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

An attention mechanism-based method for marking lesions in optical laryngoscope images

An attention and laryngoscope technology, applied in the field of image understanding, can solve the problems of no labeling of the lesion area in the optical laryngoscope image, and the small data set is prone to overfitting, etc.

Active Publication Date: 2021-11-23
XIDIAN UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to propose a method for labeling lesion areas of optical laryngoscope images based on attention mechanism, so as to solve the problems in the prior art that there is no lesion area labeling for optical laryngoscope images and overfitting easily occurs on small data sets

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
  • An attention mechanism-based method for marking lesions in optical laryngoscope images
  • An attention mechanism-based method for marking lesions in optical laryngoscope images
  • An attention mechanism-based method for marking lesions in optical laryngoscope images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] Below in conjunction with accompanying drawing, specific embodiment of the present invention and effect are further explained and illustrated:

[0029] refer to figure 1 , the implementation steps of this example are as follows:

[0030] Step 1: Data preparation;

[0031] 1a) Obtain the laryngoscope image of the patient, which is recorded as the laryngoscope data set P, and obtain the medical report of the laryngoscope image, which is recorded as the label data set R; the laryngoscope data set P is for example figure 2 shown;

[0032] 1b) Get the training dataset T:

[0033] 1b1) Reduce each image in the laryngoscope dataset P to 224*224;

[0034] 1b2) Center each reduced image, that is, subtract (104, 116, 122) from the pixel value of the reduced image to obtain the pixel value (x', y', z') of the centered image:

[0035] (x',y',z')=(x-104,y-116,z-122)

[0036] Among them, (x, y, z) is the pixel value of the original image;

[0037] 1b3) The laryngoscope data s...

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 labeling lesion areas of optical laryngoscope images based on an attention mechanism, which mainly solves the problems in the prior art that there is no labeling of lesion areas for optical laryngoscope images and overfitting easily occurs on small data sets. Its implementation scheme: obtain laryngoscope image dataset and image dataset label; scale and centralize the image dataset, record the centralized image dataset and image dataset label together as a training dataset; construct a 18 Layer network, using the training data set as the training data of the network, using the adaptive learning rate optimization algorithm to optimize the network to obtain a trained network; input a laryngoscope image to the trained network, according to the generated report in the label The corresponding lesion area is obtained and marked in the data set. The invention avoids the overfitting of small data sets, can obtain and mark the lesion area of ​​the optical laryngoscope image, and is convenient for doctors to diagnose the optical laryngoscope image.

Description

technical field [0001] The invention belongs to the field of image understanding, and in particular relates to a method for labeling lesion areas of optical laryngoscope images, which can be used for labeling lesion areas of optical laryngoscope images, and improves the diagnostic efficiency and accuracy of doctors for optical laryngoscope images. Background technique [0002] Laryngoscope images play an important role in the diagnosis and treatment of diseases in the nasal cavity and throat. However, the diagnosis of laryngoscope images generally requires professional doctors to carefully check each area in the original image, which is an important tool for doctors. Challenging tasks. Because there is no suggestive label in the original image, all lesions can only be observed by the doctor's naked eyes. Long-term observation will cause the doctor's attention to decrease and mental fatigue, which will lead to missed and misjudgment during the examination and a decrease in di...

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): G06T7/00
CPCG06T7/0012G06T2207/20081G06T2207/20084G06T2207/30204
Inventor 缑水平李国栋谭瑶毛莎莎许成陈佳伟焦昶哲焦李成
Owner XIDIAN 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