Training method and verification method for lesion identification model, and lesion image identification device

An image recognition device and recognition model technology, applied in the field of deep learning, can solve the problems of high professional technical requirements and poor diagnostic accuracy of various endoscopic techniques, and achieve the effect of overcoming strong subjectivity

Inactive Publication Date: 2017-11-21
BEIJING HOTWIRE MEDICAL TECH DEV CO LTD
View PDF2 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] At present, the diagnosis of various gastrointestinal lesions under manual electronic endoscopy is highly subjective

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
  • Training method and verification method for lesion identification model, and lesion image identification device
  • Training method and verification method for lesion identification model, and lesion image identification device
  • Training method and verification method for lesion identification model, and lesion image identification device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] In order to enable those skilled in the art to better understand the solution of the present invention, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0036] An embodiment of the present invention provides a lesion model training method, the basic idea is to use the acquired lesion image samples to train the Faster RCNN network to obtain a lesion recognition model. The lesion identification model is obtained based on deep learning. When using this model for lesion identification, the location and type of the lesion can be judged more accurately.

[0037] The Faster RCNN network in the embodiment of the present invention includes a candidate window network (Region ProposalNetworks, RPN) and a fast region convolutional neural network (Fast Region-Based Convolutional NeuralNetworks, FRCN), which combines the RPN network and the Fast RCNN network , directly connect the proposal obtain...

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 training method for a lesion identification model. The training method comprises the following step of utilizing an obtained lesion image sample to train a Faster RCNN (Recurrent Neural Network) to obtain the lesion identification model. The invention also correspondingly provides a lesion image device based on the lesion identification model and a verification method for the lesion identification mode. The training method for the lesion identification model is designed on the basis of a deep learning concept, and the position and the type of a focus can be accurately judged when the model is used for carrying out lesion identification.

Description

technical field [0001] The present invention relates to the technical field of deep learning, in particular to a training method for a lesion recognition model, a lesion image recognition device using a data model obtained by the training method, and a verification method for the lesion recognition model. Background technique [0002] Deep learning aims to imitate the neural network of the human brain. It acts like the "neocortex" in the human brain that is in charge of perception, motor commands, consciousness, and language. It can learn to recognize sounds, images, and other data by itself, thereby helping computers to crack some Humans rely almost entirely on intuition to solve trivial problems, from recognizing faces to understanding language. [0003] Deep learning itself also grew out of an even older computing idea: neural networks. These neural network systems simulate the tight connections between nerve cells in the human brain. These nerve cells can communicate w...

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): G06K9/66G06K9/62G06F19/00
CPCG06V30/194G06F18/214
Inventor 李洪涛成冠举
Owner BEIJING HOTWIRE MEDICAL TECH DEV CO LTD
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