Search

The alarmingly high mortality rate and increasing global prevalence of cardiovascular diseases (CVDs) signify the crucial need for early detection schemes. Phonocardiogram (PCG) signals have been historically applied in this domain owing to its simplicity and cost-effectiveness. In this article, we propose CardioXNet, a novel lightweight end-to-end CRNN architecture for automatic detection of five classes of cardiac auscultation namely normal, aortic stenosis, mitral stenosis, mitral regurgitation and mitral valve prolapse using raw PCG signal. The process has been automated by the involvement of two learning phases namely, representation learning and sequence residual learning. Three parallel CNN pathways have been implemented in the representation learning phase to learn the coarse and fine-grained features from the PCG and to explore the salient features from variable receptive fields involving 2D-CNN based squeeze-expansion. Thus, in the representation learning phase, the network extracts efficient time-invariant features and converges with great rapidity. In the sequential residual learning phase, because of the bidirectional-LSTMs and the skip connection, the network can proficiently extract temporal features without performing any feature extraction on the signal. The obtained results demonstrate that the proposed end-to-end architecture yields outstanding performance in all the evaluation metrics compared to the previous state-of-the-art methods with up to 99.60% accuracy, 99.56% precision, 99.52% recall and 99.68% F1- score on an average while being computationally comparable. This model outperforms any previous works using the same database by a considerable margin. Moreover, the proposed model was tested on PhysioNet/CinC 2016 challenge dataset achieving an accuracy of 86.57%. Finally the model was evaluated on a merged dataset of Github PCG dataset and PhysioNet dataset achieving excellent accuracy of 88.09%. The high accuracy metrics on both primary and secondary dataset combined with a significantly low number of parameters and end-to-end prediction approach makes the proposed network especially suitable for point of care CVD screening in low resource setups using memory constraint mobile devices.

This paper proposes an unsupervised way for Phonocardiogram (PCG) analysis, which uses a revised auto encoder based on distribution density estimation in the latent space. Auto encoders especially Variational Auto-Encoders (VAEs) and its variant β−VAE are considered as one of the state-of-the-art methodologies for PCG analysis. VAE based models for PCG analysis assume that normal PCG signals can be represented by latent vectors that obey a normal Gaussian Model, which may not be necessary true in PCG analysis. This paper proposes two methods DBVAE and DBAE that are based on estimating the density of latent vectors in latent space to improve the performance of VAE based PCG analysis systems. Examining the system performance with PCG data from the a single domain and multiple domains, the proposed systems outperform the VAE based methods. The representation of normal PCG signals in the latent space is also investigated by calculating the kurtosis and skewness where DBAE introduces normal PCG representation following Gaussian-like models but DBVAE does not introduce normal PCG representation following Gaussian-like models.

Deep Learning based heart sound classification is of significant interest in reducing the burden of manual auscultation through the automated detection of signals, including abnormal heartbeats. This work presents a method for classifying phonocardiogram (PCG) signals as normal or abnormal by applying a deep Convolutional Neural Network (CNN) after transforming the signals into 2D color images. In particular, a new methodology based on fractal theory, which exploits Partitioned Iterated Function Systems (PIFS) to generate 2D color images from 1D signals is presented. PIFS have been extensively investigated in the context of image coding and indexing on account of their ability to interpolate and identify self-similar features in an image. Our classification approach has shown a high potential in terms of noise robustness and does not require any pre-processing steps or an initial segmentation of the signal, as instead happens in most of the approaches proposed in the literature. In this preliminary work, we have carried out several experiments on the database released for the 2016 Physionet Challenge, both in terms of different classification networks and different inputs to the networks, thus also evaluating the data quality. Among all experiments, we have obtained the best result of 0.85 in terms of modified Accuracy (MAcc). [Display omitted] • PIFS encode self-similarities in a 1D signal and transcode them into a 2D image. • The transcoding is noise-resistant and enables segmentation-free analysis of PCG. • The 2D image is a very compact representation of the signal that can feed a CNN. [ABSTRACT FROM AUTHOR]

This paper proposes an unsupervised way for Phonocardiogram (PCG) analysis, which uses a revised auto encoder based on distribution density estimation in the latent space. Auto encoders especially Variational Auto-Encoders (VAEs) and its variant β-VAE are considered as one of the state-of-the-art methodologies for PCG analysis. VAE based models for PCG analysis assume that normal PCG signals can be represented by latent vectors that obey a normal Gaussian Model, which may not be necessary true in PCG analysis. This paper proposes two methods DBVAE and DBAE that are based on estimating the density of latent vectors in latent space to improve the performance of VAE based PCG analysis systems. Examining the system performance with PCG data from the a single domain and multiple domains, the proposed systems outperform the VAE based methods. The representation of normal PCG signals in the latent space is also investigated by calculating the kurtosis and skewness where DBAE introduces normal PCG representation following Gaussian-like models but DBVAE does not introduce normal PCG representation following Gaussian-like models., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2021 Li and Tian.)

This study concerns the task of automatic structural heart abnormality risk detection from digital phonocardiogram (PCG) signals aiming at pediatric heart disease screening applications. Recently, various systems based on convolutional neural networks trained on time-frequency representations of segmental PCG frames have been presented that outperform systems using hand-crafted features. This study focuses on the segmentation and time-frequency representation components of the CNN-based designs. We consider the most commonly used features (MFCC and Mel-Spectrogram) used in state-of-the-art systems and a time-frequency representation influenced by domain-knowledge, namely sub-band envelopes as an alternative feature. Via tests carried on two high quality databases with a large set of possible settings, we show that sub-band envelopes are preferable to the most commonly used features and period synchronous windowing is preferable over asynchronous windowing., (Copyright © 2018 Elsevier Ltd. All rights reserved.)