Vertebral plasty bone cement leakage detection method and system and storage medium

A technology of vertebroplasty and detection methods, applied in the fields of deep learning, computer vision, and intelligent medical image analysis technology, can solve problems such as easy judgment errors, low efficiency, and difficulty in ensuring the accuracy and consistency of detection and analysis results. Achieve the effect of reducing prediction time and solving poor consistency

Active Publication Date: 2020-07-24
王银璇
View PDF10 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] X-ray images of vertebroplasty are one of the conventional medical images. At present, the inspection and analysis of bone cement leakage during surgery are mostly done manually. The analysis takes a long time, is inefficient, and is easy to misjudge. The accuracy of the analysis results depends on the level of the analyst, and it is difficult to guarantee the accuracy and consistency of the detection and analysis results

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
  • Vertebral plasty bone cement leakage detection method and system and storage medium
  • Vertebral plasty bone cement leakage detection method and system and storage medium
  • Vertebral plasty bone cement leakage detection method and system and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0041] The implementation method of bone cement leakage detection in vertebroplasty based on deep learning includes marking X-ray images of bone cement, image key point detection model training, leakage judgment model training, and model deployment.

[0042] The X-ray images of bone cement during vertebroplasty in a tertiary hospital are used as a data set. The data set includes 6,000 X-ray images, including 5,000 images without leakage and 1,000 images with leakage.

[0043] According to attached image 3 As shown, mark the obtained intraoperative bone cement X-ray photos, use the marked images as a data set, and randomly select 80% of the pictures (4800) from the data set as the training set, and the remaining pictures (1200) as the test set. image 3 The middle circle marks the outline of the vertebra, and the triangle marks the outline of the bone cement.

[0044] as attached figure 2 Construct two neural networks, both based on the ResNet-50 model, remove the fully co...

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 vertebroplasty bone cement leakage detection method and system and a storage medium. The vertebroplasty bone cement leakage detection method comprises the following steps: acquiring a vertebroplasty X-ray image for marking contour key points and whether bone cement leaks or not in advance; for a pre-collected X-ray image, constructing a data set according to the marked contour key point position information and an identifier for identifying whether bone cement leaks or not; constructing two neural networks based on an ImageNet pre-training model, using the first neural network for learning vertebra contour information through language, and using the second neural network for learning bone cement contour coordinate information; solidifying the model parameters of the two neural networks, outputting the output values of the two models to an SVM classifier, and learning whether bone cement leakage occurs or not. Compared with manual detection, the prediction timeis greatly shortened, and the problem that the consistency of detection results is poor due to the fact that manual detection of doctors depends on experience of the doctors is solved.

Description

technical field [0001] The present invention relates to the fields of deep learning, intelligent medical image analysis technology, and computer vision, and in particular to a method, system, and storage medium for detecting bone cement leakage in vertebroplasty. Background technique [0002] X-ray images of vertebroplasty are one of the conventional medical images. At present, the inspection and analysis of bone cement leakage during surgery are mostly done manually. The analysis takes a long time, is inefficient, and is easy to misjudge. The accuracy of the analysis results depends on the level of the analyst, and it is difficult to guarantee the accuracy and consistency of the detection and analysis results. Contents of the invention [0003] The technical problem to be solved by the present invention is to provide a method, system and storage medium for detecting bone cement leakage in vertebroplasty in view of the deficiencies in the prior art, so as to quickly and ac...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06T7/00G06N3/04
CPCG06T7/0012G06T2207/10116G06T2207/20081G06T2207/20084G06T2207/30012G06V2201/03G06N3/045G06F18/214G06F18/2411
Inventor 王银璇
Owner 王银璇
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products