The CoVawe application is inspired from TimBre. TimBre is a Tuberculosis screening application, developed by Docturnal, where the sound of cough of an individual is recorded by a medical practitioner along with their demographic, clinical and socio-economic variables and processed in real time, leveraging machine learning to detect if the cough is TB positive or negative. To combat the COVID-19 pandemic, TimBre has been repurposed for pre-screening of COVID-19 as a point-of-care solution for active case finding through cough signatures and other biomarkers, relevant to specific lung condition.

It is currently available as a pre-screening tool and Docturnal is currently working to obtain labelled COVID-19 datasets to make it available as a screening tool approximately 1 month afterwards. It can be used for both symptomatic and asymptomatic cases and provides a screening solution from the comfort of one’s home (DIY) or with a microphone array at a clinical setup, with both alternatives requiring certain protocol to ensure safety and minimise the infectious nature of the virus.

This solution takes less than five minutes to provide results. These are binary results (yes and no) to asses if the subject is Covid-19 positive or negative. CoVawe has a patent applied for our TB solution with the Indian Patent Office and provisional patents have been filed for both SPO2 and RR detection in real time.

The areas this solution aims to address is 

• Reduction of burden on healthcare centers
• Active Case Finding to channelize the traffic to healthcare centers
• Screening solution that is affordable and accessible to all We have multiple discussions with prospects on the deployment of the solution.

The solution is available in two variants:

1. DIY (from the comfort of one’s home)
2. Clinical Setting (at a medical facility) “CoVawe” Simple & easy to use screening solution, that works on a mobile phone and can be conducted from the comfort of one’s home anytime anywhere, leveraging Acoustics (spectral analysis), Machine Learning & offer near real time results.

It is built on the scientific foundation we have gained from our ready solution for Pulmonary Tuberculosis. For “coVawe” we have conducted a global pre-beta release & know that the achieved sensitivity & specificity will significantly increase once labelled data is acquired. Docturnal has a wide spectrum of training data being collected since 2016 encompassing almost all lung disorders, not limited to Pneumonia, Asthma, TB, COPD, Emphysema, Pleural Effusion, Controls etc. that make the Machine Learning Model accurate without bias. We intend to expose the API for integration with Ecosystem Partners across the Globe.

Once available on the market, Docturnal will provide support for Android, iOS & Integration readiness. After installation, the subject will register, input their demographic and clinical variables, finally record the sound of their cough. After which the result will be provided to them in less than five minutes, which can also be shared with a medical practitioner as well.