CN117853334B

CN117853334B - Medical image reconstruction method and system based on DICOM image

Info

Publication number: CN117853334B
Application number: CN202410257310.5A
Authority: CN
Inventors: 李志�; 刘玲; 王芳芳
Original assignee: People's Liberation Army Navy Navy Qingdao Special Service Sanatorium
Current assignee: People's Liberation Army Navy Navy Qingdao Special Service Sanatorium
Priority date: 2024-03-07
Filing date: 2024-03-07
Publication date: 2024-05-14
Anticipated expiration: 2044-03-07
Also published as: CN117853334A

Abstract

The invention relates to the technical field of medical image processing, in particular to a medical image reconstruction method and system based on DICOM images. The method comprises the following steps: obtaining DICOM medical image data, performing key element mining analysis and pixel structure analysis, and obtaining medical image pixel structure information data; performing pixel point sampling processing and region segmentation on the DICOM medical image data to obtain pixel sampling point region image data; performing pixel equivalent connection processing on the image data of the pixel sampling point region based on the medical image pixel structure information data to obtain a DICOM medical image pixel equivalent surface; and carrying out equivalent image reconstruction on the image data of the pixel sampling point region according to the equivalent of the DICOM medical image pixel to obtain DICOM medical reconstruction image data, and carrying out reconstruction quality evaluation analysis and adjustment optimization to obtain a medical reconstruction image optimization result. The invention can improve the quality and reconstruction accuracy of medical images.

Description

Medical image reconstruction method and system based on DICOM image

Technical Field

The invention relates to the technical field of medical image processing, in particular to a medical image reconstruction method and system based on DICOM images.

Background

DICOM is an international standard in the medical imaging arts for storing, transmitting and sharing medical image information. In the field of medical image processing technology, image reconstruction is a key technology for improving and optimizing the quality and information extraction of medical images. However, the conventional DICOM image reconstruction method is limited by problems in terms of pixel resolution, image pixel structure, etc., and limits the quality of medical images and the accuracy of image reconstruction.

Disclosure of Invention

Accordingly, the present invention is directed to a medical image reconstruction method and system based on DICOM images, which solve at least one of the above problems.

In order to achieve the above object, a medical image reconstruction method based on DICOM image includes the following steps:

step S1: acquiring DICOM medical image data, and performing key element mining analysis on the DICOM medical image data to obtain medical image key element information data; performing pixel structure analysis on the DICOM medical image data according to the medical image key element information data to obtain medical image pixel structure information data;

Step S2: performing pixel point sampling processing on the DICOM medical image data to obtain DICOM medical image pixel sampling points; performing region segmentation on the DICOM medical image data based on the DICOM medical image pixel sampling points to obtain pixel sampling point region image data; performing image layer thickness analysis on the image data of the pixel sampling point region based on the medical image pixel structure information data to obtain image layer thickness level data of the sampling point region;

Step S3: carrying out pixel color statistics calculation on the image data of the pixel sampling point region to obtain a pixel color value of the image of the sampling point region; performing pixel equivalent connection processing on the DICOM medical image pixel sampling points based on the image layer thickness horizontal data of the sampling point area and the image pixel color values of the sampling point area to obtain a DICOM medical image pixel equivalent surface;

step S4: performing equivalent image reconstruction on the pixel sampling point region image data according to the equivalent of the DICOM medical image pixels to obtain DICOM medical reconstructed image data; performing reconstruction quality evaluation analysis on the DICOM medical reconstruction image data to obtain a medical reconstruction image quality correction factor; and adjusting and optimizing the DICOM medical reconstruction image data according to the medical reconstruction image quality correction factors to obtain a medical reconstruction image optimization result.

According to the invention, the corresponding DICOM medical image data are acquired firstly, so that subsequent analysis and processing are carried out to extract key information in medical images. Meanwhile, through carrying out key element mining analysis on the DICOM medical image data, important elements in the DICOM medical image, such as patient information, scanning equipment information, image types and the like, can be identified, and basic data are provided for subsequent analysis. Through the pixel structure analysis of the DICOM medical image data by using the medical image key element information data, the pixel structure characteristics of different types of medical images can be understood more deeply, so that the structural characteristics in the images can be identified, and more information support is provided for the subsequent processing process. Secondly, through carrying out pixel extraction processing and sampling on DICOM medical image data, the preset sampling density frequency is used in the sampling process, so that the data volume is reduced while the image characteristics are reserved, the processing efficiency is improved, and the key point of the step is that a high-efficiency data processing basis can be provided for subsequent analysis. The DICOM medical image pixel sampling points are used for carrying out region segmentation on the DICOM medical image data, so that a region medical image segmentation result corresponding to the pixel sampling points can be successfully obtained, the step is helpful for segmenting the medical image corresponding to the DICOM medical image pixel sampling points into individual medical image regions, more local data is provided for subsequent analysis, and the method is beneficial for further researching the characteristics of specific regions in the medical image. By carrying out image layer thickness analysis on the image data of the pixel sampling point region according to the image pixel structure information data of the medical image, the layer thickness level information of different regions in the medical image can be more comprehensively known through the analysis, more detailed data support is provided for subsequent processing and research, and the implementation of the step is helpful for enabling the extraction of information with depth and comprehensiveness from the medical image to be possible. Then, through carrying out pixel color statistics calculation on the image data of the pixel sampling point region, accurate and comprehensive pixel color values of the image of the sampling point region can be obtained, so that important information about image content can be provided, color characteristics of different structures and tissues in medical images can be revealed, and a foundation is laid for subsequent image processing and interpretation. The DICOM medical image pixel sampling points are subjected to equivalent screening division and connection processing by combining the sampling point region image layer thickness horizontal data and the sampling point region image pixel color values, the medical image pixels can be effectively divided into equivalent point regions with different values based on layer thickness information by the equivalent screening division, and the scattered equivalent point regions are connected, so that a more continuous and complete medical image expression form is formed, the structure and feature distribution in the different value regions can be more comprehensively understood, and more powerful support is provided for further analysis of medical images. And finally, performing equivalent image reconstruction on the basis of the pixel equivalent surface of the DICOM medical image to obtain DICOM medical reconstruction image data. And then, carrying out reconstruction quality evaluation analysis on the reconstructed image data to obtain a medical reconstruction image quality correction factor, and carrying out adjustment and optimization according to the correction factor to finally obtain an optimization result of the medical reconstruction image, wherein the series of operations can effectively improve the quality of the medical image, thereby ensuring the accuracy and the reliability of the reconstruction of the medical image data.

