CN116863068A - Method and system for realizing intelligent segmentation and reconstruction of skeletal muscle system image - Google Patents

Method and system for realizing intelligent segmentation and reconstruction of skeletal muscle system image Download PDF

Info

Publication number
CN116863068A
CN116863068A CN202310642505.7A CN202310642505A CN116863068A CN 116863068 A CN116863068 A CN 116863068A CN 202310642505 A CN202310642505 A CN 202310642505A CN 116863068 A CN116863068 A CN 116863068A
Authority
CN
China
Prior art keywords
bone
slice
femur
density
qct
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202310642505.7A
Other languages
Chinese (zh)
Other versions
CN116863068B (en
Inventor
李建涛
徐成
刘婉姮
王道峰
张武鹏
张�浩
李睿
张子程
唐佩福
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fourth Medical Center General Hospital of Chinese PLA
Original Assignee
Fourth Medical Center General Hospital of Chinese PLA
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fourth Medical Center General Hospital of Chinese PLA filed Critical Fourth Medical Center General Hospital of Chinese PLA
Priority to CN202310642505.7A priority Critical patent/CN116863068B/en
Publication of CN116863068A publication Critical patent/CN116863068A/en
Application granted granted Critical
Publication of CN116863068B publication Critical patent/CN116863068B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
    • A61B6/5211—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10081—Computed x-ray tomography [CT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30008—Bone
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/03—Recognition of patterns in medical or anatomical images
    • G06V2201/033—Recognition of patterns in medical or anatomical images of skeletal patterns

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Databases & Information Systems (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Biophysics (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Geometry (AREA)
  • Optics & Photonics (AREA)
  • Pathology (AREA)
  • Computer Graphics (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Image Processing (AREA)

Abstract

The application relates to a method and a system for realizing intelligent segmentation and reconstruction of a skeletal muscle system image, which are used for acquiring a plurality of femoral bone scanning fault bone slices based on the existing QCT scanning system; according to the bone density distribution diagram on the bone slice, unifying the bone density on the bone slice to obtain a new bone slice with consistent bone density; and reforming each new bone slice, and performing three-dimensional bone modeling according to the new bone slices to obtain a femur bone model. The QCT can be utilized to realize the segmentation of the femur bones and the muscles and the three-dimensional reconstruction of the femur bone model, the reconstructed model separates out the muscle image, the femur bone structure can be clearly displayed, and the medical cooperation cost is reduced.

