Synthesis of Normal Heart Sounds Using Generative Adversarial Networks and Empirical Wavelet Transform

A novel heart sound segmentation algorithm via multi-feature input and neural network with attention mechanism

Objective. Heart sound segmentation (HSS), which aims to identify the exact positions of the first heart sound(S1), second heart sound(S2), the duration of S1, systole, S2, and diastole within a cardiac cycle of phonocardiogram (PCG), is an indispensable step to find out heart health. Recently, some neural network-based methods for heart sound segmentation have shown good performance.Approach. In this paper, a novel method was proposed for HSS exactly using One-Dimensional Convolution and Bidirectional Long-Short Term Memory neural network with Attention mechanism (C-LSTM-A) by incorporating the 0.5-order smooth Shannon entropy envelope and its instantaneous phase waveform (IPW), and third intrinsic mode function (IMF-3) of PCG signal to reduce the difficulty of neural network learning features.Main results. An average F1-score of 96.85 was achieved in the clinical research dataset (Fuwai Yunnan Cardiovascular Hospital heart sound dataset) and an average F1-score of 95.68 was achieved in 2016 PhysioNet/CinC Challenge dataset using the novel method.Significance. The experimental results show that this method has advantages for normal PCG signals and common pathological PCG signals, and the segmented fundamental heart sound(S1, S2), systole, and diastole signal components are beneficial to the study of subsequent heart sound classification.

Deep Learning-Based Heart Sound Analysis for Left Ventricular Diastolic Dysfunction Diagnosis

The aggravation of left ventricular diastolic dysfunction (LVDD) could lead to ventricular remodeling, wall stiffness, reduced compliance, and progression to heart failure with a preserved ejection fraction. A non-invasive method based on convolutional neural networks (CNN) and heart sounds (HS) is presented for the early diagnosis of LVDD in this paper. A deep convolutional generative adversarial networks (DCGAN) model-based data augmentation (DA) method was proposed to expand a HS database of LVDD for model training. Firstly, the preprocessing of HS signals was performed using the improved wavelet denoising method. Secondly, the logistic regression based hidden semi-Markov model was utilized to segment HS signals, which were subsequently converted into spectrograms for DA using the short-time Fourier transform (STFT). Finally, the proposed method was compared with VGG-16, VGG-19, ResNet-18, ResNet-50, DenseNet-121, and AlexNet in terms of performance for LVDD diagnosis. The result shows that the proposed method has a reasonable performance with an accuracy of 0.987, a sensitivity of 0.986, and a specificity of 0.988, which proves the effectiveness of HS analysis for the early diagnosis of LVDD and demonstrates that the DCGAN-based DA method could effectively augment HS data.

Conditional GAN based augmentation for predictive modeling of respiratory signals

Respiratory illness is the primary cause of mortality and impairment in the life span of an individual in the current COVID-19 pandemic scenario. The inability to inhale and exhale is one of the difficult conditions for a person suffering from respiratory disorders. Unfortunately, the diagnosis of respiratory disorders with the presently available imaging and auditory screening modalities are sub-optimal and the accuracy of diagnosis varies with different medical experts. At present, deep neural nets demand a massive amount of data suitable for precise models. In reality, the respiratory data set is quite limited, and therefore, data augmentation (DA) is employed to enlarge the data set. In this study, conditional generative adversarial networks (cGAN) based DA is utilized for synthetic generation of signals. The publicly available repository such as ICBHI 2017 challenge, RALE and Think Labs Lung Sounds Library are considered for classifying the respiratory signals. To assess the efficacy of the artificially created signals by the DA approach, similarity measures are calculated between original and augmented signals. After that, to quantify the performance of augmentation in classification, scalogram representation of generated signals are fed as input to different pre-trained deep learning architectures viz Alexnet, GoogLeNet and ResNet-50. The experimental results are computed and performance results are compared with existing classical approaches of augmentation. The research findings conclude that the proposed cGAN method of augmentation provides better accuracy of 92.50% and 92.68%, respectively for both the two data sets using ResNet 50 model.