Preferably, the present invention also provides a DICOM image-based medical image reconstruction system for performing the DICOM image-based medical image reconstruction method as described above, the DICOM image-based medical image reconstruction system comprising:

The medical image pixel structure analysis module is used for acquiring DICOM medical image data, and carrying out key element mining analysis on the DICOM medical image data to obtain medical image key element information data; performing pixel structure analysis on the DICOM medical image data according to the medical image key element information data so as to obtain medical image pixel structure information data;

The pixel point image layer thickness horizontal analysis module is used for carrying out pixel point sampling processing on the DICOM medical image data to obtain DICOM medical image pixel sampling points; performing region segmentation on the DICOM medical image data based on the DICOM medical image pixel sampling points to obtain pixel sampling point region image data; performing image layer thickness analysis on the image data of the pixel sampling point region based on the medical image pixel structure information data, so as to obtain image layer thickness level data of the sampling point region;

The pixel isosurface is connected with the processing module and is used for carrying out pixel color statistics calculation on the image data of the pixel sampling point region to obtain the image pixel color value of the sampling point region; performing pixel equivalent connection processing on the DICOM medical image pixel sampling points based on the image layer thickness horizontal data of the sampling point area and the image pixel color values of the sampling point area to obtain a DICOM medical image pixel equivalent surface;

the medical image reconstruction optimization module is used for carrying out equivalent image reconstruction on the image data of the pixel sampling point region according to the equivalent surface of the DICOM medical image pixels so as to obtain DICOM medical reconstruction image data; performing reconstruction quality evaluation analysis on the DICOM medical reconstruction image data to obtain a medical reconstruction image quality correction factor; and adjusting and optimizing the DICOM medical reconstruction image data according to the medical reconstruction image quality correction factors, so as to obtain a medical reconstruction image optimization result.

The medical image reconstruction system based on the DICOM image comprises a medical image pixel structure analysis module, a pixel point image layer thickness horizontal analysis module, a pixel isosurface connection processing module and a medical image reconstruction optimization module, and can realize any medical image reconstruction method based on the DICOM image.

Drawings

Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of a non-limiting implementation, made with reference to the accompanying drawings in which:

FIG. 1 is a flow chart showing steps of a medical image reconstruction method based on DICOM images according to the present invention;

FIG. 2 is a detailed step flow chart of step S1 in FIG. 1;

fig. 3 is a detailed step flow chart of step S12 in fig. 2.

Detailed Description

The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.

Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.

It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.

In order to achieve the above objective, referring to fig. 1 to 3, the present invention provides a medical image reconstruction method based on DICOM images, the method comprises the following steps:

In the embodiment of the present invention, please refer to fig. 1, which is a schematic flow chart of steps of a medical image reconstruction method based on DICOM images of the present invention, in this example, the steps of the medical image reconstruction method based on DICOM images include:

According to the embodiment of the invention, the medical image data in the corresponding DICOM format is obtained from medical equipment, a hospital information system or other medical data information sources, and the collected medical image data is preprocessed, including denoising, enhancement, edge detection and other operations, so that the quality and definition of medical images are improved, and the DICOM medical image data are obtained. And secondly, analyzing the prefix information identifier of the corresponding DICOM file in the DICOM medical image data, judging whether the corresponding DICOM medical image file accords with the DICOM standard according to a predefined rule or a DICOM standard document rule to judge the prefix information identifier obtained through analysis, analyzing the medical image file which accords with the DICOM standard at the same time to analyze metadata in the DICOM medical image standard file, including pixel information, image attributes, patient information and the like, and then analyzing the analyzed metadata to deeply mine key features and information in the medical image data so as to obtain the medical image key element information data. Finally, structural topology analysis is carried out on the image pixel structure in the DICOM medical image data according to the medical image key element information data so as to deeply understand the pixel structural characteristics of different types of medical images, and structural characteristic information in the medical images of corresponding types is identified, so that the medical image pixel structural information data is finally obtained.

according to the embodiment of the invention, the DICOM medical image data is processed by using the pixel extraction method, so that the original pixel points of the DICOM medical image are successfully obtained from the corresponding DICOM medical image data, and the sampling is randomly carried out from the collected pixel points by using the preset sampling density frequency (for example, 5 times/second), so that part of the pixel points are selected as the sampling points, the data quantity is reduced, and the important information of the medical image can be reserved, so that the DICOM medical image pixel sampling points are obtained. And secondly, dividing the DICOM medical image area corresponding to the sampled DICOM medical image pixel sampling points by using a proper area dividing algorithm so as to divide the medical image corresponding to the DICOM medical image pixel sampling points into corresponding medical image areas, and eliminating the edge effect in the dividing process, so that the real situation of a medical image structure can be reflected better, and the image data of the pixel sampling point area can be obtained. And then, analyzing the image data of the pixel sampling point region by combining the corresponding image pixel structural features in the medical image pixel structural information data so as to measure the layer thickness distribution condition of the image of the pixel sampling point region corresponding to the DICOM medical image pixel sampling point, and finally obtaining the image layer thickness level data of the sampling point region.

According to the embodiment of the invention, firstly, RGB values of all pixels of an image in a pixel sampling point area are collected, and color values of all pixels in the sampling point area are calculated according to statistics of the RGB values, so that the color values of the pixels of the image in the sampling point area are obtained. Secondly, calculating the image layer thickness level data of the sampling point area to calculate and quantify the accurate measurement value corresponding to the medical image area layer thickness, carrying out interpolation compensation on the image pixel color value of the sampling point area in the sampling point area by using the calculated image layer thickness level value to eliminate the color change among the inner layers of the sampling point area, screening and dividing the DICOM medical image pixel sampling points with the same compensation value into the same point set, and simultaneously connecting the corresponding DICOM medical image pixel sampling points in the medical image pixel equivalent point sets with different values by using a corresponding pixel point connection method to form a more continuous and complete medical image pixel equivalent surface to finally obtain the DICOM medical image pixel equivalent surface.

According to the embodiment of the invention, the medical image data corresponding to the DICOM medical image pixel sampling points in the DICOM medical image pixel equivalent surface are reconstructed and converted into the image format which is easier to manage, transmit and read, so that the DICOM medical reconstructed image data is obtained. Secondly, detecting the image structure and the spatial resolution level of the reconstructed DICOM medical reconstruction image data to objectively evaluate the structure and the spatial resolution level of the medical reconstruction image, and carrying out correction analysis on the corresponding DICOM medical reconstruction image to comprehensively consider two aspects of structural integrity and resolution to evaluate the quality problem and improvement space of the DICOM medical reconstruction image, and simultaneously determining the corresponding quality correction factor according to the evaluation analysis result so as to obtain the quality correction factor of the medical reconstruction image. And then, the quality correction factors of the medical reconstructed image obtained through evaluation are used for adjusting and optimizing the DICOM medical reconstructed image data by using corresponding image processing technologies (including image enhancement, noise removal, sharpening processing and the like) aiming at the existing quality problems and improvement spaces so as to improve key indexes such as definition, contrast and the like of the reconstructed medical image, and finally, the medical reconstructed image optimization result is obtained.