Description

Method and system for realizing intelligent segmentation and reconstruction of skeletal muscle system image
Technical Field
The disclosure relates to the technical field of image processing, and in particular relates to an intelligent segmentation and reconstruction method, an intelligent segmentation and reconstruction system and electronic equipment for realizing skeletal muscle system images.
Background
The osteoporosis fracture is serious in harm, so that the method has important significance for evaluating the bone strength and further predicting the fracture risk.
In the process of assessing bone strength, for bone image images of a femur or the like, separation of bones and muscles is required, and muscle images are separated from bone images, such as CT slices or nuclear magnetic resonance.
In recent years, with the rise of intelligent network recognition models, such as deep convolutional neural networks, have been widely used for medical image segmentation, so that the segmentation models can be used for conveniently and rapidly processing image images. However, the software application is almost a third party research and development team, and needs to be used with high cost, limited by certain use, and has to be matched with a service provider for application upgrading and management. The convolutional network segmentation model recognizes and segments the bone image, which can cause background operation pressure, delay operation calculation memory and reduce medical efficiency.
The existing bone analysis system such as QCT can perform bone recognition, so that bone analysis such as bone and muscle segmentation can be performed based on the existing equipment, thereby saving the cost of medical research and development cooperation.
Disclosure of Invention
In order to solve the above problems, the present application provides a method, a system and an electronic device for realizing the intelligent segmentation and reconstruction of skeletal muscle system images.
In one aspect of the present application, an intelligent segmentation and reconstruction method for implementing skeletal muscle system images is provided, including the following steps:
obtaining a plurality of femoral scanning fault bone slices;
according to the bone density distribution diagram on the bone slice, unifying the bone density on the bone slice to obtain a new bone slice with consistent bone density;
reforming each new bone slice, and performing three-dimensional bone modeling according to the new bone slice to obtain a femur bone model, including:
sequentially introducing each new bone slice S' (t) into a Quantitative Computed Tomography (QCT) system according to a scanning sequence;
reforming each new bone slice S' (t), and performing three-dimensional bone modeling according to the reformed femur bone slices to obtain a femur bone model M:
M＝∏S′(t)；
the femur bone model M is stored in a background database.
As an optional embodiment of the present application, optionally, obtaining a plurality of femoral scanning tomographic bone slices further comprises:
pre-deploying a feature recognition algorithm on a background server;
scanning and acquiring QCT image data of a femur, and storing the QCT image data into background data;
calculating the QCT image data according to a feature recognition algorithm, recognizing and extracting femur outline features, and obtaining global image outline features in the femur image;
and marking the global image contour features on the original femur image in the QCT image data, and classifying the bone and muscle images to obtain femur contour positions.
As an alternative embodiment of the application, optionally, obtaining a plurality of femoral scanning tomographic bone slices comprises:
setting a scanning sequence;
quantitative computer tomography is carried out on the femur part according to the scanning sequence, and a plurality of femur scanning tomography bone slices are obtained;
the individual femur scan tomographic bone slices are stored.
As an optional embodiment of the present application, optionally, according to the bone density distribution diagram on the bone slice, performing a unification treatment on the bone density on the bone slice to obtain a new bone slice with consistent bone density, including:
quantitatively calculating QCT bone density T of each bone slice according to the quantitative computed tomography result S ) Obtaining a bone density profile of the bone slice;
and carrying out bone density unification treatment on the slice region on the bone density distribution map, wherein the slice region meets the following conditions to obtain a new bone slice with consistent bone density:
T( S )≤α*T Mmax
T( S ) Is the average value of bone density of any slice region;
alpha is a distribution coefficient, and the value is 0.3-1.0;
T Mmax is the maximum bone density value on the current bone density profile.
As an optional embodiment of the present application, optionally, the bone mineral density unification treatment for the cut region includes:
if the bone density mean value T (T) of the current slice region is less than or equal to alpha T Mmax Then T is taken as Mmax Uniformly modifying the bone density mean value T (T) on the current slice region into T as the processing standard of the front slice region Mmax 。
In another aspect of the present application, an intelligent segmentation and reconstruction system for implementing an image of a skeletal muscle system is provided, comprising:
the CT tomography module is used for acquiring a plurality of femoral bone scanning tomography bone slices;
the processing module is used for carrying out unified processing on the bone density on the bone slice according to the bone density distribution diagram on the bone slice to obtain a new bone slice with consistent bone density;
the bone reconstruction module is used for reforming each new bone slice and carrying out three-dimensional bone modeling according to the new bone slices to obtain a femur bone model, and comprises the following steps:
sequentially introducing each new bone slice S' (t) into a Quantitative Computed Tomography (QCT) system according to a scanning sequence;
reforming each new bone slice S' (t), and performing three-dimensional bone modeling according to the reformed femur bone slices to obtain a femur bone model M:
M＝∏S′(t)；
the femur bone model M is stored in a background database.
As an optional embodiment of the present application, optionally, further comprising:
the QCT scanning module is used for scanning and acquiring QCT image data of the femur and storing the QCT image data into background data;
the feature processing module is used for importing the QCT image data into the deep learning model which is deployed in advance on a background server, recognizing and extracting femur contour features, and acquiring global image contour features in femur image images;
and the feature marking module is used for marking the global image contour feature on the original femur image in the QCT image data, and classifying the bone and muscle images to obtain the femur contour position.
As an optional embodiment of the present application, optionally, the processing module includes:
the bone density calculating module is used for quantitatively calculating QCT bone density T of each bone slice according to the quantitative computed tomography result S ) Obtaining a bone density profile of the bone slice;
a distribution map formatting module, configured to perform bone density unification processing on a slice region on the bone density distribution map, where the slice region meets the following conditions, to obtain a new bone slice with consistent bone density, where if the bone density average value T (T) is less than or equal to α×t in the current slice region Mmax Then T is taken as Mmax Uniformly modifying the bone density mean value T (T) on the current slice region into T as the processing standard of the front slice region Mmax ：
T( S )≤α*T Mmax
T( S ) Is the average value of bone density of any slice region;
alpha is a distribution coefficient, and the value is 0.3-1.0;
T Mmax is the maximum bone density value on the current bone density profile.
In another aspect, the present application further provides an electronic device, including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to implement the method of implementing intelligent segmentation and reconstruction of skeletal muscle system images when executing the executable instructions.
