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

Intelligent classification of endoscopic images and detection method of irregular lesion areas

A lesion area and irregular technology, applied in the field of medical image intelligent processing, can solve the problems of manual feature difficulty and achieve the effect of improving application value and accuracy

Inactive Publication Date: 2018-10-16
FUDAN UNIV
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] Traditional salient object detection methods use more hand-crafted feature boosting effects, benchmarks published by Nankai University [3] Traditional methods that work best in [4] 34 manual features are used, however, it is very difficult to continue to add manual features

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
  • Intelligent classification of endoscopic images and detection method of irregular lesion areas
  • Intelligent classification of endoscopic images and detection method of irregular lesion areas
  • Intelligent classification of endoscopic images and detection method of irregular lesion areas

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] use figure 1 In the shown method, annotated endoscopic images are used to train a saliency detection network model to convergence, and the model is used for endoscopic image classification and irregular lesion region detection. For an endoscopic image to be detected, one can use figure 2 method shown. details as follows:

[0046] The specific implementation method is:

[0047] (1) Preprocess the image, according to the value of the R channel in the RGB channel of the endoscopic image (the range is 0-255), use a smaller threshold (such as 30) to binarize the image, and remove the lower value of the binarized image , and record the position information of the reserved area as (x, y, H, W);

[0048] (2) Input the preprocessed image to the network model to obtain the saliency map M, and the threshold T is set to 0.5. If there is M(i,j)>T in M, the endoscopic image is classified as lesion, otherwise it is classified as lesion. is normal; if the endoscopic image is a le...

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 belongs to the technical field of intelligent medical image processing, and specifically refers to an intelligent classification of endoscopic images and a detection method of irregularlesion areas. An artificial intelligence is used to apply a saliency detection to the classification of endoscopic images and the detection of irregular lesion areas, and the endoscopic image markedby senior doctors is used as a training sample and input into the saliency detection network model, so that the network model learns irregularly saliency areas, namely lesion areas. Experimental results show that according to the intelligent classification of endoscopic images and the detection method of irregular lesion areas, the diagnosis experience of senior doctors can be learned, the endoscopic image is more accurately divided into a lesion image and a normal image, an irregular lesion area is detected, and a reference is provided for the doctor so as to improve the diagnosis accuracy.

Description

technical field [0001] The invention belongs to the technical field of medical image intelligent processing, and in particular relates to a medical image classification and lesion detection method, and more specifically, to an endoscopic image intelligent classification and irregular lesion area detection method. Background technique [0002] At present, the screening, treatment, and follow-up of early gastrointestinal cancer mainly rely on endoscopy and imaging examinations such as CT / MRI, among which endoscopy is the most important. With the advancement of endoscopic diagnostic technology, various methods such as white light observation, electronic chromosomal magnifying endoscopy, ultrasonography, and iodine staining can be used. However, the detection rate of the above-mentioned inspection methods is affected by the experience of clinical endoscopists, and the subjective influence of different examiners is relatively large in judgment, especially in areas with low medica...

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): G06T7/00
CPCG06T7/0014G06T2207/10068G06T2207/20081G06T2207/20104G06T2207/20116
Inventor 颜波钟芸诗牛雪静蔡世伦谭伟敏李冰
Owner FUDAN UNIV
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