Search

CN-121999119-A - Digital modeling method, device, computing equipment and medium for bone microstructure

CN121999119ACN 121999119 ACN121999119 ACN 121999119ACN-121999119-A

Abstract

The invention discloses a digital modeling method, a device, calculation equipment and a medium of a bone microstructure, wherein the method comprises the steps of obtaining CT or MRI three-dimensional image data of the bone microstructure based on high-resolution imaging equipment, segmenting a bone tissue region from the three-dimensional image data, extracting bone microstructure features from the bone tissue region, wherein the bone microstructure features comprise the number of bone trabecular nodes, the number of connecting rods corresponding to the bone trabecular nodes, the porosity, the thickness of the bone trabecular and the distance between bone trabeculae, and constructing a ball and stick model of the bone microstructure by using balls to represent the nodes of the bone trabecular and using sticks to represent the connecting rods between the bone trabecular according to the bone microstructure features. The proposal can better understand the biological characteristics and mechanical characteristics of bones by constructing a model of bone microstructure, and provides more scientific and effective basis for preventing, diagnosing and treating bone diseases.

Inventors

  • YANG HUIFANG
  • XU XIAO
  • ZHANG GUOHAO

Assignees

  • 北京大学口腔医学院

Dates

Publication Date
20260508
Application Date
20241106

Claims (11)

  1. 1. A method for digitally modeling a bone microstructure, the method comprising: Acquiring CT or MRI three-dimensional image data of a bone tissue microstructure based on high-resolution imaging equipment; Dividing a bone tissue region from the three-dimensional image data, and extracting bone microstructure features from the bone tissue region, wherein the bone microstructure features comprise the number of bone trabecular nodes, the number of connecting rods corresponding to the bone trabecular nodes, the porosity, the thickness of bone trabecular and the interval between bone trabeculae; According to the bone microstructure characteristics, the nodes of the bone trabeculae are represented by balls, the rods represent connecting rods among the bone trabeculae, and a ball rod model of the bone microstructure is constructed.
  2. 2. The method of claim 1, further comprising, after acquiring CT or MRI three-dimensional image data of the bone tissue microstructure based on the high resolution imaging device: and (3) preprocessing at least one of noise removal, image enhancement or binarization processing is carried out on the three-dimensional image data.
  3. 3. The method according to claim 1, further comprising skeletonizing the three-dimensional image data, specifically comprising: dividing the three-dimensional image data of the trabecula to extract the boundary of bone tissue; converting the trabecula into skeleton lines by using a skeletonizing algorithm based on centerline extraction; The skeleton lines are represented in the form of nodes and links.
  4. 4. The method of claim 3, wherein skeletonizing the three-dimensional image data further comprises: Key morphological characteristics of the crossing points and bifurcation points of the bone trabeculae are reserved; Identifying nodes of the trabeculae and generating links connecting the nodes, the nodes including intersections and endpoints; A club model is formed based on the nodes and links for mechanical analysis of bone tissue.
  5. 5. The method according to claim 4, wherein the method further comprises: Extracting geometric and mechanical parameters from the club model through geometric and mechanical analysis, and determining the connection quantity of balls, wherein the parameters comprise the length, the direction and the thickness of the club; and according to the parameters and the mechanical analysis results, evaluating the state of the bone microstructure, and carrying out osteoporosis analysis, osteogenesis situation analysis or bone defect analysis.
  6. 6. The method of claim 1, further comprising, after constructing the club model of the bone microstructure: carrying out multi-scale analysis on the three-dimensional image data by utilizing PUMA software, and identifying and dividing the structure of the trabecula; further dividing the trabecula by OpenPNM software to construct a pore network model; determining the topology structure and the attribute of the pore network model, including the connection relation and the porosity of the bone trabecula; and analyzing and evaluating the state of the bone microstructure according to the bone microstructure characteristics, the ball and rod model or the pore network model, wherein the analysis comprises osteoporosis analysis, osteogenesis situation analysis, bone defect analysis or mechanical analysis.
  7. 7. The method of claim 1, wherein segmenting the bone tissue region from the three-dimensional image data comprises: performing preliminary segmentation on the three-dimensional image data by using a threshold method or a pre-trained deep learning model so as to identify cortical bone and cancellous bone regions; refining the segmented boundaries by using a conditional random field algorithm or a graph cut algorithm; further optimizing the result of the segmentation using edge detection or texture analysis; And adjusting parameters of the deep learning model through an iterative process so as to improve the segmentation accuracy.
  8. 8. The method of claim 7, wherein refining the boundaries of the segmentation using a conditional random field algorithm or a graph cut algorithm includes utilizing spatial proximity between pixels and context information to improve segmentation consistency and accuracy and/or, Further optimizing the results of the segmentation using edge detection or texture analysis includes analytically optimizing the structure of the segmentation with edge strength, texture features, and shape features to enhance the robustness of the segmentation, and/or, Adjusting parameters of the deep learning model through an iterative process to improve segmentation accuracy includes dividing a training process of the deep learning model into a training phase and a verification phase to optimize model parameters and reduce overfitting, and/or, The deep learning model comprises a convolutional neural network, a U-Net network or a Mask R-CNN network.
  9. 9. A digital modeling device for bone microstructure is characterized in that, the device for digitally modeling bone microstructure comprises: the data acquisition module is suitable for acquiring CT or MRI three-dimensional image data of the bone tissue microstructure based on the high-resolution imaging equipment; the feature extraction module is suitable for segmenting a bone tissue region from the three-dimensional image data and extracting bone microstructure features from the bone tissue region, wherein the bone microstructure features comprise the number of bone trabecular nodes, the number of connecting rods corresponding to the bone trabecular nodes, the porosity, the thickness of bone trabecular and the spacing between bone trabeculae; and the model construction module is suitable for constructing a ball and rod model of the bone microstructure according to the bone microstructure characteristics, wherein the ball represents the node of the bone trabecula, and the rod represents the connecting rod between the bone trabecula.
  10. 10. A computing device comprising a processor and a memory arranged to store computer executable instructions that when executed cause the processor to perform a method of digitally modeling a bone microstructure according to any of claims 1-8.
  11. 11. A computer readable storage medium, characterized in that it stores one or more programs, which when executed by a processor, implement a method of digital modeling of bone microstructures according to any of claims 1-8.