The application has the technical effects that:
the application obtains a plurality of femoral bone scanning fault bone slices based on the existing QCT scanning system; according to the bone density distribution diagram on the bone slice, unifying the bone density on the bone slice to obtain a new bone slice with consistent bone density; reforming each new bone slice, and performing three-dimensional bone modeling according to the new bone slice to obtain a femur bone model, including: sequentially introducing each new bone slice S' (t) into a Quantitative Computed Tomography (QCT) system according to a scanning sequence; and reforming each new bone slice S' (t), and performing three-dimensional bone modeling according to the reformed femur bone slices to obtain a femur bone model M. The QCT can be utilized to realize the segmentation of the femur bones and the muscles and the three-dimensional reconstruction of the femur bone model, the reconstructed model separates out the muscle image, the femur bone structure can be clearly displayed, and the medical cooperation cost is reduced.
Meanwhile, part of the segmentation work can be distributed to the original QCT system for processing, so that the operation pressure can be reduced and the efficiency can be improved for a background server.
Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features and aspects of the present disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1 shows a schematic flow chart of an implementation of the present application;
FIG. 2 is a schematic diagram of an application architecture of the application system of the present application;
FIG. 3 is a schematic representation of the contours of the femoral image contours identified and marked in accordance with the present application;
FIG. 4 shows a schematic representation of individual bone slices obtained for a tomographic scan of the present application;
FIG. 5 is a schematic view showing the distribution of tissue on a section of the present application;
FIG. 6 is a schematic view of a bone section of the present application after the bone density is uniform;
fig. 7 shows a schematic application diagram of the electronic device of the present application.
Detailed Description
Various exemplary embodiments, features and aspects of the disclosure will be described in detail below with reference to the drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
The word "exemplary" is used herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.
In addition, numerous specific details are set forth in the following detailed description in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, well known means, elements, and circuits have not been described in detail so as not to obscure the present disclosure.
Example 1
As shown in fig. 1, in one aspect, the present application provides a method for implementing intelligent segmentation and reconstruction of skeletal muscle system images, which includes the following steps:
s1, acquiring a plurality of femoral scanning fault bone slices; the QCT scanning system (mainly adopts a 3DQCT system) can quantitatively perform COM tomography, perform tomography on the femoral bone parts, and the bone density on each femoral scanning tomography can be automatically calculated through an image system of the QCT scanning system. Three-dimensional scanned image of the femoral bone site can be obtained by a 3D QCT system, the image comprising bones and muscles. Bone slices will be stored in sequence.
S2, according to the bone density distribution diagram on the bone slice, unifying the bone density on the bone slice to obtain a new bone slice with consistent bone density; the density distribution on the slice image can be identified in a slice mode, the muscles and bones are identified according to the density distribution, the density distribution image on the slice image where the muscles are positioned is subjected to unified coating treatment according to the bone density, the density point cloud slice image where the muscles are positioned can be uniformly modified into bone density point clouds, and the slice density image where the muscles are positioned is divided, so that a new bone slice is obtained;
the bone density of different areas on a bone slice is different from the density distribution image of the bone slice according to the bone density distribution map, and the density cloud points and the image depth displayed on the density distribution image of the bone slice are different, if modeling is carried out according to the non-uniform density, the bone model obtained later is in the bone area within the same range, and the distribution image depth is different, so that the appearance expression of the model is influenced.
When bone slice tomography is performed in the preamble, a two-dimensional or three-dimensional tomographic image is obtained, a rough bone contour image is obtained, then bone slicing is performed, and then muscles can be separated from bones according to bone density on a slice section, but the muscles still have a certain density shadow. Therefore, the part can be subjected to unified treatment according to the bone density, and the muscle layer outside the bone area is subjected to unified treatment according to the bone density, so that the condition that the external surface expression of the bone model is influenced by muscle shadow when the bone slice is subjected to the recombination model in the follow-up process is avoided.
S3, reforming each new bone slice, and carrying out three-dimensional bone modeling according to the new bone slices to obtain a femur bone model, wherein the method comprises the following steps:
sequentially introducing each new bone slice S' (t) into a Quantitative Computed Tomography (QCT) system according to a scanning sequence;
reforming each new bone slice S' (t), and performing three-dimensional bone modeling according to the reformed femur bone slices to obtain a femur bone model M:
M＝∏S′(t)；
and integrating the new bone slices S' (t), and sequentially carrying out three-dimensional bone recombination to obtain a bone slice assembly with separated muscles, namely a femur bone model. The differentiation is mainly performed by using the difference of the density values of the bone density and the surrounding muscle tissue on the density distribution map.
And (3) carrying out section recombination, carrying out section matching and recombination according to section sequence to obtain a bone model of the combination body, and carrying out parameter adjustment and optimization to obtain the finished expression effect with consistent bone surface density.
The 3D QCT system is capable of integrating new bone slices S' (t), combining the individual slices in order into one femoral bone model M.
And S4, storing the femur skeleton model M into a background database.
As shown in fig. 2, an interaction diagram between the application entities is shown. The fault scanning system is communicated with a hospital background server, the 3D QCT system can be utilized for carrying out preliminary processing on scanned image data, the configured slicing function is utilized for carrying out image slicing, and bone slice image data of each femur scanning fault are sent to the background server for further processing after slicing.
In order to obtain exact femur bone position information, the scheme also adopts a bone image feature recognition algorithm, QCT image data of femur bone parts are obtained at the initial stage by starting a QCT scanning system, the QCT image data are stored in the background, and feature recognition and contour marking are carried out on the QCT image by a feature recognition algorithm deployed in a background processor.
As an optional embodiment of the present application, optionally, obtaining a plurality of femoral scanning tomographic bone slices further comprises:
pre-deploying a feature recognition algorithm on a background server;
scanning and acquiring QCT image data of a femur, and storing the QCT image data into background data;
calculating the QCT image data according to a feature recognition algorithm, recognizing and extracting femur outline features, and obtaining global image outline features in the femur image;
and marking the global image contour features on the original femur image in the QCT image data, and classifying the bone and muscle images to obtain femur contour positions.
