Data augmentation using Variational Autoencoders for improvement of respiratory disease classification

Abstract

Computerized auscultation of lung sounds is gaining importance today with the availability of lung sounds and its potential in overcoming the limitations of traditional diagnosis methods for respiratory diseases. The publicly available ICBHI respiratory sounds database is severely imbalanced, making it difficult for a deep learning model to generalize and provide reliable results. This work aims to synthesize respiratory sounds of various categories using variants of Variational Autoencoders like Multilayer Perceptron VAE (MLP-VAE), Convolutional VAE (CVAE) Conditional VAE and compare the influence of augmenting the imbalanced dataset on the performance of various lung sound classification models. We evaluated the quality of the synthetic respiratory sounds’ quality using metrics such as Fréchet Audio Distance (FAD), Cross-Correlation and Mel Cepstral Distortion. Our results showed that MLP-VAE achieved an average FAD of 12.42 over all classes, whereas Convolutional VAE and Conditional CVAE achieved an average FAD of 11.58 and 11.64 for all classes, respectively. A significant improvement in the classification performance metrics was observed upon augmenting the imbalanced dataset for certain minority classes and marginal improvement for the other classes. Hence, our work shows that deep learning-based lung sound classification models are not only a promising solution over traditional methods but can also achieve a significant performance boost upon augmenting an imbalanced training set.

Top-down generation of low-resolution representations improves visual perception and imagination

“…Such geometry-respecting representations are also observed in high-level brain areas such as prefrontal cortex and hippocampus [40, 41]. Developing this geometry-respecting representation is essential for semi-supervised learning [42, 43], few-shot learning [44], and data augmentation via imagination [44, 45]. Using VAE as a model of the visual system, we will study the computational advantage for low-resolution top-down generation ( Figure 1d ), and investigate the inspiration of this finding to advance the technique of sketch generation ( Figure 1e, f ).…”

“…Such geometry-respecting representations are also observed in high-level brain areas such as prefrontal cortex and hippocampus [40, 41]. Developing this geometry-respecting representation is essential for semi-supervised learning [42, 43], few-shot learning [44], and data augmentation via imagination [44, 45].…”

moBRCA-net: a breast cancer subtype classification framework based on multi-omics attention neural networks

“…For better generalization of moBRCA-net, we adopted a data augmentation based on the deep generative model to enlarge the training dataset size. Several recent papers have shown that conditional variational autoencoder (CVAE)-based data generation for certain minority classes in the imbalanced dataset improved the classification performance in various domain tasks such as respiratory disease classification [ 41 ], temporal pattern prediction based on electronic health records [ 42 ], and prediction of chemical structure based on the chemical properties [ 43 ]. We constructed a conditional variational autoencoder (CVAE) composed of two-layered encoder and decoder, which estimates the conditional distribution with latent variables and data, and generates samples for specified breast cancer subtype.…”

COVID-19 respiratory sound analysis and classification using audio textures

“…However, it creates new opportunities to develop COVID-19 screening tools for telemedicine and remote monitoring (Sharma et al 2022;Villa-Parra et al 2022). In future works, we will explore data augmentation techniques (Saldanha et al, 2022), transfer learning (Bhatt et al, 2021), and interpretable deep learning models (Joshi et al, 2021) to improve the interpretability and usability of our framework to help COVID-19 diagnosis.…”