Analysis of coughs for the identification of COVID-19 cases automatically
The ongoing COVID-19 pandemic has put healthcare systems worldwide to the test. In a bid to combat the virus, a team of researchers led by Adria Mallol and Helena Cuesta have submitted a system to the Cough Sound Track of the Diagnosing COVID-19 using Acoustics (DiCOVA) Challenge at the Interspeech 2021 international congress.
Their project, titled "Cough-based COVID-19 Detection with Contextual Attention Convolutional Neural Networks and Gender Information", was published in the Proceedings of Interspeech. The research investigates whether the patient's gender is a factor in analysing coughs for COVID-19 detection.
The team developed and evaluated two different neural networks that predict positive or negative COVID-19 using one second of audio. These models use the spectrogram, a time-frequency representation of the audio signal as input. The research findings suggest that coughs generated by a man and those generated by a woman are not necessarily equivalent from the point of view of the spectrogram.
The models that incorporate information on the patient's gender obtain better results in their predictions in most scenarios. This could be a significant step forward in mass population screening, a necessary measure to detect positive cases and break virus transmission chains.
The experiments were conducted using the Coswara dataset provided by the Cough Sound Track - DiCOVA Challenge. This dataset contains 1,040 audio recordings of coughing, with metadata including positive/negative for COVID-19, the individual's gender, and nationality.
The project is part of the sustAGE (ID 826506) and TROMPA (ID 770376) projects, funded by the EU's Horizon 2020 research and innovation programme and the AGAUR, Catalan Government.
Previous systems based on AI have proven effective at detecting coughing, sneezing, and respiratory anomalies. This new research offers clues for future steps in the development of AI-based systems for COVID-19 detection. Access to vaccines against covid-19 has been making the situation more stable, but the importance of early detection and screening remains crucial in managing the pandemic.
