Automatic Classification of Large-Scale Respiratory Sound Dataset Based on Convolutional Neural Network

Minami, Koki; Lu, Huimin; Kim, Hyoungseop; Mabu, Shingo; Hirano, Yasushi; Shoji, K.

doi:10.23919/iccas47443.2019.8971689

Cited by 49 publications

(37 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some researchers combined different types of spectrogram, e.g. short-time Fourier transform (STFT) and Wavelet as proposed by Minami et al [27] or optimized S-Transformations in [28]. Although extracting good quality representative spectrograms is very important for the back-end classifier, researchers to date have not explored the settings used in this step deeply -something we aim to contribute in this paper.…”

Section: Introductionmentioning

confidence: 99%

“…Current deep learning classifiers acting on spectrograms for respiratory sound analysis are mainly based on Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), or hybrid architectures. The CNN-based systems span some diverse architectures such as LeNet6 [23], [22], VGG5 [20], two parallel VGG16s [27], and ResNet50 [28]. Inspired by the fact that respiratory indicative sounds such as Crackle and Wheeze present certain sequential characteristics, RNN-based networks have been developed in order to capture the sequential information.…”

Section: Introductionmentioning

confidence: 99%

“…By using LSTM and Gated Recurrent Unit (GRU) cells in a RNN-based network, Kochetov et al [26] proposed a novel architecture, namely the Noise Masking Recurrent Neural Network, which aimed to distinguish both noise and anomalous respiratory sounds. Regarding hybrid architectures proposed in [21], [27], a CNN was firstly used to map a spectrogram input to a temporal sequence. Then, LSTM [21] or GRU [27] layers were used to learn sequence structures before classification takes place via fully-connected layers.…”

Section: Introductionmentioning

confidence: 99%

“…Regarding hybrid architectures proposed in [21], [27], a CNN was firstly used to map a spectrogram input to a temporal sequence. Then, LSTM [21] or GRU [27] layers were used to learn sequence structures before classification takes place via fully-connected layers.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

CNN-MoE Based Framework for Classification of Respiratory Anomalies and Lung Disease Detection

Pham

Phan

Palaniappan

et al. 2021

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

This paper presents and explores a robust deep learning framework for auscultation analysis. This aims to classify anomalies in respiratory cycles and detect diseases, from respiratory sound recordings. The framework begins with front-end feature extraction that transforms input sound into a spectrogram representation. Then, a back-end deep learning network is used to classify the spectrogram features into categories of respiratory anomaly cycles or diseases. Experiments, conducted over the ICBHI benchmark dataset of respiratory sounds, confirm three main contributions towards respiratory-sound analysis. Firstly, we carry out an extensive exploration of the effect of spectrogram types, spectral-time resolution, overlapping/non-overlapping windows, and data augmentation on final prediction accuracy. This leads us to propose a novel deep learning system, built on the proposed framework, which outperforms current state-of-the-art methods. Finally, we apply a Teacher-Student scheme to achieve a trade-off between model performance and model complexity which holds promise for building real-time applications.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

CNN-MoE Based Framework for Classification of Respiratory Anomalies and Lung Disease Detection

Pham

Phan

Palaniappan

et al. 2021

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

show abstract

“…Another proposed strategy is the empirical mode decomposition method [18][19][20], which exploits instantaneous frequencies and points toward a local high-dimensional representation. In some sense, these approaches can be considered precursors of the most recent deep learning approaches [21][22][23][24]. However, the high dimensionality of these strategies impairs their statistical robustness; more importantly, deep learning provides models which can be difficult to clinically interpret.…”

Section: Introductionmentioning

confidence: 99%

Multi-Time-Scale Features for Accurate Respiratory Sound Classification

et al. 2020

View full text Add to dashboard Cite

The COVID-19 pandemic has amplified the urgency of the developments in computer-assisted medicine and, in particular, the need for automated tools supporting the clinical diagnosis and assessment of respiratory symptoms. This need was already clear to the scientific community, which launched an international challenge in 2017 at the International Conference on Biomedical Health Informatics (ICBHI) for the implementation of accurate algorithms for the classification of respiratory sound. In this work, we present a framework for respiratory sound classification based on two different kinds of features: (i) short-term features which summarize sound properties on a time scale of tenths of a second and (ii) long-term features which assess sounds properties on a time scale of seconds. Using the publicly available dataset provided by ICBHI, we cross-validated the classification performance of a neural network model over 6895 respiratory cycles and 126 subjects. The proposed model reached an accuracy of 85%±3% and an precision of 80%±8%, which compare well with the body of literature. The robustness of the predictions was assessed by comparison with state-of-the-art machine learning tools, such as the support vector machine, Random Forest and deep neural networks. The model presented here is therefore suitable for large-scale applications and for adoption in clinical practice. Finally, an interesting observation is that both short-term and long-term features are necessary for accurate classification, which could be the subject of future studies related to its clinical interpretation.

show abstract