“We don’t know that this will be the odor of the virus, per se, or the response to the virus, or a combination,” said Otto, who is leading the project. “But the dogs don’t care what the odor is. … What they learn is that there’s something different about this sample than there is about that sample.”

A similar effort is underway at the London School of Hygiene & Tropical Medicine, where researchers previously demonstrated that dogs could identify malaria infections in humans. In a statement, James Logan, head of the school’s disease control department, called canines a “new diagnostic tool” that “could revolutionize our response to covid-19.”

Logan said Tuesday that his research team expects to begin collecting covid-19 samples “within a matter of weeks” and working with the charity Medical Detection Dogs to train canines soon after. The initial goal is to deploy six dogs to airports in the United Kingdom, he said.

“Each individual dog can screen up to 250 people per hour,” Logan said in an email. “We are simultaneously working on a model to scale it up so it can be deployed in other countries at ports of entry, including airports.”

The Working Dog Center typically trains dogs, which live with foster families, at its facility in Philadelphia, but the pandemic is forcing it to adjust. To minimize social contact, the project instead is working with Labs at a K-9 training firm in Maryland, Tactical Directional Canine, Otto said.

Miss M., Poncho and six other chocolate, yellow and black Labs began the first stage of training — learning to identify an odor for a food reward — this month, she said. Next, the dogs will train using urine and saliva samples collected from patients who tested positive and negative at the Hospital of the University of Pennsylvania and the Children’s Hospital of Philadelphia.

The following step is trickier, Otto said: learning to detect the virus in a human.

“That’s going to be the next proof of concept: Can we train them to identify it when a person has it and that person’s moving? Or even standing still?” Otto said.

Exactly how covid-19 detection dogs might be put to use in the United States would depend on demands, Otto said, though no one’s talking about stationing a dog in every hospital or testing site.

If the need is lots of tests, then Penn chemists and physicists might be able to use what they learn from the dogs to create an electronic “nose,” or sensor. The goal of the Working Dog Center’s research on ovarian-cancer-detection dogs, for example, is to produce “an electronic test where thousands and thousands of samples could be screened in a short period,” Otto said.

Other settings, such as fields where the center has trained dogs to detect the eggs of invasive spotted lantern flies, call for actual canines that can quickly roam and sniff, she said.

“The exciting area is the sort of convergence with what dogs are currently doing with [the Transportation Security Administration] and screening for explosives,” she said. “If we can do a similar approach for screening humans, then there will be a large interest” in using dogs to help flag people for testing, she added.

“We don’t have enough detection dogs. And if now, all of a sudden, everyone wants a covid detection dog? It’s going to be a challenge to figure out where are the priorities,” Otto said. “But there’s a lot of opportunity.”

Read more: