CovidEnvelope: A Fast Automated Approach to Diagnose COVID-19 from Cough Signals

Abstract

The COVID-19 pandemic has a devastating impact on the health and well-being of global population. Cough audio signals classification showed potential as a screening approach for diagnosing people, infected with COVID-19. Recent approaches need costly deep learning algorithms or sophisticated methods to extract informative features from cough audio signals. In this paper, we propose a low-cost envelope approach, called CovidEnvelope, which can classify COVID-19 positive and negative cases from raw data by avoiding above disadvantages. This automated approach can pre-process cough audio signals by filter-out background noises, generate an envelope around the audio signal, and finally provide outcomes by computing area enclosed by the envelope. It has been seen that reliable datasets are also important for achieving high performance. Our approach proves that human verbal confirmation is not a reliable source of information. Finally, the approach reaches highest sensitivity, specificity, accuracy, and AUC of 0.92, 0.87, 0.89, and 0.89 respectively. The automatic approach only takes 1.8 to 3.9 minutes to compute these performances. Overall, this approach is fast and sensitive to diagnose the people living with COVID-19, regardless of having COVID-19 related symptoms or not, and thus have vast applicability in human well-being by designing HCI devices incorporating this approach.

Exploring Machine Learning Strategies in COVID-19 Prognostic Modelling: A Systematic Analysis of Diagnosis, Classification and Outcome Prediction

“…Researchers such as Hossain and colleagues have proposed signal processing and ML techniques to analyse cough signals and identify unique features related to COVID-19 infection [50]. Their automated approach, called Covidenvelope , demonstrated a classification accuracy of 95.10%, a sensitivity of 94.64%, and a specificity of 96.32% for COVID-19 diagnosis.…”

A ResNet attention model for classifying mosquitoes from wing-beating sounds

“…As male mosquitoes have a higher wing-beating rate than female mosquitoes 46 , our model can be widely used to classify females mosquitoes, which are solely responsible for transmitting pathogens. Besides mosquitoes, the current model will serve as a baseline model to classify other insect-pest species, based on their unique sound features, e.g.—wing-beating, movement, feeding sounds, etc, and can be adopted for other audio-based classification tasks 47 . Lastly, the current study is based on audios, and we can only detect one species at one time.…”

Machine learning to analyse omic-data for COVID-19 diagnosis and prognosis

“…Coronaviruses, particularly those of the genus beta coronavirus (e.g. Middle East Respiratory Syndrome Coronavirus, aka MERS-CoV and Severe Acute Respiratory Syndrome Coronavirus, aka SARS-CoV), are highly pathogenic agents of respiratory disease whose highly variable genetic diversity and diverse host-adaptive features make them lethal and devastating worldwide [1]. This diversity has led to the global spread and transmission to millions of people, with more than 6.4 million deaths from COVID-19 (SARS-CoV-2) worldwide [2].…”

“…detect the onset of diseases, which cannot be visible in physiological responses or behaviours. Previous studies[1,7] used different physiological datasets, which is cost-effective, but often lack the reliability of onset of the disease. Khanday et al[25] build models based on clinical text, which sophisticated feature engineering and can be improved further with deep learning methods.…”