Local startup DetectNow is using AI to detect COVID-19 through cough audios

Avatar Written by Hamza Zakir ·  1 min read >

There has been no dearth of efforts made by startups, innovators, and healthcare experts around the world to identify COVID-19 and tackle its consequences. A similarly grand attempt has been made by local startup DetectNow using the power of AI.

Currently based in National Incubation Center (NIC), DetectNow’s novel AI tool is able to detect the presence of the coronavirus from cough audios. While coughing is one of the most prominent symptoms of COVID-19, it is important to distinguish between it and a regular cough.

When an individual records their cough on DetectNow’s platform, an algorithm is executed over the audio to analyze it and hence determine the likelihood of the individual suffering from COVID-19. The audio file of the individual’s cough is compared against a database of cough audios from patients who tested either positively or negatively for the disease in order to enable the system to reach a conclusion. You can check out the application in action here.

DetectNow started out as an idea pitched over the course of the CodeVsCovid19 hackathon held in March. Simon Hofer submitted the initial idea, while Zeeshan Karamat developed most of the prototype during the hackathon. These individuals went on to serve as the first project leads of DetectNow. The startup has since expanded to a team of 14 engineers, doctors, and entrepreneurs.

Not only did DetectNow’s innovative solution win the CodeVsCovid19 hackathon, but it also bagged the Solve for Humanity hackathon held in April. These accomplishments simply cement the startup’s status as a force to be reckoned with.

DetectNow aims to aid in the current pandemic era and bring a reliable COVID-19 testing tool accessible to everyone. In the long run, the company hopes to use audio-based detection technology to fight other diseases than the coronavirus as well.


Written by Hamza Zakir
Platonist. Humanist. Unusually edgy sometimes. Profile