Preferably, step S1 comprises the steps of:

Step S11: obtaining DICOM medical image data;

Step S12: performing key element mining analysis on the DICOM medical image data according to the corresponding DICOM prefix to obtain medical image key element information data;

step S13: extracting pixel data and image type data from the key element information data of the medical image to obtain DICOM medical image pixel data and DICOM medical image type data;

step S14: performing pixel distribution statistical analysis on the DICOM medical image pixel data based on the DICOM medical image type data to obtain different types of medical image pixel distribution status data;

step S15: performing gray level conversion processing on the DICOM medical image pixel data to obtain medical image pixel gray level data; gray level association analysis is carried out on the gray level data of the pixels of the medical image by using a gray level co-occurrence matrix technology, so as to obtain gray level association relation data of the pixels of the medical image;

Step S16: and carrying out pixel structure analysis on the different types of medical image pixel distribution status data according to the medical image pixel gray level association relation data to obtain medical image pixel structure information data.

As an embodiment of the present invention, referring to fig. 2, a detailed step flow chart of step S1 in fig. 1 is shown, in which step S1 includes the following steps:

Step S11: obtaining DICOM medical image data;

According to the embodiment of the invention, the corresponding DICOM-format medical image data is obtained from medical equipment, a hospital information system or other medical data information sources, and the collected medical image data is preprocessed, including denoising, enhancement, edge detection and other operations, so that the quality and definition of medical images are improved, and the DICOM medical image data is finally obtained.

According to the embodiment of the invention, the prefix information identifier of the corresponding DICOM file in the DICOM medical image data is analyzed, the prefix information identifier obtained through analysis is judged according to a predefined rule or a DICOM standard document rule, so that whether the corresponding DICOM medical image file accords with the DICOM standard is judged, meanwhile, the medical image file which accords with the DICOM standard is analyzed, metadata in the DICOM medical image standard file including pixel information, image attributes, patient information and the like are analyzed, and then, key features and information in the medical image data such as important medical image attributes, patient information, image pixel structure conditions and the like are deeply mined through analysis of the analyzed metadata, so that the medical image key element information data is finally obtained.

According to the embodiment of the invention, the pixel data of the corresponding medical image, namely the specific image information of the medical image, is extracted from the key element information data of the medical image, so that the DICOM medical image pixel data is obtained. And then extracting type information of the corresponding medical image from the medical image key element information data, including CT, MRI, X-ray and the like, so as to obtain DICOM medical image type data.

According to the embodiment of the invention, the corresponding DICOM medical image pixel data is subjected to statistical analysis according to different medical image types in the DICOM medical image type data so as to identify the pixel distribution characteristics and differences of different medical images, so that the pixel distribution conditions of different medical images can be better understood, and finally, the different medical image pixel distribution condition data is obtained.

According to the embodiment of the invention, the extracted DICOM medical image pixel data is processed by using an image gray level conversion algorithm so as to convert the corresponding medical image pixel data into gray level values, so that the medical image pixel gray level data is obtained. And then, analyzing the converted medical image pixel gray scale data by using a gray scale co-occurrence matrix technology so as to analyze and calculate the association relation between different pixel gray scales in the medical image, and finally obtaining the medical image pixel gray scale association relation data.

According to the embodiment of the invention, the structural topology analysis is carried out on the pixel distribution status data of the medical images of different types by combining the association relation among the gray scales of different pixel points in the pixel gray scale association relation data of the medical images, so that the pixel structural characteristics of the medical images of different types are deeply understood, structural characteristic information in the medical images of corresponding types is identified, and finally the pixel structural information data of the medical images is obtained.

According to the invention, the corresponding DICOM medical image data is firstly obtained, the step is the first step of the whole medical image reconstruction process, DICOM is a standard of medical image digitization and communication, and the DICOM medical image data is obtained for subsequent analysis and processing so as to extract key information in medical images. Meanwhile, key element mining analysis is carried out on the DICOM medical image data by using the corresponding DICOM prefix, so that important elements in the DICOM medical image, such as patient information, scanning equipment information, image types and the like, can be identified, and basic data are provided for subsequent analysis. Secondly, through extracting pixel data and image type data from the key element information data of the medical image, the content and the characteristics of the medical image can be known in depth, and the pixel data and the image type data comprising the DICOM medical image can be obtained as a result of the step, so that basic data is provided for subsequent analysis and processing. Then, by performing statistical analysis on pixel distribution of the DICOM medical image pixel data based on the DICOM medical image type data, the pixel distribution situation of different types of medical images can be better understood, which is helpful for identifying the characteristics and differences of different types of medical images and provides a reference for further analysis. Then, the pixel data can be converted into gray values by performing gray conversion processing on the DICOM medical image pixel data, so that gray correlation analysis is facilitated. In addition, gray scale correlation analysis is performed on the gray scale data of the pixels of the medical image obtained after conversion by using a gray scale co-occurrence matrix technology, so that gray scale correlation among pixels in the medical image can be deeply explored, and further image characteristic information can be extracted. Finally, through carrying out pixel structure analysis on the pixel distribution status data of the medical images of different types by combining the pixel gray level association relation data of the medical images, the pixel structure characteristics of the medical images of different types can be understood more deeply, thus being beneficial to identifying the structural characteristics in the images and providing more information support for the subsequent processing process.

Preferably, step S12 comprises the steps of:

Step S121: performing file standard judgment on the DICOM medical image data according to the corresponding DICOM prefix to obtain DICOM medical image standard file data;

step S122: performing metadata analysis processing on the DICOM medical image standard file data through a DICOM analysis library to obtain DICOM medical image metadata;

Step S123: performing association calculation on the DICOM medical image element data by using a pixel association degree calculation formula to obtain a DICOM medical image pixel association degree value;

Step S124: comparing and judging the DICOM medical image pixel association degree value according to a preset image association degree threshold, and marking the DICOM medical image metadata corresponding to the DICOM medical image pixel association degree value as strong association metadata when the DICOM medical image pixel association degree value is larger than or equal to the preset image association degree threshold; when the DICOM medical image pixel association degree value is smaller than a preset image association degree threshold, marking the DICOM medical image metadata corresponding to the DICOM medical image pixel association degree value as weak association metadata;

Step S125: and screening the DICOM medical image metadata marked as the strong-association metadata, and carrying out key element mining analysis to obtain medical image key element information data.

As an embodiment of the present invention, referring to fig. 3, a detailed step flow chart of step S12 in fig. 2 is shown, in which step S12 includes the following steps:

according to the embodiment of the invention, the prefix information identifier of the corresponding DICOM file in the DICOM medical image data is analyzed, then the prefix information identifier obtained through analysis is judged according to a predefined rule or a DICOM standard document rule, so that whether the corresponding DICOM medical image file accords with the DICOM standard is judged, the file which accords with the DICOM standard is extracted, and finally the DICOM medical image standard file data is obtained.

