In an increasingly competitive global marketplace, high-quality, flavorful protein is no longer a luxury—it is a consumer expectation. But how can you objectively measure it?
PIC unveiled Pork Chop Studio, a new technology using advanced imaging and artificial intelligence to help deliver on that expectation, at World Pork Expo. By combining advanced imaging with artificial intelligence, PIC wants to move beyond the limitations of human subjectivity to deliver a more consistent and predictable product for producers and consumers alike.
How Can AI Measure Pork Quality?
For years, the industry has relied on expert-trained evaluators to manually assign marbling and color scores and specialized tools to measure attributes like pH, color and tenderness, explains Dan Hamilton, PIC director of product performance. While these experts are highly skilled, visual scoring is inherently subjective.
He believes this can be a tool to get more data automatically collected to predict what ultimately the consumer sees as a quality pork chop.
“If you gave the same loin image to a human scorer 10 minutes apart, you could easily get different results depending on mood, focus or fatigue,” says Eric Psota, PIC digital innovation senior manager.
The Pork Chop Studio eliminates this variability. A custom imaging station equipped with a high-end 4K camera and studio lighting, captures fine details—such as muscle fiber striations—that are nearly invisible to the human eye, Psota says. These images are then processed by a neural network trained on thousands of expert-scored samples to provide an objective, “expert-level” score every time.
“Historically we’ve always collected this data,” points out Clay Eastwood, PIC applied meat scientist. “But anytime we can do this in a more objective manner, the better off we are.”
Can Digital Phenotyping Improve Swine Genetics?
The true power of the Pork Chop Studio lies in how the data is used, Hamilton says. Unlike a one-off research project, this technology is used on an ongoing basis to generate pork quality data that helps inform PIC’s genetic program.
“This is set up in the real world, in a packing plant where we cut chops every week and measure it,” Hamilton says. “We do routine data collection that feeds our genetic program to help us make decisions on which animals to mate for the next generation.”
“We dream it, and then engineers bring it to fruition,” Hamilton says about the digital phenotyping technology. “It’s repeatable and exciting.”
How Can You Predict Eating Experience?
While eliminating manual marbling scoring was the initial goal, the technology is rapidly expanding. The system is being trained to:
• Predict pH and Color: Using visual cues like surface moisture and reflectance to replace time-consuming manual meters.
• Measure Dimensions: Automatically calculating loin height, width and circumference.
• Predict Tenderness: Utilizing machine learning to estimate tenderness from images, a task that traditionally required destructive mechanical testing.
The ultimate goal? Predicting consumer preference before the meat is ever cooked. By pairing images with trained sensory taste panels, PIC hopes to identify the most flavorful pieces of pork with unprecedented accuracy.
“We intend to conduct trained taste panel evaluations where we have images of pork before it’s cooked along with human taste responses,” says Brandon Fields, PIC applied meat sciences global director. “With the data, we plan to train the system to identify the most flavorful piece of pork before it’s ever cooked and given to the tasting panel. That’s our ultimate pursuit.”
Eastwood says it all goes back to the concept that ‘you can’t manage what you don’t measure.’
“In everything that we do, we need good data to support decisions,” she says. “In the future, if we’re dreaming big, if we could take these measurements and pictures, and actually combine them with trained sensory evaluation, maybe we can get to the point where we’re actually predicting consumer preference.”
How Could Pork Chop Studio Drive Demand?
For pork producers, this technology represents a shift toward a more secure and profitable future.
“I want producers to know that not only are we producing a pig that has a lower cost of production, but we’re also thinking about the endpoint -- the consumer,” Hamilton says. “We’re trying to produce a product they desire, want to buy at the meat case, and can take home and have a good eating experience.”
There’s more to pork production than just the live animal, he adds.
“We all want to have more margin opportunity,” Hamilton says. “If we could have a product that has more consumer demand, they’ll be willing to pay for it ultimately, and we should be able to either sell more pounds of pork or get more premium.”
By adding objectivity to data collection, PIC wants to narrow the gap between the barn and the dinner plate, making sure that pork remains attractive to consumers worldwide.
