2008 5th International Conference on Electrical Engineering, Computing Science and Automatic Control 2008
DOI: 10.1109/iceee.2008.4723403
|View full text |Cite
|
Sign up to set email alerts
|

Classification of heart sounds using wavelets and neural networks

Abstract: A heart sound feature extraction and classification method has been developed. It used the discrete wavelet decomposition and reconstruction to produce the envelopes of details of the signals for further extracting the features. Some statistical variables were extracted from the processed signals and used as the features for the heart sounds classification. A Multilayer Perceptron Neural Network has been used for classification of heart sounds. The performance of the proposed method has been evaluated using 25… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2

Citation Types

0
6
0

Year Published

2012
2012
2022
2022

Publication Types

Select...
4
3

Relationship

0
7

Authors

Journals

citations
Cited by 12 publications
(6 citation statements)
references
References 8 publications
0
6
0
Order By: Relevance
“…Most of them use neural networks (NNs), support vector machines (SVMs) or some complex preprocessing algorithms to carry out this task [7]- [17], [18], [19]. Many studies like [10], [11], [15] have used a processing step where a person selects the best portion of the sound signal that should be used as input to the system, making this solution not ideal for a real scenario because of the need of human interaction. Some of them have used NNs to classify between different kinds of heart murmurs [7], [9]- [11], but have only trained the network with simulated heart sounds with no noise, obtaining very bad accuracy results when testing the classifier with real heart sounds (48.5%).…”
Section: Table I Comparative Study Between State-of-the-art Studies About Heart Sound Diagnosis Systemsmentioning
confidence: 99%
See 1 more Smart Citation
“…Most of them use neural networks (NNs), support vector machines (SVMs) or some complex preprocessing algorithms to carry out this task [7]- [17], [18], [19]. Many studies like [10], [11], [15] have used a processing step where a person selects the best portion of the sound signal that should be used as input to the system, making this solution not ideal for a real scenario because of the need of human interaction. Some of them have used NNs to classify between different kinds of heart murmurs [7], [9]- [11], but have only trained the network with simulated heart sounds with no noise, obtaining very bad accuracy results when testing the classifier with real heart sounds (48.5%).…”
Section: Table I Comparative Study Between State-of-the-art Studies About Heart Sound Diagnosis Systemsmentioning
confidence: 99%
“…Many studies like [10], [11], [15] have used a processing step where a person selects the best portion of the sound signal that should be used as input to the system, making this solution not ideal for a real scenario because of the need of human interaction. Some of them have used NNs to classify between different kinds of heart murmurs [7], [9]- [11], but have only trained the network with simulated heart sounds with no noise, obtaining very bad accuracy results when testing the classifier with real heart sounds (48.5%). Others have used only a small amount of real heart sounds [10], [12], [14]- [17], which is not representative when it comes to testing it in a real scenario.…”
Section: Table I Comparative Study Between State-of-the-art Studies About Heart Sound Diagnosis Systemsmentioning
confidence: 99%
“…Most of them use neural networks [1]- [7], support vector machines [8], [9] and some complex preprocessing steps such as statistical analysis [10] to perform classification tasks. Studies such as literature [4] have adopted a number of processing steps, which show that the researchers can select the best part of the heart sound signal as the input of the system and these programs are not ideal for processing real scenes. Some of them use neural networks to classify the different types of heart sound signals, but they only use simulated heart sounds that do not contain noise in order to train the networks [1].…”
Section: Introductionmentioning
confidence: 99%
“…Research projects concerning wavelet may include general areas such as digitize spatial profits of structured cracks but since the area of interest is in medical area, Wavelet has been applied in phonocardiography, pulse oximetry traces (photoplyethysmograms) and the electrocardiogram (ECG). Wavelet transform has been used in order to extract features to determine the abnormality of the heart signal [2], [3]. There is also work in this area on denoising method for heart sound using thresholding function in wavelet domain [4].…”
Section: Introductionmentioning
confidence: 99%