As shown in fig. 3, not all femoral bone parts need to be analyzed and bone treated for the scanned femoral part. Therefore, the femur bone image obtained from the current scanning position needs to be subjected to position marking (the position marked by the contour line is used as the target position), and the contour feature of the target bone is identified and extracted from the position marking, so that the characteristic identification algorithm deployed on the background server can be used for identifying and extracting the characteristic global image contour feature (such as the femur tail end contour in fig. 3) of the QCT image, and the global image contour feature of the target position is obtained.
Image contour addressing, identification and marking can be performed through an attention mechanism algorithm or an image contour detection algorithm, and global image contour features in the femur image can be found.
Attention mechanisms, such as calculation and recognition of contour features by attention input key values, etc., can be referred to in the existing, e.g., transducer models.
If an image contour detection algorithm is used, for example, CHAIN_APPROX_SIMPLE may be used.
According to the global image contour features, bones on the original femur image in the QCT image data can be identified, bone and muscle image classification and image segmentation can be carried out, tissue gray level images such as muscles on the outer side of the contour are segmented out, femur gray level images in the contour are stored, and the femur contour position is obtained. The femur image contour is internally bone image and externally muscle image, so that the separation of bones and muscles in one longitudinal direction is realized (but bones and muscles are also present in the transverse direction). For example, as shown in fig. 3, in the current direction, the contour outside the contour is removed, so that there is no muscle gray image in the longitudinal direction, but there is still a muscle gray image in the transverse direction.
As an alternative embodiment of the application, optionally, obtaining a plurality of femoral scanning tomographic bone slices comprises:
setting a scanning sequence;
quantitative computer tomography is carried out on the femur part according to the scanning sequence, and a plurality of femur scanning tomography bone slices are obtained;
the individual femur scan tomographic bone slices are stored.
As shown in fig. 4, the synchronization may adjust the scan mode of the 3D QCT system for quantitative computed tomography, mainly 3D tomography (transverse tomography, needs to be distinguished from the longitudinal orientation described above). The scanning parameters of the scanning faults are set on the 3D QCT system by staff, or the scanning parameters are directly issued to the numbered 3D QCT system through a background after being set by a background manager. A tomographic scan is initiated and bone slices of each femoral scan slice are obtained.
In this embodiment, a background-controlled scanning scheme is preferentially adopted, equipment codes of each 3D QCT system are stored on a background server, a background administrator presets and stores corresponding tomographic parameters, the scanning parameters are issued to the 3D QCT system of the equipment codes from the background in a specified time, and the 3D QCT system receives and executes the scanning parameters to perform tomographic scanning. And for the femur slices generated by each tomographic scan, the serial numbers of each tomographic scan are carried out, and each femur slice is sent and stored in a background database according to the serial numbers, so that the later recombination of each tomographic scan slice is facilitated.
The bone slices generated by each scan will be automatically marked with the serial number of the scan slice and stored by number, and the cut planes of each bone slice will show gray scale images of the femur bone and surrounding muscles. The schematic diagram shown in fig. 5 is a schematic diagram of the tissue structure on the section of the transverse section, and the middle position is the bone tissue. Muscle tissues are arranged on the outer sides of two sides of the skeleton, and on a scanning tomography QCT (full-face tomography) image, different tissue parts of each section can show different distribution densities, so that a bone density distribution map on the section can be simultaneously scanned and acquired through the QCT image.
For each bone slice, density calculation can be performed on the gray scale image where each slice is located according to the QCT system. The outer side of the contour formed by the maximum density value can be used as a gray level image of the muscle tissue from the slice image, and the gray level image of the bone tissue on each slice in the current direction can be obtained by processing the gray level image, such as hiding or deleting, the muscle image. The density profile on the bone slice section can show the contour of each tissue site, thus distinguishing the muscle tissue on the current slice well, and finally leaving the femur bone image.
After the femur bone image is obtained, the bone density of the femur bone image can be calculated based on the QCT system, and the bone density distribution map of the current slice is obtained.
In the femur bone image of the current slice, not only hard bone but also bone marrow gray level image exists, so that the integration treatment is required according to the bone density on the bone slice, and a new bone slice with consistent bone density is obtained.
As an optional embodiment of the present application, optionally, according to the bone density distribution diagram on the bone slice, performing a unification treatment on the bone density on the bone slice to obtain a new bone slice with consistent bone density, including:
quantitatively calculating QCT bone density T of each bone slice according to the quantitative computed tomography result S ) Obtaining a bone density profile of the bone slice;
and carrying out bone density unification treatment on the slice region on the bone density distribution map, wherein the slice region meets the following conditions to obtain a new bone slice with consistent bone density:
T( S )≤α*T Mmax
T( S ) Is the average value of bone density of any slice region;
alpha is a distribution coefficient, and the value is 0.3-1.0;
T Mmax is the maximum bone density value on the current bone density profile.
As shown in FIG. 6, the QCT bone density T (S) of a bone slice can be obtained in real time on a QCT system according to quantitative computed tomography results, and the density value can be obtained by checking the attribute of a certain position of the slice, or the QCT system can output the density distribution map of each bone slice and find the maximum bone density value T on the bone density distribution map Mmax 。
Browsing each section area on the femur bone image of the current section and calculating to obtain the bone density average value T of each section area S ) Judging T% S ) Whether or not to meet T% S )≤α*T Mmax Since the non-bone contour region of the slice region is represented, it is necessary to perform bone density processing on the slice region, and the density values on the grayscale image of the slice region are uniformly set to the maximum bone density value.
Specifically, as an optional embodiment of the present application, optionally, the bone mineral density unification treatment is performed on the cut region, including:
if the bone density mean value T (T) of the current slice region is less than or equal to alpha T Mmax Then T is taken as Mmax Uniformly modifying the bone density mean value T (T) on the current slice region into T as the processing standard of the front slice region Mmax 。
The slice area may be plural on a bone slice, because bone marrow gray matter or the like may be present on a bone slice, and is shown as a different slice area (such as the structure inside the bone tissue of fig. 5) on a slice gray scale. In this case, the whole bone tissue may be a whole slice region, and the gray-scale image processing may be performed according to the maximum bone density.
