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

Image detection method based on ultrasonic imaging and Bayesian optimization

A technology of ultrasonic imaging detection and ultrasonic imaging, which is applied in the directions of ultrasonic/sonic/infrasonic image/data processing, organ movement/change detection, ultrasonic/sonic/infrasonic diagnosis, etc. Infection and other issues

Pending Publication Date: 2021-02-19
ANHUI UNIV OF SCI & TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the test time of repeated breathing method is longer, and the test technique is more complicated
In addition, the method based on differential pressure flow sensor measurement is also one of the commonly used methods in lung function testing. This method is simple to operate, but its calibration is cumbersome and there is a risk of cross-infection

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
  • Image detection method based on ultrasonic imaging and Bayesian optimization
  • Image detection method based on ultrasonic imaging and Bayesian optimization
  • Image detection method based on ultrasonic imaging and Bayesian optimization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0077] The concrete steps of described step S6 are as follows:

[0078] S61: Initialization of model hyperparameters: Bayesian optimization KPCANet model hyperparameters initialization, initialization parameters mainly include the first layer filter size PatchSize1, the number of first layer filters NumFilters1, the second layer filter size PatchSize2, the first layer The number of filters NumFilters2, the local histogram block size HistBlockSize1, HistBlockSize2 and the overlapping block area ratio BlockOverLapRatio;

[0079] S62: Model performance calculation: According to the initialized model hyperparameters, determine them as the initialization parameters of the KPCANet model, and then input the training set samples into the model for training of the analysis model. After the model is trained, use the verification set samples for preliminary verification The performance of the model, and record the detection error rate E of the verification set;

[0080] S63: Bayesian al...

Embodiment 2

[0089] In described step S62, utilize KPCANet model to miner's lung ultrasonic imaging data analysis, its specific steps are as follows:

[0090] S621: Construct the first layer of KPCA;

[0091] S622: Using the output of the first layer KPCA to construct the second layer KPCA;

[0092] S623: Perform binary hashing and block histogram processing;

[0093] S624: SVM classification of lung ultrasound imaging characteristics: input the features of the lung ultrasound imaging information extracted by the two-layer KPCANet into the SVM classifier, wherein the Gaussian function is selected as the kernel function of the SVM.

Embodiment 3

[0095] The step of constructing the first layer of KPCA in step S621 is as follows:

[0096] S6211: The preprocessed ultrasound image I i The pixels are divided into k 1 *k 2 The size of the image block, after Patch vectorization, all the point information is arranged in an orderly manner, which is recorded as

[0097] S6212: Then convert each image block matrix into a column vector and perform mean removal operation to obtain:

[0098]

[0099] S6213: Perform the above steps S5211 and S5212 on all input images to obtain:

[0100]

[0101] S6214: Calculate the input kernel matrix K 1 :

[0102] K 1 =ConstructKernelMatrix(X) (3)

[0103] and K 1 Center to get K c1 , the number of filters in the first layer is NumFilters1, the purpose of the KPCA algorithm is to extract K c1 The eigenvectors corresponding to the largest eigenvalues ​​of the first NumFilters1 are used as convolution filters:

[0104]

[0105] The main information of these zero-mean training sa...

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 an image detection method based on ultrasonic imaging and Bayesian optimization, and relates to the technical field of image detection. An advanced ultrasonic imaging nondestructive detection means is adopted when the lung function of miners is detected, meanwhile, in the ultrasonic imaging lung function analysis process, a Bayesian optimization KPCANet algorithm is adopted, so that the accuracy and reliability of lung function analysis of the miners are guaranteed, the lung function conditions of the miners are accurately and reliably mastered, the body health states of the miners can be accurately and timely mastered, and the method has important significance in early discovery of occupational pneumoconiosis. The lung function detection and analysis are carried out on the special group of the miners, the accurate and real-time measurement and analysis of the lung health state of the miners are achieved, the early warning of occupational pneumoconiosis and other occupational diseases is completed, and the life health of the miners is guaranteed.

Description

technical field [0001] The invention relates to the technical field of image detection, in particular to an image detection method based on ultrasonic imaging and Bayesian optimization. Background technique [0002] Occupational coal worker's pneumoconiosis (CWP) refers to the general term for occupational pneumoconiosis caused by coal mine workers who inhale productive dust for a long time during occupational activities. It is one of the most common occupational diseases in my country. The annual number of new cases of CWP accounts for the first of new occupational diseases in my country, and it is a research hotspot of pneumoconiosis single disease in my country. The main pathological basis of CWP is diffuse fibrosis of lung tissue. The main clinical manifestations are chest tightness, chest pain, shortness of breath, cough, cough Phlegm, fatigue, dyspnea, etc. Due to the gradual aggravation of respiratory symptoms, the patient's activities are limited, which seriously affe...

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): A61B8/08A61B8/00
CPCA61B8/08A61B8/52
Inventor 周孟然胡锋卞凯曹珍贯凌六一梁喆闫鹏程
Owner ANHUI UNIV OF SCI & TECH
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