-3 The calculation formula of the internal edge blur condition unit is as follows: ; Among them: BH is the current blur degree; H0 is the initial blur degree, which reflects the blur degree of the current edge texture ring-closed region in the initial state after the implant is implanted. Specifically, after the implant is implanted, the image is acquired by the image acquisition device, and then the blur degree of the region is determined by the processing of the image processing analysis module; H max is the maximum blur degree, H max reflects the maximum blur degree known to exist in the total endocrine region; TW min is the minimum titanium alloy texture degree TW, TW min reflects the minimum titanium alloy texture degree TW known to exist in the total endocrine region; TW max is the maximum titanium alloy texture degree TW, TW max reflects the maximum titanium alloy texture degree TW known to exist in the total endocrine region.

[0027] The embodiment first The calculation part reflects the clarity of the internal edge texture information by subtracting the difference between the initial blur degree H0 and the maximum blur degree H max , and multiplying the titanium alloy organization degree C. The smaller the ratio, the more blurred the internal edge texture information is; The calculation part calculates the product of , and divides by the maximum titanium alloy texture degree TW max ​With titanium alloy texture minimal degree TW min The difference between the two is used to get an adjustment term, which is subtracted from the result of the blur effect part to reflect the effect of the texture degree range on the blur degree of the internal edge; The initial blur level H0 is the blur level of the annular area in the image immediately after surgery. The calculation method is: use the Laplace operator to detect the edge of the image, and measure the blur level by the standard deviation of the edge gradient amplitude. The formula is: ,in The image processed by the Laplace operator, the maximum blur level H max Determined by clinical extreme case statistics: 20 images with severe artifact interference were selected, their BH values ​​were calculated, and the maximum value was taken as H max (such as H max =25.8); This algorithm unit can quantify the blur degree of the titanium alloy implant edge in the current closed ring-shaped edge texture area, and then provide intuitive information about the clarity of the boundary between the implant and the surrounding tissue after image processing; It is worth noting that tissue changes in inflamed areas and during the healing process often lead to increased edge blur. The internal edge blur status unit can reflect these changes by calculating the current blur level BH, thereby providing clues about the patient's inflammation and healing process.

[0028] See also Figs. 1-3 The storage module extracts the judgment thresholds of the titanium alloy texture degree TW, the titanium alloy tissue degree C, and the current blur degree BH. The titanium alloy texture degree TW reflects the complexity and uniformity of the titanium alloy texture in the area where the current edge texture is a ring-shaped closed shape. If the titanium alloy texture degree TW is greater than the judgment threshold of the titanium alloy texture degree TW, it indicates that the area has pathological changes and abnormal tissue structure. If the titanium alloy texture degree TW is less than the judgment threshold of the titanium alloy texture degree TW, it indicates that the area is normal and stable. The titanium alloy microstructure degree C reflects the degree of difference in microstructure between the current ring-shaped closed area with edge texture and other surrounding areas. If the titanium alloy microstructure degree C is greater than the judgment threshold of the titanium alloy microstructure degree C, it indicates that the microstructure of the area has undergone abnormal changes. If the titanium alloy microstructure degree C is less than the judgment threshold of the titanium alloy microstructure degree C, it indicates that the area is consistent with other areas in microstructure. The current blur degree BH reflects the blur degree inside the circular closed area of ​​the current edge texture. If the current blur degree BH is greater than the judgment threshold of the current blur degree BH, it indicates that the edge texture information of the area is not clear. If the current blur degree BH is less than the judgment threshold of the current blur degree BH, it indicates that the edge texture information of the area is clear.

[0029] In this embodiment, the quantitative analysis of the titanium alloy texture degree TW obtained by the implant texture feature unit provides objective data about the current edge texture ring-like closed region titanium alloy implant state. Through comparison with the normal texture degree, abnormal regions can be more easily identified, thereby improving the accuracy of diagnosis. The titanium alloy organization degree C calculated by the organization structure difference degree unit reflects the difference degree of the current edge texture ring-like closed region and other surrounding regions in the organization structure. By comparing these differences, it can be evaluated whether the organization structure of the current edge texture ring-like closed region has changed significantly, thereby judging whether there is a pathological process. The current blur degree BH calculated by the internal edge blur condition unit reflects the blur degree of the internal edge texture information of the current edge texture ring-like closed region. This result is crucial for evaluating edge sharpness, because the sharpness of the edge directly affects the interpretation of the image and the accuracy of the diagnosis. In addition, the result of the current blur degree BH can also be used to guide the optimization of image quality during medical image acquisition and analysis. Specifically, if the current blur degree BH is high, the acquisition parameters need to be adjusted and more advanced image processing techniques need to be used to improve the sharpness of the image. In summary, the results of the implant texture feature unit, the organization structure difference degree unit, and the internal edge blur condition unit have brought significant beneficial effects to the medical image acquisition and analysis method for endocrine patients.

[0030] Embodiment two, please refer to Figs. 1-3 , titanium alloy texture normal degree TW norm As a benchmark for analysis after image acquisition, when the titanium alloy texture degree TW of the current edge texture ring-like closed region is higher than the titanium alloy texture normal degree TW norm , the titanium alloy texture degree TW, texture direction and difference of the current edge texture ring-like closed region are introduced into the organization structure difference degree unit for processing. Similarly, when the titanium alloy texture degree TW of the current edge texture ring-like closed region is lower than and equal to the titanium alloy texture normal degree TW norm , the titanium alloy texture degree TW, texture direction and difference of the current edge texture ring-like closed region do not need to be introduced into the organization structure difference degree unit for processing. The regions processed by the organization structure difference degree unit and obtaining the titanium alloy organization degree C all need to be introduced into the internal edge blur condition unit, calculate and generate the current blur degree BH of the current edge texture ring-like closed region.

[0031] In this embodiment, the closed structure is a characteristic of the endocrine region, but not all endocrine regions with closed structures are abnormal. The present invention compares the titanium alloy texture level TW to be determined with the titanium alloy texture normal level TW of the known normal region. norm When making a comparison, it is necessary to ensure that both are measured under the same conditions and using the same analysis methods and evaluation criteria, and to compare the texture degree TW of the titanium alloy to be determined with the normal texture degree TW of the titanium alloy in the known normal area. norm The results of the comparison can also quickly divide and generate endocrine areas with abnormal closed structures from the image acquisition and processing, and further introduce these areas into the tissue structure difference degree unit and the internal edge blur condition unit for processing, so as to quickly generate the results of tissue structure differences and blur in the image.

[0032] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the scope defined by the claims.

Claims

1. A medical image acquisition and analysis method for endocrine patients, comprising an image acquisition module, a storage module, an image processing and analysis module, and a result generation module, characterized in that: The image processing and analysis module includes a unit for quantifying implant texture features, a unit for determining the degree of tissue structure differences, and a unit for determining the internal edge fuzziness. The specific steps for image acquisition and analysis are as follows: Step S1, using the image acquisition module to acquire images of each region in the total endocrine region of the current patient in real time; Step S2: Based on the collected images of each region, identify and extract images of all regions with closed circular edge textures, and introduce them into the image processing and analysis module; Step S3, using the image processing and analysis module, sequentially calculating and outputting the titanium alloy texture degree TW, the titanium alloy organization degree C and the current blur degree BH; Step S4: Based on the titanium alloy texture degree TW, the titanium alloy organization degree C and the current blur degree BH, the result and analysis generated by the result generation module are outputted; Step S5: The storage module stores all the results collected, processed and generated in steps S1 to S4.

2. The medical image acquisition and analysis method for endocrine patients according to claim 1, characterized in that: The equipment used by the image acquisition module includes magnetic resonance imaging equipment and computer tomography scanner; The devices used by the storage module include storage devices; The equipment used by the image processing and analysis module includes an image processing workstation; The equipment used by the result generation module includes a display.

3. The medical image acquisition and analysis method for endocrine patients according to claim 2, characterized in that: The calculation formula for the quantitative implant texture feature unit is as follows: ； ； in: TW is the texture level of titanium alloy; N is the total number of pixels, which reflects the number of all pixels in the circular closed area of ​​the current edge texture; T i is the titanium alloy texture value of the i-th pixel, T i Reflects the titanium alloy texture value of any pixel in the total endocrine area of ​​the patient where the implant is implanted, in which the edge texture of an image is a ring-shaped closed area; T avg is the average texture value of titanium alloy, T avg Reflects the titanium alloy texture value T of the i-th pixel of all pixels in the ring-shaped closed area of ​​the current edge texture i the average degree of BR is the current radius, which reflects the radius of the ring-shaped closed area of ​​the current edge texture; BR max is the maximum radius, BR max Reflects the maximum value of the radius of the total endocrine area; Used to measure the discrete degree of titanium alloy texture value of pixels in the ring-shaped closed area of ​​the current edge texture; The result ranges from 0 to 1 and is used to adjust The value of .

