Postdoctoral researcher Paulina Feire Vasconez was the guest of The AI Conversation hosted by Bao Johri, vice president for information technology and chief information officer at Fresno State, on Nov. 7.
Vasconez, a lecturer in food science and nutrition at Fresno State, is currently focusing her research on dairy products. She is exploring how machine learning, sensory analysis and rapid detection methods can be used to optimize the quality, safety, sustainability and innovation of dairy products.
Vasconez’s work with dairy combines expertise in machine learning, spectroscopy and bioprocess optimization.
Spectroscopy is the study of how matter interacts with electromagnetic radiation, like light and is used to identify substances and their compositions. Bioprocess optimization is the process of fine-tuning biological manufacturing processes to increase efficiency, yield, quality and sustainability while reducing costs.
“Dairy is one of my favorite products,” Vasconez said. “It’s an amazing matrix that has different compounds such as water, proteins and fats with everything in equilibrium.”
Vasconez said that while she was working on her doctoral degree, she noticed that spectroscopy and related technologies used a lot of data, which was tedious and time-consuming.
“In my mind, I always thought there should be a faster and better way,” Vasconez said.
She said that when she started learning about artificial intelligence (AI) and machine learning algorithms, she found a way to improve the efficiency of dairy product data. Using AI, she could predict functionalities for milk and whey.
“I could identify early defects and reduce testing time from hours to two seconds,” Vasconez said.
Vasconez said a spectroscopy captures a product’s fingerprint. For example, it can identify two different mozzarella products made by different manufacturers that have the same moisture, protein and fat content, and determine which creamery it was made in.
It’s a very powerful technology, and the results are quickly available without needing the preparation of a sample,” Vasconez said.
She said that the combination of spectroscopy and machine learning algorithms can detect patterns that are difficult for humans to see at first. The technology can provide real-time quality control in production lines without chemical testing, facilitating a safer environment and a more consistent, sustainable product.
“I hope the audience is learning as much as I’m learning,” Johri said. “Especially, here in the Central Valley, where agriculture is so prevalent.”
Johri said that the Central Valley produces 25% of the nation’s food supply, and that sustainability and food production will be key.
Vasconez said that AI technology in agriculture is an interdisciplinary field, and she collaborates with the engineering department to develop new technologies with students and faculty. Their support is invaluable in advancing AI in food production.
She said students need to be willing to learn a little bit about AI technology in agriculture. She added that AI is here to make our lives easier, it’s easy to use and is something everyone needs to discover.
“The future of this technology makes me really happy,” Vasconez said.
