Extraction method of suspected lymph nodes in gastric CT images based on sparse dynamic integration selection

A CT image and lymph node technology, applied in the field of medical image processing, can solve the problems of high computational complexity and time-consuming, and achieve the effect of good classification performance and reduced computational complexity.

Active Publication Date: 2016-05-25
XIDIAN UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Dynamic ensemble selection is the latest achievement of multi-classifier ensemble. It has received extensive attention because it can achieve better classification performance. However, in the prior art, when dynamic ensemble selection is used to classify adipose tissue, the computational complexity is high. takes a lot of time

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
  • Extraction method of suspected lymph nodes in gastric CT images based on sparse dynamic integration selection
  • Extraction method of suspected lymph nodes in gastric CT images based on sparse dynamic integration selection
  • Extraction method of suspected lymph nodes in gastric CT images based on sparse dynamic integration selection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] The present invention is a method for extracting suspected lymph nodes from a stomach CT image based on sparse dynamic integration selection. Current CT images are commonly used imaging methods. To extract suspected lymph node regions from adipose tissue regions of CT images, the adipose tissue regions must be extracted first . Lymph nodes are mainly distributed in the adipose tissue around the stomach wall, so accurate segmentation of the adipose tissue is the first priority.

[0047] The process of the present invention for extracting the suspected lymph node region from the image of the gastric adipose tissue is as follows figure 1 Shown. The specific description is as follows:

[0048] Step 1. Extract a CT image of the stomach, see Figure 4 (a). The stomach CT images used in this example are from Beijing Cancer Hospital. Each CT image has 512x512 pixels. All image processing experiments are implemented on the MATLAB2009a experimental platform.

[0049] Step 2: Mark par...

Embodiment 2

[0066] The method for extracting suspected lymph nodes from a stomach CT image based on sparse dynamic ensemble selection is the same as that in Example 1. The process of using the sparse dynamic ensemble selection method to classify the extracted adipose tissue region sample set to be classified in step 5 includes training and testing. stage:

[0067] The flow chart of the training phase is as figure 2 As shown, the specific process is:

[0068] 5.1 Select training samples from some labeled images of adipose tissue that have been successfully classified. The training samples are denoted as X = {(x i , Y i )|x i ∈R F , Y i ∈{1, 2,..., L}, i=1, 2,..., r}, where x i Is the gray histogram feature of the sample extracted from the labeled image of adipose tissue, y i Is the mark of the sample, F is the dimension of the feature of the gray histogram, L is the number of segmentation categories, r is the number of samples extracted; the number of selected training samples should be greater...

Embodiment 3

[0091] The method for extracting suspected lymph nodes from the stomach CT image selected by the sparse dynamic integration is the same as in Example 1-2. In this example, the effectiveness of the present invention is verified through simulation experiments using stomach CT images from Beijing Cancer Hospital. Each image consists of 512x512 pixels, and 3 images are selected for experimentation. All experiments are implemented on the MATLAB2009a experimental platform. Figure 4-Figure 6 The results of adipose tissue extraction from three images. in Figure 4-Figure 6 In each picture, (a) is the original image, (b) is the result after inputting interactive information, (c) is the accurate adipose tissue region extraction result, (e) is the fat of the method used in the present invention The results of tissue region extraction, (d) is the partial enlarged view corresponding to (c), and (f) is the partial enlarged view corresponding to (e). Figure 4-Figure 6 The similarity between ...

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 method for extracting a stomach CT (Computed Tomography) image suspected to be a lymph node based on sparse dynamic integrated selection and belongs to the field of medical image processing. According to the method, the problems that effective background information is difficult to determine and is high in complexity when a suspected lymph node area is extracted are mainly solved. The realization process of the method comprises the following steps of extracting an adipose tissue area from the stomach CT image by using an interactive segmentation method; over segmenting the extracted adipose tissue area by using mean shift; extracting the characteristics of a grey level histogram in the over-segmentation area to form a sample set to be classified; utilizing a training sample learning dictionary and a classifier integration system; generating an atom integration system for each atom; sparsely encoding a sample to be classified and outputting a final classifying result according to the atom integration system; combining the over-segmentation areas of the same type to obtain the stomach CT image suspected to be the lymph node area. According to the method disclosed by the invention, adipose tissues can be automatically extracted under the condition that no background is marked and the time complexity of extraction of the suspected lymph node area is effectively reduced.

Description

Technical field [0001] The invention belongs to the technical field of medical image processing, and particularly relates to medical CT image segmentation and classification, in particular to a method for extracting suspected lymph nodes from a stomach CT image selected by sparse dynamic integration. It can be used for segmentation and classification of CT images of stomach fat tissue, other types of CT image segmentation and classification, and also for segmentation or classification of other types of images. Background technique [0002] Gastric cancer ranks second in the global mortality rate of malignant tumors. Its rapid development and easy metastasis are the important reasons for the high mortality rate. Lymph node metastasis is an important independent factor affecting the prognosis of gastric cancer. Before medical and surgical treatment, it is necessary to understand the condition of the lymph node as accurately as possible to determine the treatment plan and evaluate t...

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 Patents(China)
IPC IPC(8): G06T7/00A61B6/03
Inventor 刘芳周治国李玲玲郝红侠戚玉涛焦李成李婉尚荣华马文萍马晶晶
Owner XIDIAN UNIV
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