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

Heart sound classification method and system based on CNN combined with improved frequency wavelet slice transformation

A classification method and heart sound technology, applied in neural learning methods, instruments, biological neural network models, etc., can solve problems such as large differences in heart sound signals, inability to achieve classification, and multiple interference signals in heart sound signals, and achieve improved accuracy and good performance. The effect of classification ability

Active Publication Date: 2020-07-03
SHANDONG UNIV
View PDF3 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These problems create factual difficulties in classifying heart sounds
[0004] The inventors of the present disclosure have found that (1) most of the current heart sound classification researchers have adopted feature selection, some algorithms of machine learning and deep learning, such as neural networks, support vector machines, random forests, decision trees and k-nearest neighbors, but The heart sound signal contains many interference signals, and it is often impossible to train a better model algorithm by directly extracting it, so that accurate classification cannot be achieved; (2) The difference between the heart sound signals of different individuals and between different sampling devices The feature extraction and training of heart sound signals are simple, and the trained network model often cannot accurately classify various heart sound signals.

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
  • Heart sound classification method and system based on CNN combined with improved frequency wavelet slice transformation
  • Heart sound classification method and system based on CNN combined with improved frequency wavelet slice transformation
  • Heart sound classification method and system based on CNN combined with improved frequency wavelet slice transformation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] Such as figure 1 As shown, Embodiment 1 of the present disclosure provides a heart sound classification method based on CNN combined with improved frequency wavelet slice transform, divides the data into two parts according to the sample entropy value, and trains two different parameters on the two parts of data respectively. Convolutional Neural Network (Convolutional Neural Network) models. When performing ten-fold cross-validation, different convolutional neural network models (model 1 and model 2 ) to classify, the heart sound signal greater than the sample entropy threshold adopts the model 1 For classification, the heart sound signal smaller than the sample entropy threshold adopts the model 2 Classification results show that this method has good classification ability for heart sound signals.

[0058] Specifically include the following aspects:

[0059] (1) Data selection

[0060] The present embodiment adopts the public database of the Heart Sound Challeng...

Embodiment 2

[0140] Embodiment 2 of the present disclosure provides a heart sound classification system combining CNN with improved frequency wavelet slice transformation, including:

[0141] The preprocessing module is configured to: preprocess the acquired heart sound signal, use the hidden semi-Markov model to find the position of each cardiac cycle signal of each heart sound signal, intercept each cardiac cycle signal, and find the first cardiac cycle signal. sound, second heart sound, systole, and diastole;

[0142] The data conversion module is configured to: utilize the improved frequency slice wavelet transform to convert each intercepted one-dimensional cardiac cycle signal into a two-dimensional time-frequency image;

[0143] The network training and classification module is configured to: respectively calculate the sample entropy of the acquired heart sound signal, and compare it with the preset sample entropy threshold; when the sample entropy of the heart sound signal is great...

Embodiment 3

[0145] Embodiment 3 of the present disclosure provides a medium on which a program is stored, and when the program is executed by a processor, the steps in the heart sound classification method of CNN combined with improved frequency wavelet slice transformation as described in Embodiment 1 of the present disclosure are realized.

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 provides a heart sound classification method and system based on CNN combined with improved frequency wavelet slice transformation, and the method comprises the steps: carrying out the preprocessing of an obtained heart sound signal, finding the position of each cardiac cycle through a hidden half Markov model, and carrying out the interception of each cardiac cycle signal; converting each intercepted one-dimensional cardiac cycle signal into a two-dimensional time-frequency image by using improved frequency slice wavelet transform; calculating sample entropies of the acquired heart sound signals respectively; comparing the two-dimensional time-frequency image with a preset sample entropy threshold, when the sample entropy of the heart sound signal is greater than the presetsample entropy threshold, performing network training and classification according to the two-dimensional time-frequency image by using a first convolutional neural network, otherwise, performing network training and classification according to the two-dimensional time-frequency image by using a second convolutional neural network; firstly, signals with different interference degrees are distinguished by using sample entropy, and then classification is performed by using different convolutional neural network models for different signals, so that the accuracy of heart sound signal classification is greatly improved.

Description

technical field [0001] The present disclosure relates to the technical field of heart sound classification, in particular to a heart sound classification method and system based on CNN combined with improved frequency wavelet slice transformation. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. [0003] Cardiovascular and cerebrovascular diseases have already been the diseases with the largest number of deaths in the world. Heart sound is an important physiological signal reflecting the health of the heart. Abnormal heart sounds can reflect many heart diseases and can help doctors make a diagnosis of a patient's condition. According to the heart activity, the heart sound signal can be divided into four phases, namely the first heart sound (S1 phase), the systolic phase, the second heart sound (S2 phase) and the diastolic phase. Abnormal heart ...

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): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F2218/06G06F2218/12Y02A90/10
Inventor 魏守水陈永超马彩云
Owner SHANDONG 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