According to the embodiment of the invention, the extracted DICOM medical image standard file is analyzed by using a special DICOM analysis library or software so as to analyze metadata in the DICOM medical image standard file, including pixel information, image attributes, patient information and the like, and finally the DICOM medical image metadata is obtained.

According to the embodiment of the invention, a proper pixel association degree calculation formula is formed by combining an index function, a pixel abscissa parameter of a pixel point, a pixel abscissa parameter of a reference point, a pixel abscissa distribution standard deviation, a pixel ordinate parameter of a pixel point, a pixel ordinate parameter of a reference point, a pixel ordinate distribution standard deviation, a pixel coordinate association function, a DICOM medical image gray level distribution adjustment weight, a DICOM medical image brightness level adjustment weight, a pixel abscissa distribution influence association parameter, a pixel ordinate distribution influence association parameter and a correlation parameter, so as to carry out association calculation on DICOM medical image element data, and finally obtain a DICOM medical image pixel association degree value. In addition, the pixel association degree calculation formula can also use any association detection algorithm in the field to replace the association calculation process, and is not limited to the pixel association degree calculation formula.

Comparing and judging the calculated DICOM medical image pixel association degree value by using a preset image association degree threshold, and marking the DICOM medical image metadata corresponding to the DICOM medical image pixel association degree value as strong association metadata if the DICOM medical image pixel association degree value is larger than or equal to the preset image association degree threshold; if the DICOM medical image pixel association degree value is smaller than the preset image association degree threshold value, which indicates that the pixel association degree of the DICOM medical image metadata corresponding to the DICOM medical image pixel association degree value is smaller or is not associated, marking the DICOM medical image metadata corresponding to the DICOM medical image pixel association degree value as weak association metadata.

According to the embodiment of the invention, the DICOM medical image metadata marked as the strong association metadata is screened out, and then the screened strong association metadata is analyzed by using an association rule mining method, so that key features and information in the medical image data, such as important medical image attributes, patient information, image pixel structure conditions and the like, are deeply mined, and finally the medical image key element information data is obtained.

According to the invention, the corresponding DICOM prefix is used for judging the file standard of the corresponding DICOM file in the DICOM medical image data, and the accurate judgment of the standard of the DICOM file is the basis for ensuring the analysis accuracy of subsequent data, so that the medical image file conforming to the DICOM standard can be accurately judged, and the accuracy and the reliability of subsequent processing are ensured. The execution of this step ensures the quality and standardization of the processed data, providing a reliable data basis for the subsequent steps. And secondly, metadata analysis processing is carried out on the DICOM medical image standard file data by using a DICOM analysis library, the DICOM standard file contains rich medical image metadata information which is very critical for deep understanding of the medical image data, and the metadata information in the standard file can be accurately and efficiently extracted by carrying out analysis processing on the standard file by using the DICOM analysis library, so that necessary data support is provided for subsequent association calculation and analysis. Then, the DICOM medical image data is subjected to association calculation by using a pixel association calculation formula, so that a pixel association value between each two pixel points is obtained, and the key point of the step is that important data support can be provided for subsequent image analysis, so that the characteristics and rules of the medical image data can be better understood. And then, setting an image association degree threshold value and judging the calculated medical image pixel association degree value according to the threshold value, so that the medical image metadata can be marked as strong association or weak association, the marking is helpful for distinguishing the association degree between pixels in the medical image, and further guiding the subsequent image analysis and processing, thereby providing basis for classifying and classifying the image data and being helpful for improving the utilization efficiency and accuracy of the data. Finally, key features and information in the medical image data can be deeply mined by screening DICOM medical image metadata marked as strong-association metadata and carrying out key element mining analysis on the DICOM medical image metadata, and the implementation of the step is helpful for extracting valuable information from massive medical image data, so that important support and help are provided for subsequent processing steps.

Preferably, the image association degree calculation formula in step S123 is specifically:

；

In the method, in the process of the invention, For DICOM medical image pixel association degree value,/>As an exponential function,/>For the pixel abscissa parameter of the pixel point in the DICOM medical image corresponding to the DICOM medical image metadata,/>For the pixel abscissa parameter of the DICOM medical image metadata corresponding to the reference point in the DICOM medical image,/>Is the standard deviation of pixel abscissa distribution,/>For the pixel ordinate parameter of the pixel point in the DICOM medical image corresponding to the DICOM medical image metadata,/>For the pixel ordinate parameter of the DICOM medical image metadata corresponding to the reference point in the DICOM medical image,/>Is the standard deviation of the pixel ordinate distribution,For the pixel coordinate correlation function,/>Weight is adjusted for DICOM medical image gray level distribution,/>Weight adjustment for DICOM medical image brightness level,/>Influencing the associated parameters for the pixel abscissa distribution,/>Influence the associated parameters for the pixel ordinate distribution,/>And the correction value is the correction value of the DICOM medical image pixel association degree value.

According to the invention, a specific mathematical model is used and verified to obtain a pixel association degree calculation formula, which is used for carrying out association calculation on DICOM medical image metadata, and the pixel association degree calculation formula is based on integral calculation of a two-dimensional Gaussian function, so that the spatial distribution relation among pixel points in an image is considered, and the association degree of pixels in the medical image is evaluated. By integrating the whole image area, the correlation condition among all pixel points is comprehensively considered, so that the correlation degree evaluation of the whole image is provided, and the correlation structure and the characteristics in the image are helpful to be identified. By using a two-dimensional Gaussian function to consider the distribution of pixel points on an image plane and the distance relation between the pixel points and a reference point, the consideration enables a calculation formula to capture local features in the image more accurately and evaluate the association degree between the pixel points. And secondly, the association degree of the pixel coordinates is adjusted by using a pixel coordinate association function, wherein the gray distribution adjustment weight, the brightness level adjustment weight and the influence association parameters of the pixel abscissa distribution are considered, and the adjustment of the parameters can carry out finer regulation and control on the pixel association degree according to actual conditions, so that the applicability and the accuracy of a calculation formula are improved. In addition, the pixel association degree is corrected by introducing a correction value, so that the influence of errors or other factors on a calculation result can be considered, the stability and the robustness of a calculation formula are improved, and the calculation result is ensured to be more reliable. In combination, the pixel association degree calculation formula can comprehensively evaluate the association degree of pixels in the medical image, thereby being beneficial to identifying important information associated with other metadata. In summary, the formula fully considers the DICOM medical image pixel association degree valueExponential function/>The DICOM medical image metadata corresponds to the pixel abscissa parameters/>, of pixel points in the DICOM medical imageThe DICOM medical image metadata corresponds to the pixel abscissa parameters/>, of the reference points in the DICOM medical imageStandard deviation of pixel abscissa distribution/>The DICOM medical image metadata corresponds to the pixel ordinate parameter/>, of a pixel point in the DICOM medical imagePixel ordinate parameter/>, of DICOM medical image metadata corresponds to a reference point in a DICOM medical imageStandard deviation of pixel ordinate distribution/>Pixel coordinate correlation function/>DICOM medical image gray level distribution adjustment weight/>DICOM medical image brightness level adjustment weight/>Pixel abscissa distribution affects the associated parameters/>The pixel ordinate distribution affects the correlation parameter/>Correction value/>, of DICOM medical image pixel association degree valueWherein the standard deviation/>, is distributed by using the pixel abscissaStandard deviation of pixel ordinate distribution/>DICOM medical image gray level distribution adjustment weight/>DICOM medical image brightness level adjustment weight/>Pixel abscissa distribution affects the associated parameters/>The pixel ordinate distribution affects the associated parameters/>Constitutes a pixel coordinate correlation function/>Functional relation of (2)According to DICOM medical image pixel association degree value/>The interrelationship between the parameters constitutes a functional relationship:

