Innovations in Food Safety: A Q&A with Data Science Experts on Predictive Analytics

By the bioMérieux Connection Editors

Is the food manufacturing industry ready for the data revolution? Our country’s biggest regulatory bodies seem to think so—in July of this year, the FDA announced The New Era of Smarter Food Safety Blueprint, a new initiative that introduces modern approaches for modern times. The blueprint outlines the approach the FDA will take over the next decade, leveraging technology and other tools to create a safer, more digital, traceable food system.

A major aspect of the FDA’s plan includes embracing predictive analytics to help find root causes of problems and avoid identified risks. Traditional microbiology methods and more modern diagnostic solutions have generated data streams for decades, and the FDA is ready to explore the preventive value in the collected data.

Many other areas of the food industry have embraced predictive technology—for example, grocery stores utilize data science to generate personalized offers and tailored pricing to customers, even going so far as to use infrared body-heat sensors to track how customers navigate stores and predict how many cashiers are needed. The FDA and NASA use Geographical Information System (GIS) Technology to assess environmental risks for microbial contamination of crops, helping growers predict when and in which part of their farms contamination may occur.

We spoke with three food safety and data science experts to explore the potential of predictive analytics in food manufacturing and the impact it can have on product quality and consumer safety.


Meet The Experts

Dr. Hannes Pouseele, one of the minds behind the CDC’s Pulsenet, is a Senior Expert in Bioinformatics and Data Science at bioMérieux.

Dr. Nicholas Siciliano, a scientific innovator and entrepreneur, is the CEO of Invisible Sentinel.

Dr. Vik Dutta, an inventor, diagnostician, and renowned microbiologist, is a Senior Staff Scientist at bioMérieux, Inc.

How can predictive analytics and modeling help food manufacturers?

Dr. Pouseele: We can start taking some of the randomness out of the events that seem to happen to us. If you think about pathogen contamination in a product, you can think of it as something that happens to you and you have no control over it. But actually, what we’re starting to see: these events are not just random. Pathogens don’t just turn up in your factory one day and contaminate your product. There’s a system and many factors happening at the same time, and data science is there to help the human brain figure out what is going on.”

Dr. Siciliano: “We all aspire to prevent foodborne (illness) outbreaks. The overarching goal is to make sure these types of things don’t happen. That’s ultimately what’s going to help consumers and (food manufacturers) from any associated risks of recalls.”

Dr. Pouseele: “It’s not just about foodborne illness. It’s about quality. It’s about spoilage and shelf life. I think those are key areas that will be very important for food manufacturers to increase productivity and make more food.”

Why is now a good time for the food industry to embrace predictive technologies and data science?

Dr. Siciliano: “With the FDA’s new initiative, the timing is critical.”

Dr. Dutta: “Regulators are continuing to evolve their mindsets, strategies, and their ideas of what the world will look like in the next decade or so.”

Dr. Pouseele: “We are at a time where we have collected many single data points and have made daily decisions on just one data point. I think now we are in a position where we have enough data to start doing something more than that.”

Dr. Dutta: “Traditional microbiology is sort of a very 19th century concept we carried into the 20th century. Fast forward—we are now in the age of data: the data revolution. How we utilize and collect information is more critical.”

How do traditional microbiology methods, tools, and diagnostics fit into the new future of food safety?

Dr. Pouseele: “The marriage between food manufacturing, data science, and traditional microbiology goes in all three directions.”

Dr. Dutta: “The association of data points with certain events defines how much data we need to be collecting. In other words, if there is a strong association, it has been found you don’t need to collect a lot of data. That takes me back to traditional microbiology—food safety scientists have been collecting data points for many decades now. We just haven’t ever looked at the data as one entity.”

Dr. Pouseele: “How can we create a simple test that is based on bioinformatics and data science that helps us control the problem? That takes us back to traditional microbiology testing.”


The food system is undergoing major changes as data science, new technologies, and new foods disrupt the way manufacturers tackle food safety and quality. New approaches like predictive analytics are quietly but unquestionably transforming food science labs and food manufacturing facilities. In the long-term, predictive scientific data has the potential to not only protect consumers from foodborne illness, but to improve product quality.


Opinions expressed in this article are not necessarily those of bioMérieux, Inc.

Leave a Reply

Your email address will not be published. Required fields are marked *