When most of us go to the grocery store and pick up a carton of eggs or a package of chicken, we probably expect those products will not get us sick — as long as they’re cooked properly.

For the most part that’s true, but the food supply chain isn’t perfect, and contamination sometimes happens.

Each year, the United States government investigates many foodborne disease outbreaks traced to germs like salmonella, listeria, E. coli and others. The Centers for Disease Control and Prevention estimates that 48 million people fall ill annually as a result.

Of those who get sick, 128,000 are hospitalized and 3,000 die. Salmonella is one of the top culprits, and it’s often associated with eggs or chicken.

The poultry industry processes around 10 billion chickens annually and has wanted a better way to detect the bacteria for years, said Jim Dickson, a professor in Iowa State University’s animal science department.

“People are constantly looking for better ways to sample, better ways to test and get faster results,” he said. “The quicker a processor gets results, the quicker they can react to them.”

The way chicken is tested for pathogens right now is effective, with relatively few notices of contamination.

The USDA Food Safety and Inspection Service tests for the bacteria, and major poultry processors also typically test their chicken every day too, Dickson said. But testing laboratories are often off-site and preparing a sample for analysis can take a day or two.

“That’s one of the frustrating things with microbiology,” he said. “Realistically, in most cases, we’re talking about three days. By the time they get results back, that product is gone. It’s already been shipped.”

Most chickens carry some form of salmonella, and the CDC estimates it’s in one out of every 25 packages of raw chicken. When cooked properly, there’s usually not a problem, but there’s more risk if there’s a higher load of salmonella in that poultry.

In response to this challenge, a team of researchers received an initial $750,000 from the National Science Foundation Convergence Accelerator to find ways to more quickly detect the presence of salmonella in the entire chicken supply chain.

“We just want to make a safe food supply for everyone,” said Kate Trout, one of the project’s principal investigators. “(Including) rural communities, low-income communities, which we now have higher rates of salmonella infections, but we really don’t know the root causes of that.”

The research includes the development of biosensors, like the ones inside a University of Missouri lab, that significantly cut down the time it takes to detect small salmonella concentrations.

Doctoral student Mai Abuhelwa demonstrated one that uses a laser and fiber optic cables.

“I just inject my sample, wait a couple of minutes,” she said, motioning to a computer screen with a graph on it. “We can read the signal in: this point, this point and this point.”

The molecular components of salmonella will produce a certain kind of peak on the graph, said Mahmoud Almasri, Abuhelwa’s advisor and Lead Principal Investigator in this research.

“If you see a different peak, you know you are not detecting salmonella,” he said.

Almasri added they’re also creating two other sensors, including one that will be able to detect the specific kind of salmonella present in a sample.

“There are maybe 2,500 different types of salmonella,” Almasri said. “But not all of them are in poultry.”

This level of granularity would go further than the current industry standard, which only detects salmonella in general, said Kantha Channaiah, an assistant professor of food science at the University of Missouri.

“This is going to play an important role in saving a lot of resources and time,” he said. “For industries, for commercial manufacturers, time is money.”

Plus, the underlying science behind the separate sensors can work on other bacteria, like listeria, E. coli or staph, Channaiah said. “We can apply this principle and concept and we can extend the scope of this technology to other pathogens,” he said.

These sensors are not ready to be used in a chicken processor yet, but the idea behind the research is to make it possible to test for salmonella anywhere in the chicken supply chain.

But first, the researchers have to decide which of the new biosensors work best at specific points in the production chain.

“Having the biosensor developed is not the end of the question. That is not enough,” said Haitao Li, another principal investigator, and supply chain analytics department chair at the University of Missouri-St. Louis.

Li is developing a mathematical programming model to help decide which of his colleagues’ sensors are best suited for a specific company’s needs. The model can also help predict where to place the sensors at nodes in the vast poultry supply chain, he said.

A node can represent a facility, like a farm, or a process, like transportation or storage, Li said. And it can be very granular, like different steps in chicken processing.

“At every operation,” he said. “For example, carcass washing, defeathering, the cutting.”

It would be a way to test chicken many more times beyond its time at the processor, where afterward there’s an assumption it stays properly cold on the truck to the grocery store, at the grocery store and afterward, said Tim Safranski, a professor at the University of Missouri’s animal science research center.

“If the cold chain is maintained, it remains safe. If it’s properly cooked afterward, it would remain safe,” he said. “But there’s lots of opportunities along the way where things might not work correctly.”

With sensors at more points, Li said researchers can begin to see salmonella contamination along the whole supply chain in real-time. He added they plan to incorporate the testing data with others, like those on population, health and geography, into a software system that can visualize and predict the risk of the bacteria.

“If we know the level of contamination risk, we might speed up the shipment of some goods or shorten their storage time,” he said. “We might issue a recall, we might redistribute some of the product depending on the market demand or the need of different population groups.”

And catching contamination early or at more points along that chain could mean fewer people get sick, saving money. The researchers cite the statistic that illness from salmonella contamination in 2018 cost more than $10 billion in the U.S. alone.

But these sensors will need to be easy to use and cost-friendly if the companies that process, ship or distribute chicken are going to buy in. The NSF held these same priorities as well when awarding the funding.

Still, Carlton Adams likes the idea.

“It could potentially be game changing for us,” he said. “First of all, we would have better information up front.”

Adams is the chief operating officer of Operation Food Search, a food bank in St. Louis.

On most weekday mornings, cars and trucks line up outside the organization’s warehouse to load boxes of food that will go to people in need in Missouri and Illinois.

It’s imperative to him that the food they distribute to food pantries and other local agencies is high quality, dense with nutrition and reliable for the people who depend on it, Adams said.

“This notion that you don’t deserve to have good food, because you didn’t get straight As in capitalism, that’s crazy,” he said. “The knock-on effects of having contaminated food in a neighborhood is really troubling.”

Contamination can get whole families sick, Adams added, keeping kids from school or adults from work.

Plus, the sensors could help Operation Food Search throw away less food in case something goes wrong at their St. Louis facility, Adams said.

“Things happen. If we have a power outage, if the generator doesn’t kick in and we lose cold storage for a period, that food goes into the landfill,” he said. “That’s much less food that’s distributed to the community.”

There’s promise in this new testing technology, but it’s still years away from hitting the market. The research team is submitting a proposal for the next phase of funding from the NSF to be reviewed later this year, as well as seeking feedback from potential investors, program managers and researchers from other funding agencies.

In the meantime, Adams said his food bank will be ready for the sensors and software when they eventually become available.