[ad_1]
Researchers at Cambridge University are developing a smartphone app that can be used to determine if a user has Covid-19 based on the sound of their voice, breathing, and cough. In the first step, scientists want to collect and evaluate speech, breath and cough samples using the mobile phone’s microphone. In the end, machine-oriented algorithms should be able to automatically determine disease based on sounds.
Request a lot of data
Since the Covid-19 triggered by the new coronavirus is a respiratory disease, the tones produced by those affected are very specific. “According to doctors, one of the most common things they see in patients with the virus is how they gasp when they speak, a dry cough and the intervals between their breathing patterns,” explains Cecilia Mascolo of the Cambridge Department of Computer and Technology. The professor and her team received a grant from the European Research Council of almost 2.5 million euros for the development of the application.
In late 2019, a new disease began to develop in China, COVID-19 lung disease triggered by the new SARS-CoV-2 coronavirus. Quarantine measures to contain the virus lead to production downtime and increasingly severe consequences for the economy and daily life.
“There are very few large data sets of breath sounds,” says Mascolo. Therefore, she asks for data donations. “To get better algorithms that can be used for early detection, we need as many samples from as many participants as possible.” Even if there are no crown cases, the readings can also help identify links to other complaints. The project’s research objective, which began in October 2019 and will run until 2024, is generally related to the diagnosis of respiratory diseases.
Users of the Covid-19 Sounds app also ask researchers to provide them with demographic and medical information. The app also performs a rough location once when users have generally posted location data. Information should only be stored on Cambridge University servers and used for research purposes only. Users will not be able to receive medical advice through the application.
Much is still unknown about the virus.
The additional roadmap stipulates that the development team will first carry out an initial analysis of the information collected and then release the dataset to other researchers. This should help to better understand the course of the disease and possible complications, for example, based on the patient’s history and past illnesses. “There is still a lot we don’t know about this virus and the disease it causes,” Mascolo said. “In a pandemic situation like today, it helps to get as much reliable information as possible.”
The application is currently available for Android phones in the Google Play Store, an iOS version with several other compatible languages will be released soon. The study was approved by the ethics committee of the university’s computer and technology faculty.
(mho)
