Back

Effective community screening for asymptomatic and symptomatic COVID-19 with a fast and extremely low cost COVID-Anosmia Checker tool

Basu, B.; Riya, P. A.; Issac, J.; Parvathy, S.; Nair, B. S.; Tokdar, P.; Kumar, D. S.; Kulkarni, P. R.; Hanumanram, G.; Jagadeesan, M.; Suthakaran, P. K.; Nair, L. D. V.; Vennila, R.; Kannan, R.; Kaimal, B.; Anoop, G. K.; Joseph, I.; Nair, R.; George, S.; James, J.; Pillai, M. R.

2020-11-03 public and global health
10.1101/2020.10.28.20221200 medRxiv
Show abstract

BackgroundCOVID-19 curve can be flattened by adopting mass screening protocols with aggressive testing and isolating infected populations. The current approach largely depends on RT-PCR/rapid antigen tests that require expert personnel resulting in higher costs and reduced testing frequency. Loss of smell is reported as a major symptom of COVID-19, however, a precise olfactory testing tool to identify COVID-19 patient is still lacking. MethodsTo quantitatively check for the loss of smell, we developed an odor strip, "COVID-Anosmia checker", spotted with gradients of coffee and lemon grass oil. We validated its efficiency in healthy and COVID-19 positive subjects. A trial screening to identify SARS-CoV-2 infected persons was also carried out to check the sensitivity and specificity of our screening tool. FindingsIt was observed that COVID positive participants were hyposmic instead of being anosmic when they were subjected to smelling higher odor concentration. Our tool identified 97% of symptomatic and 94% of asymptomatic COVID-19 positive subjects after excluding most confounding factors like concurrent chronic sinusitis. Further, it was possible to reliably predict COVID-19 infection by calculating a loss of smell score with 100% specificity. We coupled this tool with a mobile application, which takes the input response from the user, and can readily categorize the user in the appropriate risk groups. ConclusionLoss of smell can be used as a reliable marker for screening for COVID-19. Our tool can be used for first-line screening to trace out COVID-19 infection effectively. It can be used in difficult to reach geographical locations.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.