Method for improving prediction performance of computer-aided stress fracture prediction system

A computer-aided and stress-based technology, applied in computer parts, computing, kernel methods, etc., can solve problems such as poor system performance, weak feature extraction ability, and noise data influence, so as to improve training stability and increase generalization. Sex, the effect of speeding up training

Inactive Publication Date: 2021-09-24
BEIJING INSTITUTE OF TECHNOLOGYGY +1
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The purpose of the present invention is to overcome the technical problems of the existing computer-aided stress fracture prediction system, such as weak feature extraction ability and too much influence of noise data, which lead to poor system performance, and propose a new solution

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
  • Method for improving prediction performance of computer-aided stress fracture prediction system
  • Method for improving prediction performance of computer-aided stress fracture prediction system
  • Method for improving prediction performance of computer-aided stress fracture prediction system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0047] A method to improve the predictive performance of computer-aided predictive stress fracture systems such as figure 1 shown, including the following steps:

[0048] Step 1: Training phase.

[0049] Step 1.1: Denoise the training data, set the noise ratio to 10%, randomly construct the isolated forest multiple times, select the 10% sample points with the least average number of splits, and eliminate them;

[0050] Step 1.2: Select the largest attribute vector and the smallest attribute vector of the cleaned training data, and then use each sample to perform normalization according to formula (1).

[0051] Step 1.3: Set the dimension of ICA feature extraction to 2, and use the normalized data to train the ICA feature extractor to obtain the features extracted from the training set.

[0052] Step 1.4: For the extracted features, we set the number of resampled positive and negative samples to be 300, then select an appropriate amount of feature samples x, and calculate the...

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 relates to a method for improving the prediction performance of a computer-aided stress fracture prediction system, aims to overcome the technical problem that the existing computer-aided stress fracture prediction system is poor in system performance due to the defects that the feature extraction capability is not strong, noise data is greatly influenced and the like, and belongs to the technical field of computer application. According to the method, for a computer-aided stress fracture prediction system, abnormal value detection and elimination are realized through an isolated forest algorithm, then feature extraction is performed by using an independent component analysis (ICA) technology, then data resampling is performed by using a comprehensive minority class oversampling technology (SMOTE), and finally classification is realized through a support vector machine. Therefore, the auxiliary prediction effect of the system is greatly improved. Experiments prove that the method can effectively improve the model accuracy and sensitivity of the system, the performance of the system can be improved to 78.7% of accuracy and 82.5% of sensitivity, and the method is superior to the prior art.

Description

technical field [0001] The invention relates to a method for improving the prediction performance of a computer-aided stress fracture prediction system. Specifically, the accuracy rate of stress fracture prediction of the system is improved by improving the automatic analysis of gait data, and belongs to the technical field of computer-aided diagnosis. Background technique [0002] At present, computer-aided diagnosis technology and system are widely used in various fields of medical diagnosis. Through medical processing technology and other possible physiological and biochemical means, combined with computer analysis and calculation, the accuracy of diagnosis and prediction can be improved. Computer-aided diagnosis is also one of the basic problems of computer application. [0003] In the existing computer-aided diagnosis system for stress fractures, the classification methods adopted by the system are mainly divided into two categories: [0004] 1) A stress fracture predi...

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): G06K9/62G06N20/10
CPCG06N20/10G06F18/2411G06F18/24323G06F18/214
Inventor 刘峡壁廖东海汪爱媛彭江周皓赵燕旭冯勇
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap