Key result
A stacked autoencoder and a 1-D convolutional neural network achieved higher accuracy, sensitivity, and specificity in recognizing S1 and S2 heart sounds compared to well-known classifiers.
Answers are grounded in this paper and cite their sources.
A novel deep learning approach using stacked autoencoders and 1-D CNNs improves the automated identification of S1 and S2 heart sounds without relying on ECG reference or timing intervals.
This paper proposes a new technique for the identification of fundamental heart sounds (HSs), namely, S1and S2from the cardiac cycle, the first and foremost step in the automated HSs analysis for the detection of pathological events, without incorporating time-interval informations between S1and S2or electrocardiogram signal as reference. The motive of this paper is to demonstrate that the reliable S1and S2classification performances based on the combinatory feature (CF) derived from higher order moments and cepstral-based domain can still be achieved, under the circumstances where the timing interval information might not be easily understood due to cardiac abnormalities. Using deep neural networks approach, a stacked autoencoder (SAE) based on the CF is proposed for the classification of fundamental HSs. Experiments are conducted on both publicly available and recorded HSs signals for the validation of the proposed method. The SAE using the proposed CF achieves better classification results in recognizing S1and S2in comparison to well-known classifiers such as deep belief neural network, support vector machine, Naive Bayes, linear discriminant analysis, and boosting ensemble. The proposed method shows higher classification rate in terms of accuracy, sensitivity, and specificity by considering CF, which uses Mel-frequency cepstral coefficients and its derivative features. A second approach for addressing the problem of S1and S2identifications is carried out by employing 1-D convolutional neural network that uses the signals directly to learn the relevant features by its own for the recognition.
No takes yet. Share an insight, caveat, or question.
Mishra et al. (2018) studied Heart sound classification. Stacked autoencoder (SAE) and 1-D convolutional neural network vs. Deep belief neural network, support vector machine, Naive Bayes, linear discriminant analysis, and boosting ensemble was evaluated on Classification of fundamental heart sounds (S1 and S2). A stacked autoencoder and a 1-D convolutional neural network achieved higher accuracy, sensitivity, and specificity in recognizing S1 and S2 heart sounds compared to well-known classifiers.
Synapse has enriched 10 related papers. Consider them for comparative context: