AI BREAKTHROUGH COULD HELP PREVENT 600 MILLION CASES OF FOOD POISONING

HSI captures an optical footprint of mycotoxins

Artificial intelligence may soon prevent hundreds of millions of food poisoning cases each year, with researchers developing technology that can identify contaminated crops before they reach consumers.

A team led by the University of South Australia has shown how hyperspectral imaging (HSI) combined with machine learning (ML) can rapidly detect mycotoxins - toxic compounds produced by fungi that contaminate crops during growth, harvest and storage.

These compounds are no minor threat. The World Health Organisation estimates foodborne contamination, including from mycotoxins, causes 600 million illnesses and 4.2 million deaths globally every year. The UN’s Food and Agriculture Organisation says roughly a quarter of the world’s crops are affected.

“In contrast, hyperspectral imaging – a technique that captures images with detailed spectral information – allows us to quickly detect and quantify contamination across entire food samples without destroying them,” says lead author and UniSA PhD candidate Ahasan Kabir.

Kabir and colleagues in Australia, Canada and India reviewed more than 80 studies covering wheat, corn, barley, oats, almonds, peanuts and pistachios — staples that are both widely produced and highly vulnerable to fungal contamination in warm, humid environments.

“HSI captures an optical footprint of mycotoxins, and when paired with machine learning algorithms, it rapidly classifies contaminated grains and nuts based on subtle spectral variations,” Kabir says.

The study found AI-driven HSI consistently outperformed conventional techniques, particularly in detecting aflatoxin B1, described by project lead Professor Sang-Heon Lee as “one of the most carcinogenic substances found in food.”

“It offers a scalable, non-invasive solution for industrial food safety, from sorting almonds to inspecting wheat and maize shipments,” says Prof Lee.

One of the most significant advantages is speed. The technology can work in real time, giving food processors the ability to screen entire batches without delays or waste. Researchers believe it could soon be deployed on factory lines or even in handheld devices, reducing both health risks and trade losses.

The team is now refining the system using deep learning and AI to improve accuracy and reliability. With further development, it could reshape how global food supplies are monitored, helping ensure only safe, uncontaminated produce reaches consumers.