Scientists are perfecting plant-based meats with AI

Meat lovers can attest that the move to plant-based alternatives isn’t always a simple transition. However, thanks to a team of innovative Stanford engineers, life is becoming a little tastier for those wishing to embrace greener diets.

Working from a unique mechanical engineering perspective, they’ve developed an intriguing approach to food texture evaluation.

This could potentially lead to plant-based steaks that even the staunchest meat enthusiasts will find impossible to resist.

Innovation in plant-based meat texture

The team from Stanford University recently showcased how mechanical testing, combined with machine learning, can essentially mimic the palate of human taste testers. Their ingenious method could potentially accelerate the evolution of superior plant-based meats.

Interestingly, they discovered that several plant-based products are already successfully emulating the texture of the meats they strive to replicate. According to Ellen Kuhl, the senior author of the recent study, it is possible to close the gap between animal meat texture and that of plant-based meats.

“We were surprised to find that today’s plant-based products can reproduce the whole texture spectrum of animal meats,” noted Kuhl.

This is phenomenal progress since the days when tofu was the only meat substitute available.

The environmental impact of our food choices

Industrial animal agriculture’s environmental toll is heavy; it leads to climate change, pollution, habitat loss, and antibiotic resistance. An effective way to lighten this is by shifting from animal proteins to plant proteins.

Previous studies have shown that plant-based meats cause, on average, 50% less environmental impact than animal meats. However, most meat eaters are hesitant to make this change.

For instance, in one survey, only a third of Americans stated their likelihood of purchasing plant-based alternatives as “very likely’ or ‘extremely likely.”

“People love meat. If we want to convince the hardcore meat eaters that alternatives are worth trying, the closer we can mimic animal meat with plant-based products, the more likely people might be open to trying something new,” explained Skyler St. Pierre, the lead author of the paper.

Unfortunately, conventional food testing methods are neither standardized nor transparent, making it tougher for scientists to collaborate on creating new recipes for alternatives.

The future of food texture testing

This novel and futuristic food texture test that makes use of AI came to life as part of a class project by St. Pierre. He was looking for affordable materials to use in engineering tests on stress, load and stretching. He turned to tofu and hot dogs as possible study materials.

Through the summer of 2023, student researchers experimented with the texture and mechanics of these and other foods, and then trialed their findings on eight products including animal and plant-based hot dogs, sausages, turkey, and firm and extra firm tofu.

A machine was used to simulate the act of chewing, and to test how the materials reacted when pulled, pushed, and sheared.

This data was then processed using machine learning to design a new type of neural network, and resulted in equations defining the properties of the different meats.

A survey was later conducted to test if these equations could truly replicate the sensation of texture.

The survey respondents, after sampling the eight products, rated them in 12 categories including softness, hardness, brittleness, chewiness, gumminess, viscosity, springiness, stickiness, fibrousness, fattiness, moistness, and meat-likeness.

Plant-based findings and hope for the future

The results were thrilling. For instance, plant-based hot dogs and sausages were very similar to their animal counterparts in terms of their stiffness. Human testers ranked the stiffness of the hot dogs and sausages very similarly to the mechanical tests.

“What’s really cool is that the ranking of the people was almost identical to the ranking of the machine. That’s great because now we can use the machine to have a quantitative, very reproducible test,” said Kuhl.

Such discoveries suggest that data-driven methodologies may accelerate the development of delectable plant-based products.

The team even suggests using generative AI for creating plant-based meat recipes that have specific desired properties.

Open-source collaboration and future prospects

In an effort to push this field forward, the team has opted for open-source data, allowing other researchers to view, add to, and learn from their work.

“Historically, some researchers, and especially companies, don’t share their data and that’s a really big barrier to innovation,” Kuhl added.

The team continues in their food testing endeavors as they look to establish a public database of their findings.

They’re now testing deli slices, both vegetable and meat-based. They also plan to test engineered fungi that has been developed by a new addition to their team.

Kuhl has generously extended an invitation to the wider community to contribute to the ongoing study.

“If anybody has a plant-based meat they want to test, we’re so happy to test it to see how it stacks up,” said Kuhl.

In their quest to save the planet one bite at a time, these Stanford engineers are transforming our perception of plant-based diets.

Their commitment to open-source information and collaboration is setting the stage for a more sustainable future.

The study is published in the journal Science of Food.

—–

Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates. 

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.

—–