Serving It Up
Strictly omakase
It’s been a while since my last post. Clients and real life keep demanding attention . . .
A big welcome to all my new subscribers. What can you expect to find here? Think of me as your sushi chef: There’s no fixed menu, it’s strictly omakase. I serve up whatever market-fresh material sparks my imagination.
Sometimes it’s a new scientific paper. Other times it’s an item about technology, perfume culture, or cluelessly bogus conventional wisdom. The gravitational center is always smell.
The house style is skeptical and critical—you can get your gee-whiz press release science newz at the usual pop outlets. My readers get a frank and fair assessment based on my years of experience in commercial research and academic science. Expect plenty of wasabi.
Your dish for today:
A paper from researchers in France has made a big splash—it’s been out a week and drawn nearly 17,000 views. It claims trained dogs can, with relatively high sensitivity and specificity, identify samples from COVID-19 positive patients. (The dogs sniffed nasopharyngeal swabs along with sweat and saliva samples.)
Let’s assume the study was conducted with all necessary controls, that the statistical results are watertight, and that COVID-19 infection is linked to specific volatile organic compounds (VOCs). The question then becomes: so what?
There are two possible answers.
(1) We can now move on to develop instrumental VOC sensors for quick, non-invasive COVID screening.
(2) We can now train lots of dogs to perform mass sniff-screening for COVID.
Answer (2) is impractical. Even the authors admit it takes time and money to train and maintain dogs, and using them makes sense in only certain situations. The same objection holds for dogs that can sniff out lung cancer, melanoma, diabetes, C. difficile, etc.
I prefer answer (1) because it pockets the canine results as proof-of-principle and goes to work on a mass-produced, low-cost device.
Sadly, the French team seems all-in on answer (2). I say bonne chance.
Add some wasabi: The assumption is that rapid screening is an effective way to limit the spread of COVID. Have a look at Alex Berenson’s book and get back to me on that.
Pickled ginger palette cleanser: Are we sure that dog-sniffable VOC signatures for cancer, diabetes, COVID, etc. are each unique and not a common metabolic response to disease?
Chug a Kirin Ichiban: Canine sense of smell is overrated: humans are often just as sensitive. Why don’t we have a COVID Sniff-Off between Bobby the med-tech and Fluffy the Belgian Malinois shepherd?
Dominique Grandjean, Caroline Elie, Capucine Gallet, Clotilde Julien, Vinciane Roger, Loïc Desquilbet, Guillaume Alvergnat, Séverine Delarue, Audrey Gabassi, Marine Minier, Laure Choupeaux, Solen Kerneis, Constance Delaugerre, Jérôme Le Goff, Jean-Marc Treluyer. (2022). Diagnostic accuracy of non-invasive detection of SARS-CoV-2 infection by canine olfaction. PLoS ONE 17(6): e0268382.