In this way, each bone slice is subjected to gray level image separation treatment of bone and muscle, only the rest bones (the gray level image in the outline where the maximum bone density is located) are subjected to muscle separation on the transverse section, and meanwhile, bone density unification is carried out, so that a solid bone model is conveniently obtained, and the reconstruction of the three-dimensional bone model is conveniently carried out.
And obtaining each new bone slice S' (t) by adopting the slice gray level processing. And (3) carrying out three-dimensional reformation according to the marked scanning sequence, and reconstructing a three-dimensional model of the current bone part to obtain a complete bone model without muscle tissues.
The background can send each new slice of orderly marks to a 3D QCT system, reform each new bone slice S' (t), and perform three-dimensional bone modeling according to the reformed femur bone slice to obtain a femur bone model M:
M＝∏S′(t)。
the slice reforming is specifically implemented by the 3D QCT system in cooperation with control parameters of the background server, which is not described in detail in this embodiment.
Therefore, the application can perform quantitative computer tomography on the femur bone part based on the existing QCT scanning system, can realize the segmentation of femur bones and muscles and the three-dimensional reconstruction of a femur bone model by using QCT, separates muscle images from the reconstructed model, can clearly display the femur bone structure, and reduces the medical cooperation cost.
It should be noted that although the above has been described as an example, those skilled in the art will appreciate that the present disclosure should not be limited thereto. In fact, the user can flexibly set the device according to the actual application scene, so long as the technical function of the application can be realized according to the technology.
It should be apparent to those skilled in the art that implementing all or part of the above-described embodiments may be accomplished by computer programs to instruct related hardware, and the programs may be stored in a computer readable storage medium, which when executed may include the processes of the embodiments of the controls described above. It will be appreciated by those skilled in the art that implementing all or part of the above-described embodiments may be accomplished by computer programs to instruct related hardware, and the programs may be stored in a computer readable storage medium, which when executed may include the processes of the embodiments of the controls described above. The storage medium may be a magnetic disk, an optical disc, a Read-only memory (ROM), a random access memory (RandomAccessMemory, RAM), a flash memory (flash memory), a hard disk (HDD), or a Solid State Drive (SSD); the storage medium may also comprise a combination of memories of the kind described above.
Example 2
Based on the implementation principle of embodiment 1, another aspect of the present application proposes an intelligent segmentation and reconstruction system for implementing an image of a skeletal muscle system, including:
the CT tomography module is used for acquiring a plurality of femoral bone scanning tomography bone slices;
the processing module is used for carrying out unified processing on the bone density on the bone slice according to the bone density distribution diagram on the bone slice to obtain a new bone slice with consistent bone density;
the bone reconstruction module is used for reforming each new bone slice and carrying out three-dimensional bone modeling according to the new bone slices to obtain a femur bone model, and comprises the following steps:
sequentially introducing each new bone slice S' (t) into a Quantitative Computed Tomography (QCT) system according to a scanning sequence;
reforming each new bone slice S' (t), and performing three-dimensional bone modeling according to the reformed femur bone slices to obtain a femur bone model M:
M＝∏S′(t)；
the femur bone model M is stored in a background database.
As an optional embodiment of the present application, optionally, further comprising:
the QCT scanning module is used for scanning and acquiring QCT image data of the femur and storing the QCT image data into background data;
the feature processing module is used for importing the QCT image data into the deep learning model which is deployed in advance on a background server, recognizing and extracting femur contour features, and acquiring global image contour features in femur image images;
and the feature marking module is used for marking the global image contour feature on the original femur image in the QCT image data, and classifying the bone and muscle images to obtain the femur contour position.
As an optional embodiment of the present application, optionally, the processing module includes:
the bone density calculating module is used for quantitatively calculating QCT bone density T of each bone slice according to the quantitative computed tomography result S ) Obtaining a bone density profile of the bone slice;
a distribution map formatting module, configured to perform bone density unification processing on a slice region on the bone density distribution map, where the slice region meets the following conditions, to obtain a new bone slice with consistent bone density, where if the bone density average value T (T) is less than or equal to α×t in the current slice region Mmax Then T is taken as Mmax Uniformly modifying the bone density mean value T (T) on the current slice region into T as the processing standard of the front slice region Mmax ：
T( S )≤α*T Mmax
T( S ) Is the average value of bone density of any slice region;
alpha is a distribution coefficient, and the value is 0.3-1.0;
T Mmax is the maximum bone density value on the current bone density profile.
The operation functions and principles of the above respective modules are specifically described with reference to embodiment 1.
The modules or steps of the application described above may be implemented in a general-purpose computing device, they may be centralized in a single computing device, or distributed across a network of computing devices, or they may alternatively be implemented in program code executable by a computing device, such that they may be stored in a memory device and executed by a computing device, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps within them may be fabricated into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.
Example 3
As shown in fig. 7, in still another aspect, the present application further provides an electronic device, including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to implement the method of implementing intelligent segmentation and reconstruction of skeletal muscle system images when executing the executable instructions.
Embodiments of the present disclosure provide for an electronic device that includes a processor and a memory for storing processor-executable instructions. Wherein the processor is configured to implement any of the foregoing methods of implementing the intelligent segmentation and reconstruction of skeletal muscle system images when executing the executable instructions.
Here, it should be noted that the number of processors may be one or more. Meanwhile, in the electronic device of the embodiment of the disclosure, an input device and an output device may be further included. The processor, the memory, the input device, and the output device may be connected by a bus, or may be connected by other means, which is not specifically limited herein.
The memory is a computer-readable storage medium that can be used to store software programs, computer-executable programs, and various modules, such as: the embodiment of the disclosure relates to a program or a module corresponding to an intelligent segmentation and reconstruction method for realizing skeletal muscle system images. The processor executes various functional applications and data processing of the electronic device by running software programs or modules stored in the memory.
The input device may be used to receive an input number or signal. Wherein the signal may be a key signal generated in connection with user settings of the device/terminal/server and function control. The output means may comprise a display device such as a display screen.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the technical improvement of the prior art in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims (9)

