Coughvid app will tell if you have Covid-19

covid19-cough-app-neuroscienews-public (1)

Your smartphone will soon be able to listen to your cough and predict whether you\’ve contracted COVID-19.

Coronavirus patients have a distinctive cough that sounds different than other illnesses and now researchers are building an app that can listen to coughs and use AI to tell if they have COVID-19.

Coughvid has been in development for the past month at École Polytechnique Fédérale de Lausanne and has gathered more than 15,000 audio samples of coughs to train its AI. You can log on to their website and record your cough to help the researchers.

The current oropharyngeal swab test is physically invasive and must be performed by a trained clinician which puts a strain on the resources of governments battling the spread of the virus. Ideally, testing would be performed noninvasively at no cost and administered at the homes of potential patients to minimise contamination risk.

Reports from doctors that COVID-19 patients had a cough with a distinctive sound — a chirping intake of breath at the end — that differed from other illnesses, inspired the Swiss-based researchers to use this information coupled with AI to develop the app.

The World Health Organization (WHO) has reported that 67.7% of COVID-19 patients exhibit a “dry cough”, meaning that no mucus is produced, unlike the typical “wet cough” that occurs during a cold or allergies [2]. Dry coughs can be distinguished from wet coughs by the sound they produce, which raises the question of whether the analysis of the cough sounds can give some insights about COVID-19. Such cough sounds analysis has proven successful in diagnosing respiratory conditions like pertussis [3], asthma, and pneumonia [4].

The team behind Coughvid is still collecting data to train its AI and have gathered more than 15,000 audio samples of people coughing, 1,000 of which came from people who reported being diagnosed with COVID-19. Once completed, the Coughvid app could be used as a tool to recommend whether users should seek out a coronavirus test or further treatment.

Coughvid is just one of many potential coronavirus solutions being pulled together by AI labs eager to find algorithmic solutions to the epidemic. According to the The Wall Street Journal reported last week, at least three other labs are also developing AI-powered apps that analyse subjects\’ breathing, speaking, and coughing in an attempt to predict coronavirus, partnering with researchers at schools including Carnegie Mellon University and New York University.

The app is still in early development, and public health experts are being consulted to determine how best to deploy it. Data collection will likely continue for at least two months before any product is deployed.

At the Embedded Systems Laboratory (ESL) at EPFL, they propose to leverage signal processing, pervasive computing, and machine learning to develop an Android application and website to automatically screen COVID-19 from the comfort of people’s homes. Test subjects will be able to simply download a mobile application, enter their symptoms, record an audio clip of their cough, and upload the data anonymously to their servers. They will then use audio signal processing and machine learning techniques to evaluate if there is some room for automatic or assisted COVID-19 screening.

The objective of this website is to collect a large number of sample recordings from patients that are known to have COVID-19. That’s why they are asking everybody that can provide them with a few seconds of cough sound to collaborate. It’s so easy!

Disclaimer from EPFL

All the data collected by the website is anonymous, and it is stored on a private server at the premises of the École Polytechnique Fédérale de Lausanne (EPFL), Switzerland. Each recording is associated with the timestamp in which it was received, and the geolocalisation information if the user grants the corresponding permissions.

The data will be exclusively used for research purposes, and under no circumstances will they be sold or shared with third parties. Eventually, the dataset will be made publicly available to the research community.