Detecting COVID-19 from Breathing and Coughing Sounds using Deep Neural Networks

Abstract

The COVID-19 pandemic has affected the world unevenly; while industrial economies have been able to produce the tests necessary to track the spread of the virus and mostly avoided complete lockdowns, developing countries have faced issues with testing capacity. In this paper, we explore the usage of deep learning models as a ubiquitous, low-cost, pre-testing method for detecting COVID-19 from audio recordings of breathing or coughing taken with mobile devices or via the web. We adapt an ensemble of Convolutional Neural Networks that utilise raw breathing and coughing audio and spectrograms to classify if a speaker is infected with COVID-19 or not. The different models are obtained via automatic hyperparameter tuning using Bayesian Optimisation combined with HyperBand. The proposed method outperforms a traditional baseline approach by a large margin. Ultimately, it achieves an Unweighted Average Recall (UAR) of 74.9 %, or an Area Under ROC Curve (AUC) of 80.7 % by ensembling neural networks, considering the best test set result across breathing and coughing in a strictly subject independent manner. In isolation, breathing sounds thereby appear slightly better suited than coughing ones (76.1 % vs 73.7 % UAR).

A New Model to Detect COVID-19 Coughing and Breathing Sound Symptoms Classification from CQT and Mel Spectrogram Image Representation using Deep Learning

“…Table II analyses the performance of several techniques in terms of accuracy. The authors in [23,24] employed a small dataset that included the actual COVID-19 coughing sound sample in a comparable investigation. Much of the prior study has been on distinguishing between coughing and noncoughing tones.…”

“…This paper concentrated on categorizing and identifying breathing and coughing sounds caused by COVID-19 virus infected individuals. Schuller et al [23] used CNN to create a deep learning strategy to identify raw breathing as well as coughing in COVID-19 patients. Researchers improved the CNN method, which employs breathing as well as coughing sounds to test whether a person has COVID-19 or is fit.…”

Efficient and Explainable Deep Neural Networks for Airway Symptom Detection in Support of Wearable Health Technology

“…have been used as digital audio biomarkers for early disease detection or predicting acute exacerbations in airway diseases such as asthma, chronic obstructive pulmonary diseases (COPD), and COVID-19. [2][3][4] Most wearable health devices for airway diseases are built on audio sensing technology to detect aforesaid airway symptoms with an embedded microphone. [5] These acoustic microphones are often omnidirectional and capture both, a speaker's voice and surrounding sounds.…”

“…[5] Cough detection is also helpful in predicting COVID-19 infection. [4] However, these deep neural networks (DNNs) have barely been optimized for wearable devices. Further, not many algorithms are capable of multiclass classification in detecting more than one airway symptom [18] Also, given the black-box-character of AI algorithms, explainable AI has been advocated to increase trust among users and decisionmakers, [19][20][21][22] especially in the development of health wearable devices.…”

LSTM-Based COVID-19 Detection Method Using Coughing

“…As the other example of software-based COVID-19 detection methods, in [12], MFCC, rst order MFCC, second order MFCC, and zero crossing rate (ZCR) features are extracted from breathing, coughing, and speech signals and then put these features as input into a neural network for COVID-19 detection purposes. In [13], raw audio signals are converted into spectrogram images, which are used to train their proposed machine learning algorithm to detect COVID-19. Machine learning algorithms have been applied to detect COVID-19.…”