Fresh leafy greens like spinach and lettuce can carry harmful bacteria, but today's tests are often too slow, too costly, or too destructive to use in real time where produce is processed

The Department of Agriculture National Institute of Food and Agriculture awarded Trustees of Tufts College $633,750 on September 1, 2026, under the Agriculture and Food Research Initiative (AFRI) to develop CLEAR (Comprehensive Light-Based Evaluation and AI Recognition for Pathogens), a compact sensor for rapid, non-destructive detection of harmful bacteria on fresh leafy greens.

CLEAR uses inexpensive light-emitting diodes to shine selected wavelengths—including invisible light—onto leaves and reads reflected light with a black-and-white camera and explainable artificial intelligence models. The system replaces costly hyperspectral imaging with off-the-shelf components. The project team will build a library of light signatures characteristic of pathogen-contaminated produce, identify the most useful light bands, train AI models, and test a working prototype. Current laboratory testing for bacterial contamination takes a day or more, allowing contaminated product to reach stores before results return; faster screening would enable processors to catch contamination earlier, prevent outbreaks, and reduce food safety costs for growers, processors, and rural communities dependent on produce sales.

Place of performance is Medford, Massachusetts. Period of performance runs from September 1, 2026, through August 31, 2029. The award is structured as a Project Grant under AFRI, the nation's competitive grants program for agricultural sciences supporting research to improve food safety, agricultural profitability, and the bioeconomy.

Laboratory tests can take a day or more, so contaminated product may already be on its way to stores before results come back. The cost of getting this wrong falls widely: growers and processors lose crops, sales and consumer trust in a recall, and rural communities that depend on produce lose income. A faster, cheaper way to screen greens is needed to protect public health and to reduce the burden. This project develops clear (comprehensive light-based evaluation and AI recognition for pathogens), a compact sensor that shines selected colors of light on leaves, some of them beyond what the eye can see, using inexpensive light-emitting diodes (leds), and reads the reflected light with a black-and-white camera and artificial intelligence. Instead of a costly hyperspectral camera, IT uses low-cost, off-the-shelf parts. The team will build a library of light signatures, find the most useful bands, train AI models that can be explained, and test a working prototype.if IT succeeds, clear will give food processors a fast, low-cost, and non-destructive way to screen leafy greens, helping catch contamination earlier, prevent outbreaks, and lower the cost of keeping food safe.