ScholarLens Back to search
2019•IOP Conference Series Materials Science and EngineeringOpen access

Heart Sound Diagnose System with BFCC, MFCC, and Backpropagation Neural Network

Chairisni Lubis, Felicia Gondawijaya

Open full text 8 citations

Abstract

human heart produce 2 different sound which are lub-dub. Abnormal heart sound would produce additional sound such as whoosing, roaring, rumbling or turbulence fliud noise between the normal heart sounds. In general, diagnosing heart abnormalities is relying on doctors' experience by hearing heart sound through stethoscope or using ECG. This research is using heart sound recording from The Pascal Classifying Heart Sound, Dataset B[1] as learning and testing data. Diagnosing heart sound with BFCC (Bark Frequency Cepstral Coefficients), MFCC (Mel Frequency Cepstral Coefficients), Modified BFCC, and Modified MFCC as feature extraction method and Backpropagation Neural Network as learning method. Heart sound recognition from The Pascal Classifying Heart Sound, Dataset Bwith BFCC is up to 79.167%, MFCC is up to 87.5%, Modified BFCC is up to 70.83%, and Modified MFCC is up to 95.83%.

Open-access reader

About this research paper

What this paper is about

human heart produce 2 different sound which are lub-dub. Abnormal heart sound would produce additional sound such as whoosing, roaring, rumbling or turbulence fliud noise between the normal heart sounds. In general, diagnosing heart abnormalities is relying on doctors' experience by hearing heart sound through stethoscope or using ECG. This research is using heart sound recording from The Pascal Classifying Heart Sound, Dataset B[1] as learning and testing data. Diagnosing heart sound with BFCC (Bark Frequency Cepstral Coefficients), MFCC (Mel Frequency Cepstral Coefficients), Modified BFCC, and Modified MFCC as feature extraction method and Backpropagation Neural Network as learning method. Heart sound recognition from The Pascal Classifying Heart Sound, Dataset Bwith BFCC is up to 79.167%, MFCC is up to 87.5%, Modified BFCC is up to 70.83%, and Modified MFCC is up to 95.83%.

Why it matters

OpenAlex reports 8 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

human heart produce 2 different sound which are lub-dub. Abnormal heart sound would produce additional sound such as whoosing, roaring, rumbling or turbulence fliud noise between the normal heart sounds. In general, diagnosing heart abnormalities is relying on doctors' experience by hearing heart sound through stethoscope or using ECG. This research is using heart sound recording from The Pascal Classifying Heart Sound, Dataset B[1] as learning and testing data. Diagnosing heart sound with BFCC (Bark Frequency Cepstral Coefficients), MFCC (Mel Frequency Cepstral Coefficients), Modified BFCC, and Modified MFCC as feature extraction method and Backpropagation Neural Network as learning method. Heart sound recognition from The Pascal Classifying Heart Sound, Dataset Bwith BFCC is up to 79.167%, MFCC is up to 87.5%, Modified BFCC is up to 70.83%, and Modified MFCC is up to 95.83%.

Key concepts: Mel-frequency cepstrum, Heart sounds, Speech recognition, Stethoscope, Sound (geography), Computer science, Backpropagation, Artificial neural network

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
Heart Sound Diagnose System with BFCC, MFCC, and Backpropagation Neural Network — Research Paper | ScholarLens