4. The medical image acquisition and analysis method for endocrine patients according to claim 3, characterized in that: The calculation formula of the organizational structure difference degree unit is as follows: ； in: C is the degree of titanium alloy organization; ZF diff is the texture distribution difference, ZF diff Reflects the difference in texture direction distribution between the current edge texture in a ring-shaped closed area and the surrounding edge texture in a ring-shaped closed area; ZF avg is the average difference of texture orientation distribution, ZF avg Reflects the average degree of difference in texture direction distribution between the current edge texture in a circular closed area and all surrounding edge texture in a circular closed area; TW0 is the initial degree of titanium alloy texture, which reflects the degree of titanium alloy texture TW in the initial state of the area with the current edge texture in a ring-shaped closed shape when the implant is just implanted; TW norm The texture of titanium alloy is normal, TW norm It reflects the texture degree TW of titanium alloy in the total endocrine area, in which the edge texture is also circular and closed and is known as the normal texture degree.

5. The medical image acquisition and analysis method for endocrine patients according to claim 4, characterized in that: The titanium alloy texture is normal TW norm As the benchmark for analysis after image acquisition, when the current edge texture is a ring-shaped closed area of ​​titanium alloy texture degree TW higher than the normal titanium alloy texture degree TW norm When the titanium alloy texture degree TW, texture direction and difference of the current edge texture in the ring-shaped closed area are introduced into the tissue structure difference degree unit for processing, similarly, when the titanium alloy texture degree TW of the current edge texture in the ring-shaped closed area is lower than or equal to the normal titanium alloy texture degree TW norm When the titanium alloy texture degree TW, texture direction and difference of the current edge texture in a ring-shaped closed area do not need to be introduced into the tissue structure difference degree unit for processing.

6. The medical image acquisition and analysis method for endocrine patients according to claim 4, characterized in that: The texture orientation distribution difference ZF diff and the average difference ZF of texture orientation distribution avg The calculation formulas are as follows: ； ； in: M is the total number of features; f A,i is the i-th feature dimension value of region A; f B,i is the i-th feature dimension value of region B; K is the total number of regions; Texture orientation distribution difference ZF diff Before calculation, for each ring-shaped closed area of ​​edge texture, the texture distribution characteristics are extracted according to the direction histogram and gradient direction, and for each pair of adjacent areas, the texture distribution characteristics are extracted according to the direction histogram and gradient direction. Calculate the difference.

7. The medical image acquisition and analysis method for endocrine patients according to claim 5, characterized in that: The calculation formula of the internal edge blur condition unit is as follows: ； in: BH is the current blur level; H0 is the initial degree of blur, which reflects the degree of blur in the initial state of the area with the current edge texture in a ring-shaped closed shape when the implant is just implanted; H max is the maximum blur level, H max Reflects the known and maximum ambiguity in the total endocrine region; TW min The texture of titanium alloy is the smallest, TW min Reflects the known and minimum titanium alloy texture level TW in the total endocrine area; TW max For the maximum degree of titanium alloy texture, TW max Reflects the known and maximum titanium alloy texture TW in the total endocrine area.

8. The medical image acquisition and analysis method for endocrine patients according to claim 7, characterized in that: The storage module extracts the respective determination thresholds of the titanium alloy texture degree TW, the titanium alloy tissue degree C, and the current blur degree BH, wherein the titanium alloy texture degree TW reflects the complexity and uniformity of the titanium alloy texture in the region where the current edge texture is annular and closed, and if the titanium alloy texture degree TW is greater than the determination threshold of the titanium alloy texture degree TW, it indicates that pathological changes and abnormal tissue structure exist in the region, and if the titanium alloy texture degree TW is less than the determination threshold of the titanium alloy texture degree TW, it indicates that the region is normal and stable; The titanium alloy microstructure degree C reflects the degree of difference in microstructure between the current ring-shaped closed area with edge texture and other surrounding areas. If the titanium alloy microstructure degree C is greater than the judgment threshold of the titanium alloy microstructure degree C, it indicates that the microstructure of the area has undergone abnormal changes. If the titanium alloy microstructure degree C is less than the judgment threshold of the titanium alloy microstructure degree C, it indicates that the area is consistent with other areas in microstructure. The current blur degree BH reflects the blur degree inside the circular closed area of ​​the current edge texture. If the current blur degree BH is greater than the judgment threshold of the current blur degree BH, it indicates that the edge texture information of the area is not clear. If the current blur degree BH is less than the judgment threshold of the current blur degree BH, it indicates that the edge texture information of the area is clear.

9. The medical image acquisition and analysis method for endocrine patients according to claim 7, characterized in that: The area where the organizational structure difference degree unit processes and obtains the titanium alloy organizational degree C needs to be introduced into the internal edge fuzzy condition unit to calculate and generate the current fuzzy degree BH of the area where the current edge texture is annular and closed.

Images