；

the formula can realize the association calculation process of the DICOM medical image metadata, and simultaneously, the correction value of the pixel association degree value of the DICOM medical image is used The introduction of the pixel correlation degree calculation formula can be adjusted according to the error condition in the calculation process, so that the accuracy and the applicability of the pixel correlation degree calculation formula are improved.

Preferably, step S2 comprises the steps of:

step S21: performing pixel point extraction processing on the DICOM medical image data to obtain DICOM medical image pixels;

according to the embodiment of the invention, the DICOM medical image data is processed by using the pixel extraction method, so that the original pixel point of the DICOM medical image is successfully obtained from the corresponding DICOM medical image data, and finally the DICOM medical image pixel point is obtained.

Step S22: performing pixel point sampling processing on the DICOM medical image pixel points according to a preset sampling density frequency to obtain DICOM medical image pixel sampling points;

According to the embodiment of the invention, the sampling is randomly performed from the collected DICOM medical image pixel points by using the preset sampling density frequency (for example, 5 times/second), so that part of the pixel points are selected as the sampling points, the data volume is reduced, the important information of the medical image can be reserved, and finally the DICOM medical image pixel sampling points are obtained.

Step S23: performing region segmentation on the DICOM medical image data based on the DICOM medical image pixel sampling points to obtain a pixel sampling point region medical image segmentation result;

According to the embodiment of the invention, the DICOM medical image area corresponding to the sampled DICOM medical image pixel sampling points is segmented by using a proper area segmentation algorithm, so that the medical image corresponding to the DICOM medical image pixel sampling points is segmented into corresponding medical image areas, and finally, a medical image segmentation result of the pixel sampling point area is obtained.

Step S24: performing region edge optimization on the medical image segmentation result of the pixel sampling point region to obtain image data of the pixel sampling point region;

according to the embodiment of the invention, the region edge optimization algorithm is used for processing the segmented result of the medical image of the pixel sampling point region, so that the clear and accurate boundary between the segmented regions is ensured, the edge effect existing in the segmentation process is eliminated, the real situation of the medical image structure can be better reflected, and finally the image data of the pixel sampling point region is obtained.

Step S25: and carrying out image layer thickness analysis on the image data of the pixel sampling point region based on the medical image pixel structure information data to obtain image layer thickness level data of the sampling point region.

According to the embodiment of the invention, the image data of the pixel sampling point area is analyzed by combining the corresponding image pixel structural features in the medical image pixel structural information data so as to measure the layer thickness distribution condition of the image of the pixel sampling point area corresponding to the DICOM medical image pixel sampling point, and finally the layer thickness level data of the image of the sampling point area is obtained.

According to the invention, the original pixel point of the DICOM medical image can be successfully obtained by carrying out pixel point extraction processing on the DICOM medical image data, and the step is the basis of the whole medical image processing flow, so that abundant original data is provided for subsequent analysis and processing, and detailed information in the medical image is contained. Secondly, the pixel point sampling processing is carried out on the DICOM medical image pixel points according to the preset sampling density frequency, and the preset sampling density frequency is used in the sampling process to help reduce the data volume while preserving the image characteristics, so that the processing efficiency is improved, and the key point of the step is that a high-efficiency data processing basis can be provided for subsequent analysis. Then, the DICOM medical image pixel sampling points are used for carrying out region segmentation on the DICOM medical image data, so that a region medical image segmentation result corresponding to the pixel sampling points can be successfully obtained, and the step is helpful for segmenting the medical image corresponding to the DICOM medical image pixel sampling points into individual medical image regions, so that more local data is provided for subsequent analysis, and the method is beneficial for further researching the characteristics of a specific region in the medical image. Then, the region edge optimization is carried out on the region medical image segmentation result of the pixel sampling point region, so that possible edge effects in the segmentation process can be eliminated, the accuracy and the definition of the medical image are improved, and the optimized image data can better reflect the real situation of the medical image structure. Finally, image layer thickness analysis is carried out on image data of the pixel sampling point region according to the image pixel structure information data of the medical image, and layer thickness level information of different regions in the medical image can be more comprehensively known through the analysis, more detailed data support is provided for subsequent processing and research, and the implementation of the step is helpful for enabling extraction of information with depth and comprehensiveness from the medical image to be possible.

Preferably, step S25 comprises the steps of:

Step S251: image slicing processing is carried out on the image data of the pixel sampling point region to obtain image slice surface data of the pixel sampling point region;

According to the embodiment of the invention, the image of the pixel sampling point region corresponding to the image data of the pixel sampling point region is sliced according to the specific slicing direction so as to be divided into a plurality of layers or slice planes, so that the image is converted into a slice form which is easier to process and analyze, and finally the image slice plane data of the pixel sampling point region is obtained.

Step S252: performing image slice thickness measurement on the image slice surface data of the pixel sampling point region based on the medical image pixel structure information data to obtain image slice thickness distribution data of the sampling point region;

According to the embodiment of the invention, the layer thickness measurement is carried out on the image pixel structure corresponding to each image slice surface in the image slice surface data of the pixel sampling point region by combining the image pixel structure characteristics corresponding to the medical image pixel structure information data, so that the layer thickness distribution conditions of different regions in the medical image are accurately measured, and finally the layer thickness distribution data of the image slice of the sampling point region is obtained.

Step S253: performing feature analysis on the image slice surface data of the pixel sampling point region to obtain the image slice surface feature data of the sampling point region;

according to the embodiment of the invention, the characteristic analysis method is used for analyzing each image slice surface in the image slice surface data of the pixel sampling point region so as to extract the characteristic information such as the texture, the shape, the density and the like of the corresponding image pixel structure related to the medical image, and finally the characteristic data of the image slice surface of the sampling point region is obtained.