Description

Digital modeling method, device, computing equipment and medium for bone microstructure Technical Field The invention relates to the technical field of modeling, in particular to a digital modeling method, a digital modeling device, a digital modeling computing device and a digital modeling medium for a bone microstructure. Background The microstructure of bone refers to the spatial arrangement, interaction and specific morphology formed by the internal components of bone on the micrometer to nanometer scale, has important significance for understanding the mechanical properties, biological functions and response to external stimuli of bone, and has wide application in the fields of biomedical research, clinical diagnosis, biomaterial research, forensics and the like. The microstructure of bone is a complex and delicate system, consisting of a variety of components and cells together, and determines a variety of functions and properties of bone. The analysis and modeling of the bone microstructure and other researches are not only helpful for better understanding the biological characteristics of bones, but also provide important scientific basis for preventing and treating bone diseases. However, the prior art lacks an effective means of studying the microstructure of bone. Disclosure of Invention The present invention has been made in view of the above problems, and is directed to a method, apparatus, computing device and medium for digitally modeling bone microstructures that overcomes or at least partially solves the above problems. According to one aspect of the present invention, there is provided a method of digitally modeling a bone microstructure, the method comprising: Acquiring CT or MRI three-dimensional image data of a bone tissue microstructure based on high-resolution imaging equipment; Dividing a bone tissue region from the three-dimensional image data, and extracting bone microstructure features from the bone tissue region, wherein the bone microstructure features comprise the number of bone trabecular nodes, the number of connecting rods corresponding to the bone trabecular nodes, the porosity, the thickness of bone trabecular and the interval between bone trabeculae; According to the bone microstructure characteristics, the nodes of the bone trabeculae are represented by balls, the rods represent connecting rods among the bone trabeculae, and a ball rod model of the bone microstructure is constructed. In some embodiments, after acquiring the CT or MRI three-dimensional image data of the bone tissue microstructure based on the high resolution imaging device, further comprises: and (3) preprocessing at least one of noise removal, image enhancement or binarization processing is carried out on the three-dimensional image data. In some embodiments, the method further includes skeletonizing the three-dimensional image data, specifically including: dividing the three-dimensional image data of the trabecula to extract the boundary of bone tissue; converting the trabecula into skeleton lines by using a skeletonizing algorithm based on centerline extraction; The skeleton lines are represented in the form of nodes and links. In some embodiments, skeletonizing the three-dimensional image data further comprises: Key morphological characteristics of the crossing points and bifurcation points of the bone trabeculae are reserved; Identifying nodes of the trabeculae and generating links connecting the nodes, the nodes including intersections and endpoints; A club model is formed based on the nodes and links for mechanical analysis of bone tissue. In some embodiments, the method further comprises: Extracting geometric and mechanical parameters from the club model through geometric and mechanical analysis, and determining the connection quantity of balls, wherein the parameters comprise the length, the direction and the thickness of the club; and according to the parameters and the mechanical analysis results, evaluating the state of the bone microstructure, and carrying out osteoporosis analysis, osteogenesis situation analysis or bone defect analysis. In some embodiments, after constructing the club model of the bone microstructure, further comprising: carrying out multi-scale analysis on the three-dimensional image data by utilizing PUMA software, and identifying and dividing the structure of the trabecula; further dividing the trabecula by OpenPNM software to construct a pore network model; determining the topology structure and the attribute of the pore network model, including the connection relation and the porosity of the bone trabecula; and analyzing and evaluating the state of the bone microstructure according to the bone microstructure characteristics, the ball and rod model or the pore network model, wherein the analysis comprises osteoporosis analysis, osteogenesis situation analysis, bone defect analysis or mechanical analysis. In some embodiments, segmenting the bone tissue region from the th