Phonocardiogram (PCG) signal is used as a diagnostic test in ambulatory monitoring in order to evaluate the heart hemodynamic status and to detect a cardiovascular disease. The objective of this study is to develop an automatic classification method for anomaly (normal vs. abnormal) and quality (good vs. bad) detection of PCG recordings without segmentation. For this purpose, a subset of 18 features is selected among 40 features based on a wrapper feature selection scheme. These features are extracted from time, frequency, and time-frequency domains without any segmentation. The selected features are fed into an ensemble of 20 feedforward neural networks for classification task. The proposed algorithm achieved the overall score of 91.50% (94.23% sensitivity and 88.76% specificity) and 85.90% (86.91% sensitivity and 84.90% specificity) on the train and unseen test datasets, respectively. The proposed method got the second best score in the PhysioNet/CinC Challenge 2016.

http://www.cinc.org/archives/2016/pdf/180-213.pdf

Heart sound anomaly and quality detection using ensemble of neural networks without segmentation

Heart sound classification based on improved MFCC features and convolutional recurrent neural networks.

Cardiovascular diseases currently pose the highest threat to human health around the world. Proper investigation of the abnormalities in heart sounds is known to provide vital clinical information that can assist in the diagnosis and management of cardiac conditions. However, despite significant advances in the development of algorithms for automated classification and analysis of heart sounds, the validity of different approaches has not been systematically reviewed. This paper provides an in-depth systematic review and critical analysis of all the existing approaches for automatic identification and classification of the heart sounds. All statements on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2009 Checklist were followed and addressed thoroughly to maintain the quality of the accounted systematic review. Out of 1347 research articles available in the academic databases from 1963 to 2018, 117 peer-reviewed articles were found to fall under the search and selection criteria of this paper. Amongst them: 53 articles are focused on segmentation, 72 of the studies are related to the feature extraction approaches and 88 to classification, and 56 reported on the databases and heart sounds acquisition. From this review, it is clear that, although a lot of research has been done in the field of automated analysis, there is still some work to be done to develop robust methods for identification and classification of various events in the cardiac cycle so that this could be effectively used to improve the diagnosis and management of cardiovascular diseases in combination with the wearable mobile technologies.

Algorithms for Automatic Analysis and Classification of Heart Sounds–A Systematic Review

A study of time-frequency features for CNN-based automatic heart sound classification for pathology detection.

First, the spectrograms of heart cycles are scaled for comparison.Second, tensor decomposition is utilized to the scaled spectrograms.Third, the intrinsic structure information of scaled spectrograms is extracted.Fourth, more useful physiological and pathological information is reserved.Fifth, the extracted features are more discriminative. Heart sound signal analysis is an effective and convenient method for the preliminary diagnosis of heart disease. However, automatic heart sound classification is still a challenging problem which mainly reflected in heart sound segmentation and feature extraction from the corresponding segmentation results. In order to extract more discriminative features for heart sound classification, a scaled spectrogram and tensor decomposition based method was proposed in this study. In the proposed method, the spectrograms of the detected heart cycles are first scaled to a fixed size. Then a dimension reduction process of the scaled spectrograms is performed to extract the most discriminative features. During the dimension reduction process, the intrinsic structure of the scaled spectrograms, which contains important physiological and pathological information of the heart sound signals, is extracted using tensor decomposition method. As a result, the extracted features are more discriminative. Finally, the classification task is completed by support vector machine (SVM). Moreover, the proposed method is evaluated on three public datasets offered by the PASCAL classifying heart sounds challenge and 2016 PhysioNet challenge. The results show that the proposed method is competitive.

Shi-Wen Deng

Papers

Towards heart sound classification without segmentation via autocorrelation feature and diffusion maps