Pain is often hard to spot, especially in animals that can’t tell us what they feel. That challenge has led scientists to build tools like this one.

At first glance, this setup looks like a simple plastic container. White walls, a bright orange floor, nothing fancy. You might think it belongs in a playroom.


EarthSnap

But inside this box, something serious is happening. Scientists are trying to understand how animals feel pain and how to reduce it.

The box is designed for mice used in research. It creates a dark, quiet space where the animals can behave naturally.

There’s no human hovering nearby, no sudden movements – just a controlled environment where every small detail can be observed without adding stress.

Why darkness matters

Inside the box, it’s completely dark – that’s not by accident. Mice feel safer in the dark, and when they feel safe, they act like themselves. They explore, sniff around, and eventually settle down. Some even fall asleep.

Two cameras record everything. One looks from above, the other from the front. An infrared light lets the cameras see without disturbing the mouse.

The setup captures small changes in posture and movement, but the real focus is the face.

Pain shows up in subtle ways. A mouse might narrow its eyes, shift its ears, or change how its whiskers sit. These signs are easy to miss with the naked eye, especially when the animal is moving.

Reading pain in a face

Scientists have long used something called the Mouse Grimace Scale. It scores facial features to estimate pain levels. Each sign gets a number from 0 to 2, depending on how clear it is. The method works, but it has limits.

It takes time, depends on who is doing the scoring, and sometimes the mouse’s face isn’t visible enough to judge properly.

On top of that, the presence of a human observer can stress the animal, which may affect the results.

The new system changes that. It’s called GrimACE and uses an algorithm to study the mouse’s face in real time.

As soon as the animal enters the box, recording begins. The system selects the clearest frames and analyzes them for signs of discomfort.

Improving how animals are used

The work comes from researchers at ETH Zurich, led by Oliver Sturman, Head of the 3R Hub.

The Hub focuses on improving how animals are used in science, following the principles of Replace, Reduce, and Refine.

“This box allows laboratory animals to be observed in a humane and standardized way, whether by us here in Zurich or by researchers on the other side of the world,” said Sturman.

“This is important so that the animals feel comfortable and unobserved.”

A more consistent way to measure

One of the biggest challenges in animal research is consistency. Two experts can look at the same mouse and come up with different scores. That’s not because they’re careless – it’s because human judgment varies.

In one test, three trained raters looked at the same images. Their scores didn’t match. Each person stayed consistent with their own ratings, but their overall scoring levels differed.

“We secretly gave all three raters the same images to assess, to check whether their own scores were consistent,” said Sturman. And they were. Still, the differences between them were clear.

“This is where we see the strength of the computer, as it delivers standardized results.”

That consistency matters. If pain is underestimated, animals may not get the care they need, and if it’s overestimated, experiments could stop too soon.

“If someone always assesses that an animal is not in pain, animals will suffer needlessly. And if someone always gives overly high scores, there is a risk that experiments are abandoned unnecessarily,” said Sturman.

Tested in real conditions

The team put GrimACE to the test in a study involving mice that had undergone brain surgery.

The animals received painkillers based on expert guidelines. Images were taken before and after the procedure.

An expert reviewed thousands of these images manually. At the same time, the algorithm analyzed the same data – the results were nearly identical.

That close match shows the system can keep up with trained human observers. It also works faster and without fatigue.

More than just faces

The system doesn’t stop at facial expressions. A high-resolution camera tracks body movement from above.

It measures distances between points on the body, changes in angles, and even how quickly those points move.

These details can reveal signs of discomfort that people might miss. Machine learning models pick up patterns that are too subtle for the human eyes.

Built for global use

The entire setup is designed to be shared. The hardware and software are available as an open-source kit. That means labs around the world can build the same system and collect comparable data.

“The idea is that as many users as possible can assemble and use it in a straightforward and standardized way – and that the data will then be comparable,” said Sturman.

The more data the system gets, the better it becomes. “The more people who use GrimACE, the less bias there will be.”

Researchers from countries like the United States and the United Kingdom have already shown interest. The system is also now available at the ETH Phenomics Center for wider use.

What comes next

The team plans to keep improving the technology. They are exploring whether to turn it into a commercial product or continue sharing it freely through collaborations. For now, their focus is clear.

“We’re currently sharing our knowledge and technology in collaborations and are focusing on mutual data exchange to improve the system,” said Sturman. “Our primary concern is to improve animal welfare.”

That small plastic box may not look like much. But inside it, a quiet shift is happening in how science treats the animals it depends on.

The full study was published in the journal Lab Animal.

—–

Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates. 

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.

—–