1. An intelligent segmentation and reconstruction method for realizing skeleton muscle system images is characterized by comprising the following steps:
obtaining a plurality of femoral scanning fault bone slices; wherein the femur scan tomographic bone slice is determined by quantitative computed tomography of the femoral component;
according to the bone density distribution diagram on the bone slice, unifying the bone density on the bone slice to obtain a new bone slice with consistent bone density;
reforming each new bone slice, and performing three-dimensional bone modeling according to the new bone slice to obtain a femur bone model, including:
sequentially introducing each new bone slice S' (t) into a Quantitative Computed Tomography (QCT) system according to a scanning sequence;
reforming each new bone slice S' (t), and performing three-dimensional bone modeling according to the reformed femur bone slices to obtain a femur bone model M:
M∏S′(t)；
the femur bone model M is stored in a background database.
2. The method for intelligent segmentation and reconstruction of images of the skeletal muscle system according to claim 1, wherein obtaining a plurality of femoral scan tomographic bone slices further comprises:
pre-deploying a feature recognition algorithm on a background server;
scanning and acquiring QCT image data of a femur, and storing the QCT image data into background data;
calculating the QCT image data according to a feature recognition algorithm, recognizing and extracting femur outline features, and obtaining global image outline features in the femur image;
and marking the global image contour features on the original femur image in the QCT image data, and classifying the bone and muscle images to obtain femur contour positions.
3. The method for intelligent segmentation and reconstruction of images of the skeletal muscle system according to claim 2, wherein obtaining a plurality of femoral scan tomographic bone slices comprises:
setting a scanning sequence;
quantitative computer tomography is carried out on the femur part according to the scanning sequence, and a plurality of femur scanning tomography bone slices are obtained;
the individual femur scan tomographic bone slices are stored.
4. The method for intelligent segmentation and reconstruction of skeletal muscle system images according to claim 1, wherein unifying bone densities on the bone slices according to the bone density distribution map on the bone slices to obtain new bone slices with consistent bone densities, comprising:
quantitatively calculating QCT bone density T of each bone slice according to the quantitative computed tomography result S ) Obtaining a bone density profile of the bone slice;
and carrying out bone density unification treatment on the slice region on the bone density distribution map, wherein the slice region meets the following conditions to obtain a new bone slice with consistent bone density:
T( S )≤α*T Mmax
T( S ) Is the average value of bone density of any slice region;
alpha is a distribution coefficient, and the value is 0.3-1.0;
T Mmax is the maximum bone density value on the current bone density profile.
5. The method for intelligent segmentation and reconstruction of skeletal muscle system images according to claim 4, wherein the bone mineral density unification of the segmented regions comprises:
if the bone density mean value T (T) of the current slice region is less than or equal to alpha T Mmax Then T is taken as Mmax Uniformly modifying the bone density mean value T (T) on the current slice region into T as the processing standard of the front slice region Mmax 。
6. An intelligent segmentation and reconstruction system for realizing skeletal muscle system images, comprising:
the CT tomography module is used for acquiring a plurality of femoral bone scanning tomography bone slices;
the processing module is used for carrying out unified processing on the bone density on the bone slice according to the bone density distribution diagram on the bone slice to obtain a new bone slice with consistent bone density;
the bone reconstruction module is used for reforming each new bone slice and carrying out three-dimensional bone modeling according to the new bone slices to obtain a femur bone model, and comprises the following steps:
sequentially introducing each new bone slice S' (t) into a Quantitative Computed Tomography (QCT) system according to a scanning sequence;
reforming each new bone slice S' (t), and performing three-dimensional bone modeling according to the reformed femur bone slices to obtain a femur bone model M:
M∏S′(t)；
the femur bone model M is stored in a background database.
7. The system of claim 6, further comprising:
the QCT scanning module is used for scanning and acquiring QCT image data of the femur and storing the QCT image data into background data;
the feature processing module is used for importing the QCT image data into the deep learning model which is deployed in advance on a background server, recognizing and extracting femur contour features, and acquiring global image contour features in femur image images;
and the feature marking module is used for marking the global image contour feature on the original femur image in the QCT image data, and classifying the bone and muscle images to obtain the femur contour position.
8. The system of claim 6, wherein the processing module comprises:
the bone density calculating module is used for quantitatively calculating QCT bone density T of each bone slice according to the quantitative computed tomography result S ) Obtaining a bone density profile of the bone slice;
a distribution map formatting module, configured to perform bone density unification processing on a slice region on the bone density distribution map, where the slice region meets the following conditions, to obtain a new bone slice with consistent bone density, where if the bone density average value T (T) is less than or equal to α×t in the current slice region Mmax Then T is taken as Mmax Uniformly modifying the bone density mean value T (T) on the current slice region into T as the processing standard of the front slice region Mmax ：
T( S )≤α*T Mmax
T( S ) Is the average value of bone density of any slice region;
alpha is a distribution coefficient, and the value is 0.3-1.0;
T Mmax is the maximum bone density value on the current bone density profile.
9. An electronic device, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to implement the method of intelligent segmentation and reconstruction of skeletal muscle system images of any one of claims 1-5 when executing the executable instructions.
CN202310642505.7A 2023-06-01 2023-06-01 Method and system for realizing intelligent segmentation and reconstruction of skeletal muscle system image Active CN116863068B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310642505.7A CN116863068B (en) 2023-06-01 2023-06-01 Method and system for realizing intelligent segmentation and reconstruction of skeletal muscle system image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310642505.7A CN116863068B (en) 2023-06-01 2023-06-01 Method and system for realizing intelligent segmentation and reconstruction of skeletal muscle system image