Step S254: performing layer thickness influence analysis on the image slice layer thickness distribution data of the sampling point region based on the image slice surface characteristic data of the sampling point region to obtain image slice surface layer thickness-characteristic influence relation data;

According to the embodiment of the invention, the thickness distribution data of the image slice layers of the corresponding sampling point region are analyzed according to the characteristic information of the characteristic data of the image slice layers of the sampling point region, so that the thickness data of the image slice layers and the characteristic data are combined, the correlation and the influence relation between the image slice layers and the characteristic data are analyzed, the influence relation between different characteristics and the thickness of the medical image is further analyzed and understood, and finally the thickness-characteristic influence relation data of the image slice layers are obtained.

Step S255: and carrying out influence correction analysis on the image slice thickness distribution data of the sampling point region according to the image slice thickness-characteristic influence relation data to obtain image slice thickness level data of the sampling point region.

According to the embodiment of the invention, the image slice layer thickness distribution data of the sampling point area is corrected according to the influence relation between different characteristics and layer thicknesses in the image slice layer thickness-characteristic influence relation data, so that the influence of corresponding characteristics on layer thickness measurement is reduced, the accuracy and reliability of layer thickness measurement are ensured, and finally the image layer thickness level data of the sampling point area is obtained.

According to the invention, firstly, image slicing processing is carried out on the image data of the pixel sampling point region according to a specific slicing direction, so that corresponding image slicing surface data of the pixel sampling point region is generated, the processing process is beneficial to converting medical images into a slicing form which is easier to process and analyze, and a more structural data basis is provided for subsequent steps. Secondly, image layer thickness measurement is carried out on image slice surface data of a pixel sampling point region based on medical image pixel structure information data, so that layer thickness conditions of different regions in medical images can be accurately known, and important parameters are provided for subsequent analysis. Then, by carrying out feature analysis on the image slicing surface data of the pixel sampling point region, the method is beneficial to extracting corresponding key features in the medical image slicing surface, such as information of shapes, textures and the like, so that richer data content is provided for subsequent image analysis and processing. And then, by carrying out layer thickness influence analysis on the image slice layer thickness distribution data of the sampling point region based on the extracted characteristic data of the image slice surface of the sampling point region, the layer thickness data of the image slice surface can be combined with the characteristic data to analyze the correlation and influence relation between the image slice surface and the characteristic data, so that an important clue is provided for further understanding the correlation between different characteristics and layer thicknesses in the medical image. Finally, the image slice thickness distribution data of the sampling point area can be successfully obtained by carrying out influence correction analysis on the image slice thickness distribution data of the sampling point area according to the image slice thickness-characteristic influence relation data, the step utilizes the relation data obtained by previous analysis to carry out influence correction on the layer thickness distribution data so as to obtain more accurate and reliable medical image layer thickness distribution data, and the data can provide important reference basis for the subsequent processing process.

Preferably, step S3 comprises the steps of:

Step S31: carrying out pixel color statistics calculation on the image data of the pixel sampling point region to obtain a pixel color value of the image of the sampling point region;

According to the embodiment of the invention, firstly, RGB values of all pixels of an image in a pixel sampling point area are collected, and a statistical calculation method is used for calculating color values according to the RGB values of all pixels in the pixel sampling point area so as to calculate the color values of all pixels in the pixel sampling point area in a statistical manner, and finally, the pixel color values of the image in the pixel sampling point area are obtained.

Step S32: calculating the layer thickness of the image layer thickness level data of the sampling point region by using an image layer thickness calculation formula to obtain the layer thickness value of the image layer of the sampling point region;

According to the embodiment of the invention, a proper image layer thickness degree calculation formula is formed by combining the abscissa parameter, the ordinate parameter, the central position abscissa parameter, the central position ordinate parameter, the abscissa layer thickness level standard deviation, the ordinate layer thickness level standard deviation, the layer thickness broadening amplitude adjustment parameter and related parameters of the image of the sampling point region, so that the layer thickness degree calculation formula is used for calculating the layer thickness degree of the image layer thickness of the sampling point region, the accurate measurement value of the layer thickness of the corresponding medical image region is quantized, and finally the layer thickness degree value of the image layer of the sampling point region is obtained. In addition, the image layer thickness calculation formula can also use any image layer thickness detection algorithm in the field to replace the layer thickness calculation process, and is not limited to the image layer thickness calculation formula.

Step S33: interpolation compensation is carried out on the pixel color values of the image in the sampling point area according to the layer thickness degree value of the image in the sampling point area, so that the interpolation compensation value of the image pixels in the sampling point area is obtained;

According to the embodiment of the invention, interpolation compensation is carried out on the pixel color values of the sampling point region image in the corresponding sampling point region by using the calculated sampling point region image layer thickness value, so that the color change among the inner layers of the sampling point region is eliminated, the continuity and the spatial consistency of the image in the corresponding region are ensured, and finally the pixel interpolation compensation value of the sampling point region image is obtained.

Step S34: performing equivalent screening division on DICOM medical image pixel sampling points based on the image pixel interpolation compensation values of the sampling point areas to obtain DICOM medical image pixel equivalent point sets with different values;

according to the embodiment of the invention, the interpolation compensation value of the image pixels of the sampling point region obtained after interpolation compensation is subjected to equivalent screening, so that the sampling points of the pixels of the DICOM medical image with the same compensation value are screened and divided into the same point set, the pixels of the medical image are effectively divided into equivalent regions with different values, and finally the equivalent point sets of the pixels of the DICOM medical image with different values are obtained.

Step S35: and carrying out pixel equivalent connection processing on the DICOM medical image pixel equivalent point sets with different values to obtain a DICOM medical image pixel equivalent surface.

According to the embodiment of the invention, the corresponding DICOM medical image pixel sampling points in the DICOM medical image pixel equivalent point sets with different values are connected by using the corresponding pixel point connection method, so that a more continuous and complete medical image pixel equivalent surface is formed, and finally the DICOM medical image pixel equivalent surface is obtained.

According to the invention, firstly, through carrying out pixel color statistics calculation on the image data of the pixel sampling point region, the accurate and comprehensive pixel color value of the image of the sampling point region can be obtained, so that important information about image content can be provided, the color characteristics of different structures and tissues in medical images can be revealed, and a foundation is laid for subsequent image processing and interpretation. Secondly, by calculating the layer thickness degree of the image layer thickness level data of the sampling point region by utilizing a proper image layer thickness degree calculation formula, the accurate measurement value of the medical image layer thickness can be obtained, so that the detailed understanding of the three-dimensional structure crossing the image can be provided, key information is provided for the subsequent image analysis processing process, and the method has important value particularly in the process of analyzing the thickness of the corresponding image structure when reconstructing the medical image. Then, by performing interpolation compensation on the corresponding pixel color values of the image in the sampling point region according to the calculated thickness value of the image in the sampling point region, the step can obtain more accurate and continuous interpolation compensation values of the image pixels in the sampling point region, and the interpolation compensation values can enable the color change under different layer thicknesses to be more accurately understood, improve the spatial consistency of image data and help to accurately express the color distribution characteristics of anatomical structures. And then, carrying out equivalent screening division on the DICOM medical image pixel sampling points based on the image pixel interpolation compensation values of the sampling point areas, wherein the equivalent screening division can effectively divide the medical image pixels into equivalent areas with different values based on the layer thickness information, and provides a more structural data basis for subsequent analysis. Finally, performing pixel equivalent connection processing on the DICOM medical image pixel equivalent point sets with different values, thereby obtaining the DICOM medical image pixel equivalent surface. By the method, the scattered equivalent points can be connected to form a more continuous and complete medical image expression form, so that the structure and characteristic distribution in different numerical value areas can be more comprehensively understood, and more powerful support is provided for further analysis of medical images.

