# Brief Review — Classification of heart sounds based on the combination of the modified frequency wavelet transform and convolutional neural network

## MFSWT + CNN

Classification of heart sounds based on the combination of the modified frequency wavelet transform and convolutional neural network, by Shandong University

MFSWT + CNN2020 J. MBEC, Over 40 Citations(Sik-Ho Tsang @ Medium)

Heart Sound Classification…

20132023[2LSTM+3FC, 3CONV+2FC] [NRC-Net] [Log-MelSpectrum + Modified VGGNet] [CNN+BiGRU][CWT+MFCC+DWT+CNN+MLP] [LSTM U-Net (LU-Net)]

==== My Other Paper Readings Are Also Over Here ====

**Nodified frequency slice wavelet transform (MFSWT)**and**convolutional neural network (CNN)**are combined for classifying normal and abnormal heart sounds.

# Outline

**MFSWT + CNN****Results**

**1. MFSWT + CNN**

## 1.1. Segmentation

**A hidden Markov model (HMM)**is used to**find the position of each cardiac cycle**in the heart sound signal and determine the exact position of the four periods of S1, S2, systole, and diastole.

## 1.2. Modified Frequency Slice Wavelet Transform (MFSWT)

- The MFSWT is based on the frequency slice wavelet transform (FSWT) and is improved for low-frequency biosignals in ECG [18].
- A signal-adaptive frequency slice function (FSF) in the frequency slice wavelet transform is used to accurately locate the position of each component of the biosignal in the time-frequency diagram, and the classification of ECG signals by this method had achieved good results [18].
- In the
**FSWT**, the**scale**is a*σ***constant or a function of**. But*ω*,*t*, and*u***in the MFSWT, scale**, defined as follows:*σ*is a function of ^*f*(*x*)

- (Please kindly read [18] for FWST.)

## 1.3. Sample Entropy (SampEn)

Sample entropy is denoted by

SampEn(, whereN,r,m)is theNlength of the time signal,is themdimension, andis thersimilar tolerance.

- Suppose that
**a signal with a length of**is written as the following:*N*

- The
is defined as the following:*m*-dimensional vector*Xm*(*i*)

**The distance between two**is defined as*m*-dimensional vectors*Xm*(*i*) and*Xm*(*j*)(*i*≠*j*)as the following:*Di*,*j*

- For the vector
*Xm*(*i*),**the number of vectors**, denoted as*Xm*(*j*) whose distance is less than the tolerance*r*are counted.*Βmi*(*r*)

SampEnisa measure of signal complexityas defined below:

- The lower of the SampEn, the more regular the signal is.

Two CNN modelsare applied toclassify signals according to different SampEn.

## 1.4. CNN

The structure of the two models are similar, all of which are

12-layer neural networks, consisting oftwo convolutional layers, two ReLU layers, two max-pooling layers, three fully connected layers, and input and output layers.

**Mean square error**is used for training:

# 2. Results

**PhysioNet**

The

MaccandAccachieved thehighestscores 0.94 and 0.93, respectively,when the SampEn threshold was 2.76.

The proposed method exhibited the

highest Macc of 0.94.