Publications (2)

Publication Number Publication Date
CN116863068A true CN116863068A (en) 2023-10-10
CN116863068B CN116863068B (en) 2024-02-13

Family

ID=88222357

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310642505.7A Active CN116863068B (en) 2023-06-01 2023-06-01 Method and system for realizing intelligent segmentation and reconstruction of skeletal muscle system image

Country Status (1)

Country Link
CN (1) CN116863068B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101958079A (en) * 2010-07-22 2011-01-26 西北师范大学 Positioning model of channel acupuncture point in three-dimensional virtual human anatomy texture and application thereof
KR101757642B1 (en) * 2016-07-20 2017-07-13 (주)레벨소프트 Apparatus and method for 3d face modeling
CN107320221A (en) * 2017-06-21 2017-11-07 四川大学 A kind of lopsided knee joint skeleton model preparation method based on 3D printing technique
KR20180008355A (en) * 2017-09-01 2018-01-24 한국과학기술원 Method and apparatus for high-resulation of 3d skeletal image
CN109830288A (en) * 2019-03-19 2019-05-31 博志生物科技(深圳)有限公司 Bone density three-dimensional visualization electronic report system
CN110570515A (en) * 2019-09-03 2019-12-13 天津工业大学 method for carrying out human skeleton three-dimensional modeling by utilizing CT (computed tomography) image
CN112037234A (en) * 2020-08-19 2020-12-04 大连理工大学 Method for separating liver from muscle in CT image by using bone information
KR20220029801A (en) * 2020-08-27 2022-03-10 주식회사 쓰리디코리아 3D printing method of a dummy for simulation with skeletons similar to the skeleton human body
CN114419186A (en) * 2021-12-15 2022-04-29 武汉中海庭数据技术有限公司 Method for generating road bifurcation merging points based on original track

