Wadhwani Institute for Artificial Intelligence
Participant Information for
“Collection of Solicited Cough Sounds and Symptoms to Screen for Early Signs of COVID-19"
PURPOSE OF THE STUDY
The purpose of the study is to collect symptom history and audio recordings of solicited (i.e., voluntary) cough sounds from diverse participants. These will be used to develop algorithms that could identify early signs of COVID-19. Note that while it is established that a person’s cough pattern and type of cough are indicators of their lung health, this is an exploratory project.
STEPS TO BE FOLLOWED
The study participant will first be asked to complete a study questionnaire which will include questions about basic demographics and symptom history.
The participant will be asked to submit an image or pdf of their COVID-19 test result. The participant can submit a redacted image or pdf of their lab report (with private information hidden or blacked out), or share the name of the facility they got tested. In case the participant has not been tested, or has not received their test results yet, they can skip this step.
The participant will then be asked to record samples of solicited (i.e., voluntary) coughs as well as a short speech sample.
You may refuse to participate or withdraw at any time.
No individual benefits are expected for subjects who participate in this study. However, this research project (if successful) could help potentially millions of people with limited access to testing facilities get an early diagnosis.
Because the study involves producing sounds like a cough, there is a risk of spreading germs. To reduce this risk, we will ask the study participant to produce the sounds away from other humans and disinfect the recording device and nearby surfaces when the study is complete. The participant can wear a mask while recording if they prefer.
Information about the study participant and all audio recordings are confidential.
The researchers will listen to the audio recordings to verify the type of sound in the recording, and add labels that correspond to the start and end points of certain audio portions. This data will be used by our research team to develop and improve the detection and classification algorithms. After validation of submissions and anonymisation, we will make this dataset open access and easily available to other researchers. Any sharing of data will be managed by the Wadhwani AI research team.
The study team may want to contact the study participant about taking part in additional data collection sessions. The study participant is not obligated to participate in these.
If you have questions about the study, you may contact the study team at Wadhwani AI (firstname.lastname@example.org)