Artificial intelligence distinguishes whether a lamb has grazed, but not how long it did so.

French researchers have demonstrated that artificial intelligence can identify whether a lamb has been grass-fed or raised in a pen, using only data obtained from the animal's fat and muscle. "The algorithms managed to distinguish between channels of grass-fed lambs and lambs raised in pens with up to 95% accuracy in correctly classifying lambs," note the authors of the study. The team conducted the experiment with several groups of lambs. Some remained in pens, while others were moved to alfalfa pastures for 21, 42, or 63 days before slaughter. Subsequently, nineteen variables obtained from tissues such as perirenal fat, dorsal fat, and the longissimus dorsi muscle were analyzed. The methods employed—decision trees and random forests—allowed the identification of the most relevant biomarkers. "Of the 19 variables, two stood out clearly: the skatole content and carotenoid pigment of the perirenal fat," explain the researchers. These indicators proved to be the most reliable keys ...