TY - GEN
T1 - Short-segment Heart Sound Classification Using an Ensemble of Deep Convolutional Neural Networks
AU - Noman, Fuad
AU - Ting, Chee-Ming
AU - Salleh, Sh-Hussain
AU - Ombao, Hernando
N1 - KAUST Repository Item: Exported on 2020-10-01
PY - 2019/4/17
Y1 - 2019/4/17
N2 - This paper proposes a framework based on deep convolutional neural networks (CNNs) for automatic heart sound classification using short-segments of individual heart beats. We design a 1D-CNN that directly learns features from raw heart-sound signals, and a 2D-CNN that takes inputs of two-dimensional time-frequency feature maps based on Mel-frequency cepstral coefficients. We further develop a time-frequency CNN ensemble (TF-ECNN) combining the 1D-CNN and 2D-CNN based on score-level fusion of the class probabilities. On the large PhysioNet CinC challenge 2016 database, the proposed CNN models outperformed traditional classifiers based on support vector machine and hidden Markov models with various hand-crafted time- and frequency-domain features. Best classification scores with 89.22% accuracy and 89.94% sensitivity were achieved by the ECNN, and 91.55% specificity and 88.82% modified accuracy by the 2D-CNN alone on the test set.
AB - This paper proposes a framework based on deep convolutional neural networks (CNNs) for automatic heart sound classification using short-segments of individual heart beats. We design a 1D-CNN that directly learns features from raw heart-sound signals, and a 2D-CNN that takes inputs of two-dimensional time-frequency feature maps based on Mel-frequency cepstral coefficients. We further develop a time-frequency CNN ensemble (TF-ECNN) combining the 1D-CNN and 2D-CNN based on score-level fusion of the class probabilities. On the large PhysioNet CinC challenge 2016 database, the proposed CNN models outperformed traditional classifiers based on support vector machine and hidden Markov models with various hand-crafted time- and frequency-domain features. Best classification scores with 89.22% accuracy and 89.94% sensitivity were achieved by the ECNN, and 91.55% specificity and 88.82% modified accuracy by the 2D-CNN alone on the test set.
UR - http://hdl.handle.net/10754/655968
UR - https://ieeexplore.ieee.org/document/8682668/
UR - http://www.scopus.com/inward/record.url?scp=85068993834&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2019.8682668
DO - 10.1109/ICASSP.2019.8682668
M3 - Conference contribution
SN - 9781479981311
SP - 1318
EP - 1322
BT - ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
PB - IEEE
ER -