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

Jaundice triage information analysis method and system based on automatic identification

A technology of automatic identification and information analysis, applied in the medical field, can solve the problems of large deviation of jaundice identification, accuracy and recall rate lower than judgment requirements, and inability to apply, so as to improve accuracy, optimize model structure, and simplify judgment. effect of the process

Active Publication Date: 2021-05-25
北京变色龙科技有限公司
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the deficiencies in the prior art, the present invention discloses a method and system for analyzing jaundice triage information based on automatic identification, which is used to solve the problem of The deep neural network model is directly applied to the problem of large deviations in jaundice recognition, and in the existing mainstream neural network experiments, the accuracy and recall rates are lower than the judgment requirements, which cannot provide users with valuable judgments. Applied to market issues

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
  • Jaundice triage information analysis method and system based on automatic identification
  • Jaundice triage information analysis method and system based on automatic identification
  • Jaundice triage information analysis method and system based on automatic identification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] This embodiment discloses as figure 1 A kind of jaundice triage information analysis method based on automatic recognition shown, comprises the following steps:

[0051] S1 Acquire clinically valuable triage picture data covering various light source conditions, jaundice cases and non-jaundice cases in batches;

[0052]S2 enters into data cleaning, and discards the picture data that does not have information about infant jaundice or the picture itself is in poor condition;

[0053] S3 converts the picture into HSV space, after filtering, enters the global brightness adjustment through the V component judgment of HSV space and performs formula transformation processing;

[0054] S4 divides the processed picture into subcategories by giving several thresholds or dividing the whole body and local classification;

[0055] S5 builds a network model, conducts independent training for each sub-category, and obtains different model parameters;

[0056] S6 retrieves the corr...

Embodiment 2

[0071] This embodiment discloses a network model, the thinking is as follows:

[0072] 1 is designed with an intermediate auxiliary layer, and the auxiliary layer generates output, thus having an effect similar to a regular term.

[0073] 2 Adopt a parallel structure of the pooling layer and the convolutional layer, thereby reducing the size of the feature map, that is, achieving dimensionality reduction.

[0074] 3 using the BN method

[0075] 4 The entire network design is reduced and adjusted based on Inception V3.

[0076] In this embodiment, the model training process independently trains each subclass divided in the third step to obtain different model parameters.

[0077] For each subclass of data, the image data and corresponding label data are divided into training set and test set, and multiple rounds of iterations are performed. Select the parameters before overfitting as the final model parameters.

[0078] In this embodiment, model identification includes the ...

Embodiment 3

[0093] This embodiment discloses as figure 2 A jaundice triage information analysis system based on automatic identification is shown, and the detailed description is as follows:

[0094] 1_Data Preparation Module

[0095] It mainly provides user input function, and has data reading and storage functions.

[0096] First of all, the user can specify the input location of the original image in the form of file and http, and the system can display the image, and can perform zooming, rotation, and other processing on the image.

[0097] This module provides data storage function. The read pictures are compressed and stored in the system. Each picture is given an independent and globally unique ID number.

[0098] Modifications to pictures in this module are only for operator observation and cannot be stored in the system.

[0099] Two image processing modules

[0100] This module performs image processing according to the aforementioned logic, and saves the processed pictures...

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 relates to the technical field of medical treatment, in particular to a jaundice triage information analysis method and system based on automatic identification. The method comprises the steps: cleaning image data, and discarding image data which does not have infant jaundice condition information or is poor in image conditions; converting the picture into an HSV space, filtering, judging through a V component of the HSV space, entering global brightness adjustment, and performing formula conversion processing; carrying out subclass division on the processed picture by giving a plurality of threshold values or dividing a whole body and carrying out local classification; independently training each divided subclass to obtain different model parameters; and calling corresponding model data, and inputting the processed pictures into the model for calculation to obtain a jaundice-free probability, a physiological jaundice probability and a pathological jaundice probability. According to the method, the AI technology is applied to the field of infant health, so that the process of judging the neonatal jaundice is simplified through graphic data processing and model structure optimization, and the accuracy of preliminary judgment of the neonatal jaundice is improved.

Description

technical field [0001] The invention relates to the field of medical technology, in particular to an automatic identification-based jaundice triage information analysis method and system. Background technique [0002] At present, ordinary users mainly rely on naked eye observation to judge whether newborns have jaundice symptoms. Because ordinary users, especially first-born parents, lack experience and cannot make accurate judgments, factors such as unsuitable ambient light and lack of experience may cause It can lead to failure to correctly identify jaundice, or to mistakenly distinguish between physiological jaundice and pathological jaundice. As a result, parents do not adopt the correct way to treat newborns, which may not only lead to delays in treatment, but also may lead to unnecessary waste of time, energy and medical resources. [0003] The application of mainstream neural network models in image recognition generally has a large depth, high running cost and slow ...

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): A61B5/00
CPCA61B5/0064A61B5/441A61B2503/045
Inventor 谢景帆
Owner 北京变色龙科技有限公司
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