Api Hasthanasombat (KC 2017) is one of the developers behind the COVID-19 Sounds App project, recently launched to collect data and develop machine learning algorithms that could help inform the diagnosis of COVID-19 based on the sounds of the patient’s voice, breathing and coughing.
Headed by Api’s supervisor Cecilia Mascolo, Professor of Mobile Systems in the Department of Computer Science and Technology, the study is part of a larger Mobile Health Diagnostics with Audio initiative and is partly funded by the European Research Council through Project EAR. Their ultimate plan is to provide the code as open source to the wider research community hoping to facilitate more activity in this new area.
To crowdsource audio data, the group has launched a webform for Chrome and Firefox browsers, as well as mobile applications for both Android and iOS, all of them translated to several languages. The data collected – basic demographic and medical information as well as voice, breathing and coughing samples – will be stored on University servers and used solely for research purposes. Once the first results are available the team will release the dataset to other researchers. Api says:
Being part of this team has allowed me to re-think the possible applications of my research in systems and causality to the healthcare domain. I see numerous opportunities in the future for personalised analytics and interventional platforms that are at the intersection of causal reasoning and robust systems design.
The COVID-19 Sounds team need as many samples from as many participants as they can get, so do try out the form and download the app on https://www.covid-19-sounds.org/en/
