AI Allows Predicting Meat Tenderness and Fat Content

Published Jul 11, 2025

Tridge summary

Researchers from Brazil, Canada, and the United States developed an artificial intelligence system capable of predicting the tenderness and intramuscular fat content of raw meat from photos taken with a cell phone. The study, published in Meat Science magazine, used computer vision techniques and neural networks to analyze images of beef sirloin and pork loin, captured in a controlled environment over the course of a year.

Original content

The model was trained with images associated with laboratory data, such as the force required to cut the meat and the percentage of fat between muscle fibers. The technology was able to correctly identify the most tender cuts in up to 81.5% of cases with pork and 76.5% with beef. For intramuscular fat, the accuracy reached 77% for bovines and 79% for pigs, surpassing the performance of consumers who participated in comparative tests. The practical application of the system is still limited to two specific cuts, but researchers intend to expand tests to different types of meat, lighting conditions, and origins. The expectation is that the technology could give rise to an app to help consumers choose better quality cuts directly at the point of sale. In addition to benefiting the end consumer, the tool has the potential to ...

Source: Agrolink