Preferably, the image layer thickness calculation formula in step S32 is specifically:

；/>

In the method, in the process of the invention, For position/>Image layer thickness value of sampling point region at position/>For the abscissa parameter of the image corresponding to the sampling point region,/>For the central position abscissa parameter of the image corresponding to the sampling point region,/>For the horizontal standard deviation of the layer thickness of the abscissa of the image corresponding to the sampling point region,/>For the ordinate parameter of the image corresponding to the sampling point region,/>For the ordinate parameter of the central position of the image corresponding to the sampling point region,/>For the vertical layer thickness standard deviation of the image corresponding to the sampling point region,/>Layer thickness broadening amplitude adjustment parameter for image of corresponding sampling point region,/>And (3) correcting the thickness value of the image layer in the sampling point area.

The invention obtains an image layer thickness degree calculation formula through using a specific mathematical model and verifying, which is used for calculating the layer thickness degree of image layer thickness level data of a sampling point area, and the image layer thickness degree calculation formula considers the information in the sampling point area through using double integralThe thickness of the image layer in the sampling point area at the position can be comprehensively estimated by integrating all possible positions in the image, so that comprehensive knowledge of the thickness of the layer in the whole image is provided. By using the corresponding functional form pairs to locate at/>The image of the sampling point area at the central position is given higher weight, and the influence of the sampling points at the edge position is smaller, so that the contribution of the pixels at the central position to the layer thickness degree can be reflected more accurately, the influence of the pixels at the edge position to the layer thickness degree is weakened, and the accuracy of the calculation result is improved. And secondly, parameters in the formula comprise a center position, a standard deviation, a stretching amplitude adjustment parameter and a correction value, and the parameters can be adjusted according to specific conditions so as to adapt to different image characteristics and resolutions. For example, adjusting the standard deviation may control the sensitivity of the layer thickness calculation, adjusting the broadening amplitude parameter may change the range of variation of the layer thickness, and the correction value may be used to correct the calculation result, thereby improving the accuracy and stability of the calculation result. Therefore, the image layer thickness calculation formula comprehensively considers factors such as position correlation measurement, gaussian weighting, parameter adjustment, interpolation compensation and the like, can effectively evaluate the layer thickness degree of the image of the sampling point region, and provides accurate data for subsequent pixel equivalent division and connection processing. In summary, the formula fully considers the position/>Image layer thickness degree value/>, of sampling point region at the positionAbscissa parameter/>, corresponding to sampling point region imageCenter position abscissa parameter/>, corresponding to sampling point region imageHorizontal standard deviation/>, of layer thickness of abscissa corresponding to image of sampling point regionOrdinate parameter/>, corresponding to image of sampling point regionCenter position ordinate parameter/>, corresponding to image of sampling point regionVertical-coordinate layer thickness horizontal standard deviation/>, of image corresponding to sampling point regionLayer thickness broadening amplitude adjustment parameter/>, corresponding to the image of the sampling point regionCorrection value/>, of image layer thickness degree value of sampling point regionAccording to the position/>Image layer thickness degree value/>, of sampling point region at the positionThe interrelationship between the parameters constitutes a functional relationship:

；

The formula can realize the layer thickness calculation process of the layer thickness level data of the image layer thickness of the sampling point area, and simultaneously, the correction value of the layer thickness level value of the image layer of the sampling point area The introduction of the image layer thickness degree calculation formula can be adjusted according to the error condition in the calculation process, so that the accuracy and the applicability of the image layer thickness degree calculation formula are improved.

Preferably, step S4 comprises the steps of:

step S41: performing equivalent image reconstruction on the pixel sampling point region image data according to the equivalent of the DICOM medical image pixels to obtain DICOM medical reconstructed image data;

according to the embodiment of the invention, the medical image data corresponding to the DICOM medical image pixel sampling points in the DICOM medical image pixel equivalent surface are reconstructed and converted into the image format which accords with more detail, so that the reconstructed medical image is easier to manage, transmit and read, and the important image structure information and pixel values are reserved, so that the DICOM medical reconstructed image data is finally obtained.

Step S42: performing structure accuracy detection on the DICOM medical reconstruction image data to obtain medical reconstruction image structure accuracy status data;

According to the embodiment of the invention, the image structure of the reconstructed DICOM medical reconstruction image data is detected by using an image structure detection algorithm and technology, so that whether the structure of the medical reconstruction image is consistent with the original medical image data is objectively evaluated, the corresponding structure and the structural parts in the reconstructed medical image are ensured to be accurate, and the accurate state data of the medical reconstruction image structure is finally obtained.

Step S43: performing image resolution detection on the DICOM medical reconstruction image data to obtain medical reconstruction image resolution level data;

According to the embodiment of the invention, the reconstructed DICOM medical reconstruction image data is detected by using an image resolution detection method, so that the spatial resolution level condition of the medical reconstruction image is estimated, the definition and detail degree of the reconstruction image are ensured, and finally the medical reconstruction image resolution level data is obtained.

Step S44: the quality correction evaluation analysis is carried out on the DICOM medical reconstruction image data by combining the medical reconstruction image structure accuracy status data and the medical reconstruction image resolution level data, so as to obtain a medical reconstruction image quality correction factor;

According to the embodiment of the invention, the image structure accuracy level in the medical reconstruction image structure accuracy condition data and the resolution level in the medical reconstruction image resolution level data are combined to analyze the corresponding DICOM medical reconstruction image, so that the quality problem and improvement space of the DICOM medical reconstruction image are evaluated by comprehensively considering the structural integrity and the resolution, the corresponding quality correction factor is determined according to the evaluation analysis result, and finally the medical reconstruction image quality correction factor is obtained.

Step S45: and adjusting and optimizing the DICOM medical reconstruction image data according to the medical reconstruction image quality correction factors to obtain a medical reconstruction image optimization result.