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101958079A (en) * 2010-07-22 2011-01-26 西北师范大学 Positioning model of channel acupuncture point in three-dimensional virtual human anatomy texture and application thereof
KR101757642B1 (en) * 2016-07-20 2017-07-13 (주)레벨소프트 Apparatus and method for 3d face modeling
CN107320221A (en) * 2017-06-21 2017-11-07 四川大学 A kind of lopsided knee joint skeleton model preparation method based on 3D printing technique
KR20180008355A (en) * 2017-09-01 2018-01-24 한국과학기술원 Method and apparatus for high-resulation of 3d skeletal image
CN109830288A (en) * 2019-03-19 2019-05-31 博志生物科技(深圳)有限公司 Bone density three-dimensional visualization electronic report system
CN110570515A (en) * 2019-09-03 2019-12-13 天津工业大学 method for carrying out human skeleton three-dimensional modeling by utilizing CT (computed tomography) image
CN112037234A (en) * 2020-08-19 2020-12-04 大连理工大学 Method for separating liver from muscle in CT image by using bone information
KR20220029801A (en) * 2020-08-27 2022-03-10 주식회사 쓰리디코리아 3D printing method of a dummy for simulation with skeletons similar to the skeleton human body
CN114419186A (en) * 2021-12-15 2022-04-29 武汉中海庭数据技术有限公司 Method for generating road bifurcation merging points based on original track

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
X SHERRY LIU: "Bone density, geometry, microstructure, and stiffness: Relationships between peripheral and central skeletal sites assessed by DXA, HR-pQCT, and cQCT in premenopausal women", 《HTTPS://DOI.ORG/10.1002/JBMR.111》 *
明洁: "定量CT测量骨密度研究进展", 《四川解剖学杂志》, vol. 15, no. 2, pages 22 - 24 *
肖建林;刘潼;秦彦国;李学州;刘亮;王晓玲;王金成;高忠礼;: "基于CT图像三维重建测量股骨外翻角", 中国实验诊断学, no. 12, pages 55 - 57 *

Also Published As

Publication number Publication date
CN116863068B (en) 2024-02-13

Similar Documents

Publication Publication Date Title
CN110689038B (en) Training method and device for neural network model and medical image processing system
PS et al. Extraction of texture features using GLCM and shape features using connected regions
CN110310287B (en) Automatic organ-at-risk delineation method, equipment and storage medium based on neural network
WO2019200753A1 (en) Lesion detection method, device, computer apparatus and storage medium
CN103069455B (en) Organ-specific enhancement filter for robust segmentation of medical images
CN110648337A (en) Hip joint segmentation method, hip joint segmentation device, electronic apparatus, and storage medium
EP3410393A1 (en) Comparing medical images
CN110070540B (en) Image generation method and device, computer equipment and storage medium
US20070031020A1 (en) Method and apparatus for intracerebral hemorrhage lesion segmentation
Tomakova et al. Automatic fluorography segmentation method based on histogram of brightness submission in sliding window
CN110544245A (en) Image processing method, image processing device, computer-readable storage medium and electronic equipment
CN111915584A (en) Focus follow-up assessment method and system based on CT (computed tomography) image
Chae et al. Automatic lung segmentation for large-scale medical image management
CN114841947A (en) Method and device for multi-scale feature extraction and prognosis analysis of H & E staining pathological image tumor region
CN114450716A (en) Image processing for stroke characterization
CN114332132A (en) Image segmentation method and device and computer equipment
US11257214B2 (en) Two-stage variational image segmentation of medical images using fracture mechanics
CN112102351A (en) Medical image analysis method and device, electronic equipment and readable storage medium
CN111128348A (en) Medical image processing method, device, storage medium and computer equipment
CN112200780B (en) Bone tissue positioning method, device, computer equipment and storage medium
CN116863068B (en) Method and system for realizing intelligent segmentation and reconstruction of skeletal muscle system image
CN111899273A (en) Image segmentation method, computer device and storage medium
Arica et al. A plain segmentation algorithm utilizing region growing technique for automatic partitioning of computed tomography liver images
CN113327235B (en) Image processing method and device based on CT image sarcopenia evaluation
CN116615151A (en) Method for determining the screw trajectory of a pedicle screw

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CB03 Change of inventor or designer information

Inventor after: Zhang Zicheng

Inventor after: Li Jiantao

Inventor after: Xu Cheng

Inventor after: Liu Wanheng

Inventor after: Wang Daofeng

Inventor after: Zhang Wupeng

Inventor after: Zhang Hao

Inventor after: Li Rui

Inventor after: Tang Peifu

Inventor before: Li Jiantao

Inventor before: Xu Cheng

Inventor before: Liu Wanheng

Inventor before: Wang Daofeng

Inventor before: Zhang Wupeng

Inventor before: Zhang Hao

Inventor before: Li Rui

Inventor before: Zhang Zicheng

Inventor before: Tang Peifu