Chicken ‘woody breast’ detection improved with advanced machine learning model

By John Lovett
University of Arkansas System Division of Agriculture
Arkansas Agricultural Experiment Station

FAYETTEVILLE, Ark. — It’s called “woody breast” and for consumers it can mean a chewier chicken sandwich, but for the industry it can mean up to $200 million annual yield loss.

IMPROVED METHOD — Chaitanya Pallerla, a food science graduate student, has worked to improve the accuracy of detecting the "woody breast" defect on chicken by developing a new machine learning model and hyperspectral imaging. (U of A System Division of Agriculture photo)

Work done by the Arkansas Agricultural Experiment Station is not only making woody breast easier to detect in chicken meat but is accurate up to 95 percent of the time.

The development could help improve quality assurance and customer confidence in one of the state’s most economically important agricultural products. What allows researchers to see inside the meat is a combination of a hyperspectral camera, which examines the meat through various energy wavelengths, and machine learning to interpret what the camera sees.

“We’ve been able to improve accuracy of detection of woody breast by utilizing machine learning to analyze complex data from images with a hyperspectral camera,” said Dongyi Wang, an assistant professor in the biological and agricultural engineering department for the experiment station, the research arm of the University of Arkansas System Division of Agriculture.

“The next step will be trying to integrate the system online and make this beneficial for stakeholders,” Wang said, noting this specific application of image analysis had not been done before.

Loss in premium meat

“Woody breast” meat is harder and chewier than normal chicken breast, but it is still safe to eat, according to Casey Owens, professor of poultry processing and products for the experiment station and a co-author of the study. When detected by processers, either by humans or computer-assisted imaging technology, she said the meat is diverted from whole-breast packaging for further processing into products including chicken nuggets and patties.

The loss in premium as a whole-muscle product accounts for yield loss as high as $200 million in Arkansas and over $1 billion in direct and indirect costs annually across the United States poultry industry, Owens added. Up to 20 percent of chicken breast meat can have the defect, which is more common in larger birds of 8 to 9 pounds versus 6- to 7-pound birds.

Hyperspectral imaging

Hyperspectral imaging is a rapid, non-invasive way to capture detailed data about objects and their composition. This data can be used to classify food products according to food quality, consumer preferences and other product requirements.

But hyperspectral images come with tons of data. That’s where machine learning comes in.

Chaitanya Pallerla, a food science graduate student who has been working on the project for the past two years with Wang as his adviser, said the new machine learning model is called NAS-WD. When correlated with known data about the “woodiness” of chicken breasts, the model allows for deeper and wider analysis of hyperspectral images to identify the defect.

“In hyperspectral imaging, there are common machine learning models being used, but we were able to develop a new model that could be well-suited for correlating more than two variables,” Pallerla said. “We kind of took two different models, made a few changes, and put them together to detect patterns better and correlate the hyperspectral data with hardness of the chicken meat.”

The results of their research were published in the journal Artificial Intelligence in Agriculture under the title “Neural network architecture search enabled wide-deep learning (NAS-WD) for spatially heterogenous property awared chicken woody breast classification and hardness regression.”

The results showed that NAS-WD can classify three woody breast defect levels with an overall accuracy of 95 percent, outperforming the traditional models like the Support Vector Machine and Multi-Layer Perception, which offered 80 percent and about 73 percent accuracy, respectively.

HYPERSPECTRAL IMAGING — Dongyi Wang is an assistant professor in the biological and agricultural engineering department. (U of A System Division of Agriculture)

Wang said the study offers an example of how to use new algorithms to mine data and dig into key information. The form of hyperspectral imaging used in the research is called “push broom,” which takes an image of several objects once every 40 seconds, compared to a more common industry method of a “snapshot,” which takes an image of individual objects as fast as every 30 milliseconds. The “snapshots” have a lower resolution than the “push broom” method, but software upgrades may one day provide higher resolution for “snapshot” images, Pallerla said.

Wang said his team is working on deploying this technology in the real-time system.

The study was supported in part by the Agriculture and Food Research Initiative, project award nos. 2023-70442-39232 and 2024-67022-42882, from the U.S. Department of Agriculture’s National Institute for Food and Agriculture.

To learn more about the Division of Agriculture research, visit the Arkansas Agricultural Experiment Station website. Follow us on X at @ArkAgResearch, subscribe to the Food, Farms and Forests podcast and sign up for our monthly newsletter, the Arkansas Agricultural Research Report. To learn more about the Division of Agriculture, visit uada.edu. Follow us on X at @AgInArk. To learn about extension programs in Arkansas, contact your local Cooperative Extension Service agent or visit uaex.uada.edu.

Arkansas 4-H members put chops to the test at state barbecue contest

By Rebekah Hall
U of A System Division of Agriculture

ROGERS, Ark. — Fifteen Arkansas 4-H members from 13 counties brought the heat at the 4-H State Barbecue Cook-Off, with first-place winners in chicken and turkey barbecue advancing to the national competition in Louisville, Kentucky, this fall.