According to the embodiment of the invention, the quality correction factors of the medical reconstructed image obtained through evaluation are used, corresponding image processing technologies (including image enhancement, noise removal, sharpening processing and the like) are used for adjusting and optimizing the DICOM medical reconstructed image data aiming at the existing quality problems and improvement spaces, so that key indexes such as definition, contrast and the like of the reconstructed medical image are improved, the reconstructed medical image has better analysis usability, and finally the medical reconstructed image optimization result is obtained.

According to the invention, firstly, the equivalent image reconstruction is carried out on the image data of the pixel sampling point region by combining the obtained equivalent surfaces of the DICOM medical image pixels, so that the medical image data reconstruction of the corresponding pixel sampling point region can be effectively converted into the DICOM medical reconstruction image data, and the step is essentially to reconstruct and convert the medical image data corresponding to the DICOM medical image pixel sampling points into the image format which accords with more details, so that more detailed medical image anatomical information can be provided for doctors, and medical images are easier to manage, transmit and read. In the reconstruction process, important structural information and pixel values can be reserved, so that basic data is provided for the subsequent analysis processing process. Secondly, by detecting the structural accuracy of the DICOM medical reconstruction image data, whether the structure of the medical reconstruction image is consistent with the original medical image data can be objectively evaluated, and the detection is helpful for finding out possible medical image reconstruction deviation or distortion, so that the credibility and accuracy of the medical image are improved, reliable image information is provided for doctors, and future processing work is guided better. Then, by performing image resolution detection on the DICOM medical reconstruction image data, the spatial resolution level of the medical reconstruction image can be estimated, which is one of important indexes for estimating the image quality, and the high-resolution image can present clearer detailed information, which is helpful for doctors to accurately analyze the corresponding medical image data, so that it is important to ensure the good resolution of the reconstructed medical image. And then, by combining the accuracy condition data of the medical reconstruction image structure and the image resolution level data to carry out quality correction evaluation analysis, the quality condition of the medical reconstruction image can be comprehensively evaluated, and the comprehensive evaluation considers two aspects of structural integrity and resolution, provides guidance and basis for further optimizing the image quality, and ensures the reliability and accuracy of the medical image data. Finally, the DICOM medical reconstruction image data is adjusted and optimized according to the medical reconstruction image quality correction factors, and targeted improvement of image quality can be achieved. By adjusting parameters or adopting an image enhancement technology, key indexes such as definition, contrast and the like of the reconstructed medical image can be improved, so that the reconstructed medical image has more analysis usability, more reliable image support is provided for doctors, and more comprehensive medical information is provided.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The medical image reconstruction method based on the DICOM image is characterized by comprising the following steps of:

Step S1, including:

Step S11: obtaining DICOM medical image data;

step S16: according to the medical image pixel gray level association relation data, carrying out pixel structure analysis on different types of medical image pixel distribution status data to obtain medical image pixel structure information data;

step S2, including:

Step S25, including:

Step S255: performing influence correction analysis on image slice thickness distribution data of the sampling point region according to the image slice thickness-characteristic influence relation data to obtain image slice thickness level data of the sampling point region;

2. The DICOM image-based medical image reconstruction method according to claim 1, wherein the step S12 comprises the steps of:

3. The medical image reconstruction method according to claim 2, wherein the image association degree calculation formula in step S123 is specifically:

；

In the method, in the process of the invention, For DICOM medical image pixel association degree value,/>As an exponential function,/>For the pixel abscissa parameter of the pixel point in the DICOM medical image corresponding to the DICOM medical image metadata,/>For the pixel abscissa parameter of the DICOM medical image metadata corresponding to the reference point in the DICOM medical image,/>Is the standard deviation of pixel abscissa distribution,/>For the pixel ordinate parameter of the pixel point in the DICOM medical image corresponding to the DICOM medical image metadata,/>For the pixel ordinate parameter of the DICOM medical image metadata corresponding to the reference point in the DICOM medical image,/>Standard deviation of pixel ordinate distribution,/>For the pixel coordinate correlation function,/>Weight is adjusted for DICOM medical image gray level distribution,/>Weight adjustment for DICOM medical image brightness level,/>Influencing the associated parameters for the pixel abscissa distribution,/>Influence the associated parameters for the pixel ordinate distribution,/>And the correction value is the correction value of the DICOM medical image pixel association degree value.

4. The DICOM image-based medical image reconstruction method according to claim 1, wherein the step S3 comprises the steps of:

5. The method of claim 4, wherein the image layer thickness calculation formula in step S32 is specifically:

；

6. The DICOM image-based medical image reconstruction method according to claim 1, wherein the step S4 comprises the steps of:

7. A DICOM image based medical image reconstruction system for performing the DICOM image based medical image reconstruction method of claim 1, the DICOM image based medical image reconstruction system comprising:

The medical image pixel structure analysis module is used for acquiring DICOM medical image data; performing key element mining analysis on the DICOM medical image data according to the corresponding DICOM prefix to obtain medical image key element information data; extracting pixel data and image type data from the key element information data of the medical image to obtain DICOM medical image pixel data and DICOM medical image type data; performing pixel distribution statistical analysis on the DICOM medical image pixel data based on the DICOM medical image type data to obtain different types of medical image pixel distribution status data; performing gray level conversion processing on the DICOM medical image pixel data to obtain medical image pixel gray level data; gray level association analysis is carried out on the gray level data of the pixels of the medical image by using a gray level co-occurrence matrix technology, so as to obtain gray level association relation data of the pixels of the medical image; according to the medical image pixel gray level association relation data, carrying out pixel structure analysis on different types of medical image pixel distribution status data to obtain medical image pixel structure information data;

The pixel point image layer thickness horizontal analysis module is used for carrying out pixel point extraction processing on the DICOM medical image data to obtain DICOM medical image pixels; performing pixel point sampling processing on the DICOM medical image pixel points according to a preset sampling density frequency to obtain DICOM medical image pixel sampling points; performing region segmentation on the DICOM medical image data based on the DICOM medical image pixel sampling points to obtain a pixel sampling point region medical image segmentation result; performing region edge optimization on the medical image segmentation result of the pixel sampling point region to obtain image data of the pixel sampling point region; image slicing processing is carried out on the image data of the pixel sampling point region to obtain image slice surface data of the pixel sampling point region; performing image slice thickness measurement on the image slice surface data of the pixel sampling point region based on the medical image pixel structure information data to obtain image slice thickness distribution data of the sampling point region; performing feature analysis on the image slice surface data of the pixel sampling point region to obtain the image slice surface feature data of the sampling point region; performing layer thickness influence analysis on the image slice layer thickness distribution data of the sampling point region based on the image slice surface characteristic data of the sampling point region to obtain image slice surface layer thickness-characteristic influence relation data; performing influence correction analysis on image slice thickness distribution data of the sampling point region according to the image slice thickness-characteristic influence relation data to obtain image slice thickness level data of the sampling point region;