Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Three-dimensional broken bone segmentation method and device based on deep learning

A technology of deep learning and bone crushing, applied in the field of digital medicine, can solve the problems of low degree of automation, time-consuming and labor-intensive, unable to meet the actual needs of 3D, and achieve the effect of saving manpower, reducing difficulty, and reducing preoperative planning time

Active Publication Date: 2020-07-10
HOHAI UNIV CHANGZHOU
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Under the existing conditions, the extraction and segmentation of cortical bone on the surface of crushed bone is generally done manually by doctors with the help of 3D digital software, such as Mimics and 3-matic, based on existing medical knowledge and clinical experience. Time-consuming and labor-intensive, unable to meet the actual needs of 3D preoperative planning

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Three-dimensional broken bone segmentation method and device based on deep learning
  • Three-dimensional broken bone segmentation method and device based on deep learning
  • Three-dimensional broken bone segmentation method and device based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0048] combine figure 1 The present embodiment shown provides an overall flow chart of a three-dimensional bone fragment segmentation method based on deep learning, including:

[0049] Step 1: Build a PointNet++ deep neural network, which includes:

[0050] S1.1 The feature extraction part consists of a sampling layer, a combination layer and a feature extraction layer. Each layer extracts point sets of multiple neighborhood ranges. PointNet is used as the feature extraction structure to extract local correlation features. As the layers increase, the perception The field increases, the number of feature points decreases, and each feature point contains more and more information.

[0051] The principle of PointNet++ is as follows: the point set is divided into overlapping local areas by the distance measure of the underlying space, and the local features of the captured fine geometric structure are extracted from the small area; these local features are further grouped into la...

Embodiment 2

[0088] On the basis of the above embodiments, a 3D bone fragment segmentation method based on deep learning includes: extracting vertex coordinates and vertex normal vectors based on the acquired 3D bone fragment mesh model, and generating a bone fragment point cloud model; The cloud model is input to the pre-trained PointNet++ deep neural network to predict the vertex bone fragment label probability of the bone fragment point cloud model, which includes the label probability of cortical bone and the label probability of cancellous bone; the obtained The label probability of the vertex cortical bone and the label probability of the cancellous bone are mapped to the corresponding three-dimensional bone fragment mesh model, and the graph cut method is used to further optimize the bone fragment segmentation results.

[0089] A specific embodiment includes constructing a CT medical scan image to be subjected to three-dimensional bone fragment segmentation into a three-dimensional b...

Embodiment 3

[0099] Embodiment 3, a three-dimensional bone fragment segmentation device based on deep learning, including a bone fragment point cloud model generation module, a label probability output module, and a bone fragment segmentation module; wherein

[0100] The point cloud model generation module is used for extracting vertex coordinates and vertex normal vectors based on the acquired three-dimensional bone fragment mesh model to generate a bone fragment point cloud model;

[0101] The PointNet++ deep neural network module is used to input the generated bone fragment point cloud model to the pre-trained PointNet++ deep neural network to predict the probability of the vertex bone fragment label of the bone fragment point cloud model, and the bone fragment label probability includes cortical bone The label probability of and the label probability of cancellous bone;

[0102] The bone fragment segmentation module is configured to map the obtained label probability of the vertex cort...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a three-dimensional broken bone segmentation method and device based on deep learning, and the method comprises the steps: extracting a vertex coordinate and a vertex normal vector based on an obtained three-dimensional broken bone grid model, and generating a broken bone point cloud model; inputting the generated broken bone point cloud model into a pre-trained Point Net ++ deep neural network, mapping the obtained vertex broken bone label probability to a corresponding three-dimensional broken bone grid model, and further performing segmentation optimization on the three-dimensional broken bone grid model by utilizing a graph cutting method to obtain a final broken bone segmentation result. According to the invention, the Point Net + + deep neural network in geometric deep learning is adopted to predict classification marks of broken cortical bones and cancellous bones; PointNet + + processes a point set sampled in a measurement space in a layering mode, a fine geometric structure captured by local features can be extracted, and broken cortical bone and cancellous bone segmentation is well achieved; and a segmentation result is improved by utilizing a graph cutting method according to the smoothness degree between the triangular patches, so the broken bone segmentation efficiency and the automation degree are improved.

Description

technical field [0001] The invention belongs to the field of digital medical treatment, and in particular relates to a three-dimensional bone fragment segmentation method and device based on deep learning. Background technique [0002] With the rapid development of digital medicine, the application of digital technology in surgery is becoming more and more important. Cortical bone extraction and bone fragment segmentation, as an interdisciplinary field between computer science and biomedicine, is a special application in digital medicine and plays an important role in computer-aided 3D preoperative planning. Digital 3D preoperative planning can help doctors effectively overcome visual limitations, improve the accuracy of data measurement, and make diagnosis more accurate and efficient. Among them, surface cortical bone extraction and bone fragment segmentation are in urgent need in preoperative planning. It can help doctors correctly extract medical semantic parameters, acc...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06T7/00G06T7/11G06T17/20
CPCG06T7/0012G06T7/11G06T17/20G06T2207/10081G06T2207/10028G06T2207/20081G06T2207/20084G06T2207/30008
Inventor 蒋俊锋孙晓莉黄瑞陈正鸣何坤金
Owner HOHAI UNIV CHANGZHOU
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com