BRING THE HEAT — Arkansas 4-H members tested their grilling skills at the Arkansas 4-H State Barbecue Cook-Off, held June 21-22 in Rogers at the 64th Annual Poultry Festival. Sarah Lamb, 18, was the first-place chicken barbecue winner, and Laney Slate, 18, was the first-place turkey barbecue winner. (Lamb pictured front row, right middle; Slate pictured front row, left middle. Division of Agriculture photo.) 

The contest was held June 21-22 in Rogers at the 64th Annual Poultry Festival, hosted by The Poultry Federation. Sarah Lamb, 18, was the first-place chicken barbecue winner, and Laney Slate, 18, was the first-place turkey barbecue winner. Lamb and Slate will advance to compete at the national contest, which will be held at the National 4-H Poultry and Egg Conference in Louisville, Kentucky, from Nov. 20-21.

Andrew Bolton, extension instructor of poultry science for the University of Arkansas System Division of Agriculture, said the contest teaches participants important life skills.

“Food prep and food safety being first and foremost, from good sanitation practices to being conscious of the correct temperature that different meats need to reach to be safely cooked,” Bolton said. “This contest also pushes these youth to do research into the poultry industry, as well as some healthy living aspects as to why proteins like chicken and turkey are important.

Bolton said participants have also gained confidence in “their own abilities to get a grill lit, food prepped and then cooked all on their own.”

“They’ve learned how to be great critical thinkers in how they have developed their own sauces, marinades and rubs,” Bolton said. “I have also seen their creativity flourish when designing posters as part of their presentations on their choice of meat.”

Lamb has been a member of Arkansas 4-H in Howard County since she was five years old, competing in her county and district barbecue contests for the past four years. This year, she qualified for the state contest for the first time. Lamb attributes this success to changing up her seasoning blend, which remains a secret.

“I practice two times a month, and when I practice at home, I use a lot of different seasonings to figure out which ones I like,” Lamb said.

Slate has been a member of the Dayton 4-H Club in Sebastian County for nine years and began competing in the 4-H barbecue contest at the county level when she was nine years old. In 2021, she won first place in chicken barbecue at the state contest and advanced to the national level, where she placed fourth. Because she made it to the national competition with chicken barbecue, she wasn’t allowed to compete in the same category again, so she switched to turkey burgers for the 2024 competition.

“This year I had to practice a lot more, because the cooking process for turkey burgers versus a whole chicken is very different,” Slate said. “It’s almost two hours to cook a chicken, versus 30-45 minutes for the burgers. I had to learn that it’s okay for it to not take forever.

“With the burgers, you have to put in the ingredients to flavor it and patty it yourself, so finding the right consistency was difficult at first,” Slate said. “But once I got that down, I felt like it was fine.”

Practice and preparation

Lamb’s father taught her how to cook on a charcoal grill when she was 10 years old. When she learned the state contest would feature charcoal grills, she was relieved.

“It does take a lot of patience,” she said. “I’m the only one in my family who can do it.”

For her technique, Lamb said it’s important not to fuss over the chicken.

“I’ve learned to shut the lid and not look at my chicken for about 20 minutes, then look and see where the temperature falls,” she said. “To make it juicier, you barely touch it.”

Slate said that for her turkey burgers, breadcrumbs were key.

“You have to be careful because if you add too many, it dries them out and they fall apart when you cook them,” Slate said. “Finding the right ratio of breadcrumbs to the meat was the tricky part.”

Slate said she hopes to place within the top five at the national competition. She plans to work on her speech presentation, a significant percentage of contestants’ scores, in the meantime.

Slate said her favorite part of participating in the barbecue contest has been making new friends.

“I’ve met a lot of people who the only thing we have in common is cooking in that contest,” Slate said. “Some people I met in 2021, back at the national event, I still talk to them to this day. So, I’ve just enjoyed the friendships and people you meet along the way.”

Lamb said her 4-H record book — where 4-H members log all projects, contests, activities, events and skill development — has focused on food and nutrition, which she plans to pursue as a career.

“I’m going to culinary school a year from now because I would love to have my own bakery,” Lamb said. “It’ll be a bakery and a café, where I could serve food and sweets.”

For her fellow 4-H members interested in testing and improving their own barbecue skills, Lamb said she advises not to fear the grill.

“Don’t be intimidated by the grill, because it’s just a grill,” she said. “And if something happens, you’ll always have somebody to help you. When you get to know it, it’s more fun, because you can do anything on a grill.”

Slate said it’s also important for competitors to be able to adjust to circumstances out of their control.

“Being able to think quick on my feet has been important,” Slate said. “Since we cook outside, I’ve had to cook in the rain, the sleet and the snow. At the national level, it started sleeting and snowing on us, and then after we got done cooking, the sun came out. So, you never really know, and you just have to be prepared.”

The Arkansas 4-H youth development program is operated by the Cooperative Extension Service, the outreach arm of the Division of Agriculture. To learn more about Arkansas 4-H, visit 4h.uada.edu.

To learn about extension programs in Arkansas, contact your local Cooperative Extension Service agent or visit www.uaex.uada.edu. Follow us on X and Instagram at @AR_Extension. To learn more about Division of Agriculture research, visit the Arkansas Agricultural Experiment Station website: https://aaes.uada.edu. Follow on X at @ArkAgResearch. To learn more about the Division of Agriculture, visit https://uada.edu/. Follow us on X at @AgInArk.