Deep Learning Methods for Heart Sounds Classification: A Systematic Review

The automated classification of heart sounds plays a significant role in the diagnosis of cardiovascular diseases (CVDs). With the recent introduction of medical big data and artificial intelligence technology, there has been an increased focus on the development of deep learning approaches for heart sound classification. However, despite significant achievements in this field, there are still limitations due to insufficient data, inefficient training, and the unavailability of effective models. With the aim of improving the accuracy of heart sounds classification, an in-depth systematic review and an analysis of existing deep learning methods were performed in the present study, with an emphasis on the convolutional neural network (CNN) and recurrent neural network (RNN) methods developed over the last five years. This paper also discusses the challenges and expected future trends in the application of deep learning to heart sounds classification with the objective of providing an essential reference for further study.

Generating Synthetic Fermentation Data of Shindari, a Traditional Jeju Beverage, Using Multiple Imputation Ensemble and Generative Adversarial Networks

Fermentation is an age-old technique used to preserve food by restoring proper microbial balance. Boiled barley and nuruk are fermented for a short period to produce Shindari, a traditional beverage for the people of Jeju, South Korea. Shindari has been proven to be a drink of multiple health benefits if fermented for an optimal period. It is necessary to predict the ideal fermentation time required by each microbial community to keep the advantages of the microorganisms produced by the fermentation process in Shindari intact and to eliminate contamination. Prediction through machine learning requires past data but the process of obtaining fermentation data of Shindari is time consuming, expensive, and not easily available. Therefore, there is a need to generate synthetic fermentation data to explore various benefits of the drink and to reduce any risk from overfermentation. In this paper, we propose a model that takes incomplete tabular fermentation data of Shindari as input and uses multiple imputation ensemble (MIE) and generative adversarial networks (GAN) to generate synthetic fermentation data that can be later used for prediction and microbial spoilage control. For multiple imputation, we used multivariate imputation by chained equations and random forest imputation, and ensembling was done using the bagging and stacking method. For generating synthetic data, we remodeled the tabular GAN with skip connections and adapted the architecture of Wasserstein GAN with gradient penalty. We compared the performance of our model with other imputation and ensemble models using various evaluation metrics and visual representations. Our GAN model could overcome the mode collapse problem and converged at a faster rate than existing GAN models for synthetic data generation. Experiment results show that our proposed model executes with less error, is more accurate, and generates significantly better synthetic fermentation data compared to other models.

Special Issue on Deep Learning for Applications in Acoustics: Modeling, Synthesis, and Listening

The recent introduction of Deep Learning has led to a vast array of breakthroughs in many fields of science and engineering [...]

SynSigGAN: Generative Adversarial Networks for Synthetic Biomedical Signal Generation

Automating medical diagnosis and training medical students with real-life situations requires the accumulation of large dataset variants covering all aspects of a patient's condition. For preventing the misuse of patient's private information, datasets are not always publicly available. There is a need to generate synthetic data that can be trained for the advancement of public healthcare without intruding on patient's confidentiality. Currently, rules for generating synthetic data are predefined and they require expert intervention, which limits the types and amount of synthetic data. In this paper, we propose a novel generative adversarial networks (GAN) model, named SynSigGAN, for automating the generation of any kind of synthetic biomedical signals. We have used bidirectional grid long short-term memory for the generator network and convolutional neural network for the discriminator network of the GAN model. Our model can be applied in order to create new biomedical synthetic signals while using a small size of the original signal dataset. We have experimented with our model for generating synthetic signals for four kinds of biomedical signals (electrocardiogram (ECG), electroencephalogram (EEG), electromyography (EMG), photoplethysmography (PPG)). The performance of our model is superior wheen compared to other traditional models and GAN models, as depicted by the evaluation metric. Synthetic biomedical signals generated by our approach have been tested while using other models that could classify each signal significantly with high accuracy.