Supercharge Your Innovation With Domain-Expert AI Agents!

PCA-LDA-based medical image processing system and method

A medical image and image technology, applied in the field of radiomics, to achieve the effect of low dimensionality, high recognizability, and strong generalization ability

Inactive Publication Date: 2018-08-21
CHENGDU UNIV OF INFORMATION TECH
View PDF8 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] In order to solve the technical problems existing in the existing radiomics processing process, the present invention provides a medical image processing system and method based on PCA-LDA, which uses the PCA-LDA algorithm to perform feature selection and reduction of extracted high-dimensional feature vectors. Dimensional processing to achieve supervised feature selection, and the feature vector obtained after dimension reduction has low dimensionality, linear independence between dimensions, and more discriminative ability, which can realize a better classification model and better display the classification effect

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
  • PCA-LDA-based medical image processing system and method
  • PCA-LDA-based medical image processing system and method
  • PCA-LDA-based medical image processing system and method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0038]This embodiment proposes a medical image processing system based on PCA-LDA, which uses the PCA-LDA algorithm to perform feature selection and dimensionality reduction processing on the extracted high-dimensional feature vectors, and implements a supervised feature selection process, and The feature vectors obtained after dimensionality reduction have low dimensions, are linearly independent between each dimension, and have more discriminative ability, which can realize a better classification model and better display the classification effect.

[0039] Such as figure 1 As shown, the system specifically includes an image acquisition module, a feature extraction module and a classification model establishment module;

[0040] The image acquisition module is used to acquire the original medical image and the ROI image obtained by processing the original medical image;

[0041] The feature extraction module is used to extract features from the acquired original medical ima...

Embodiment 2

[0075] The medical image processing system corresponding to the above-mentioned embodiments, such as figure 2 Shown, the present invention proposes a kind of medical image processing method based on PCA-LDA, comprises the following steps:

[0076] S01. Obtain an original medical image and an ROI image obtained by processing the original medical image;

[0077] S02, performing feature extraction on the acquired original medical image and ROI image;

[0078] It includes two parts: the extraction of clinical features and image features. The extraction of clinical features is to extract the data on the image by identifying the original medical image; the extraction of image features first performs wavelet transform on the ROI image to obtain the image after wavelet transform, and then The texture features and gray co-occurrence matrix features are extracted from the original ROI image and the ROI image after wavelet transform.

[0079] S03. In order to make the subsequent proce...

Embodiment 3

[0083] Carry out PCA-based medical image processing, LDA-based medical image processing, and adopt PCA-LDA-based medical image processing of the present invention to carry out radiomics processing on a group of tumor medical images, as shown in Table 1-1 to Table 1 The positive and negative sample test results corresponding to -3 and the corresponding Figure 3-5 Classification renderings.

[0084] Table 1-1

[0085]

[0086] Table 1-2

[0087]

[0088] Table 1-3

[0089]

[0090] Table 1-1 shows the classification prediction using only the PCA algorithm, and the obtained test accuracy rate is 0.650000, such as image 3 Shown; Table 1-2 shows the classification prediction using only the LDA algorithm, and the test accuracy rate obtained is 0.600000, such as Figure 4 Shown; Table 1-3 adopts the present invention, promptly adopts the classification prediction that PCA-LDA algorithm carries out, and the test accuracy rate that obtains is 0.850000, as Figure 5 It can...

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 PCA-LDA-based medical image processing system and method. A PCA-LDA algorithm is adopted to carry out dimension reduction processing on preprocessed features; a PCA algorithmis firstly adopted to carry out dimension reduction on a preprocessed feature vector, and the feature redundancy after dimension reduction is reduced and linear independence is achieved; an LDA algorithm is then adopted to carry out dimension reduction to obtain a low-dimension feature vector with the most discriminating ability. The PCA-LDA algorithm is adopted to carry out dimension reduction processing and feature preference on the extracted features, supervised feature selection is realized, the low-dimension feature vector after dimension reduction is more recognizable, the classification effect can be better displayed, a better classification model is obtained, and the classification is more accurate and reliable.

Description

technical field [0001] The invention relates to the field of radiomics, in particular to a PCA-LDA-based medical image processing system and method for supervised feature selection and dimensionality reduction processing. Background technique [0002] Radiomics is an emerging field of radiomics [1] , its goal is to extract and analyze a large number of high-throughput quantitative image features from medical images (CT scans, positron emission scans (PET) or magnetic resonance imaging, etc.) Phenotype model, establish the association between image features and clinical phenotype or gene molecular markers, and then carry out tumor diagnosis and clinical phenotype prediction [2] . In the clinical decision-making of thyroid cancer, for the judgment of benign and malignant conditions, the use of radiomics methods can make better clinical decisions. [0003] In 2012, Dutch scholar Lambin proposed the concept of radiomics for the first time, and his idea came from tumor heterog...

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): G06K9/62
CPCG06F18/2135G06F18/24
Inventor 符颖吴锡邢晓羊李玉莲
Owner CHENGDU UNIV OF INFORMATION TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com