top of page

Stress-free mice weighing with computer vision

Cartoon mouse

Research from the US has developed a new non-invasive method to effectively weigh mice using computer vision, enhancing animal welfare in research settings.


Weighing mice is essential in many studies, but traditional methods with scales can cause handling stress to the animals, leading to inaccurate measurements.


Researchers at EARA member The Jackson Laboratory (JAX) have announced an improved weighing system using computer vision, a field of artificial intelligence that is capable of extracting data from images and videos.


This new technology captures images of individual mice in their cages, with continuous monitoring, and automatically calculates their weight based on the images.


The system was shown to be as accurate as scale measurement, with only 5% error across sexes and strains, but without the stress caused by handling – an example of refinement in the 3Rs of animal research.


Malachy Guzman, from JAX and first author of the study published in Patterns, stated: “By training our models with genetically diverse mouse strains, we ensured that they could handle the variable visual and size distributions commonly seen in laboratory settings. Ultimately our statistical models to predict mass can be used to carry out genetic and pharmacological experiments.”


A related study from JAX – published in Nature Biotechnology – also emphasised the use of genetically diverse mice in drug discovery, which can better mimic the variability found in humans and lead to more reliable and effective drug development.

bottom of page