One false alarm every three months thanks to AI

It seems trivial. But Harro Stokman, who worked on precisely this issue as a PhD candidate at the University of Amsterdam, discovered that triviality is an illusion. Together with a few colleagues, he founded Euvision, a company that gained attention with an app that could organize photos on your phone. Dogs, cats, sailboats. In 2014, the American company Qualcomm acquired the business.

After a few years at the tech giant, it was time for something different. AI had become significantly smarter. It was now possible to derive activities from moving images to understand what people are doing. However, Stokman was seeking an application that made a societal impact.

Three nurses for a hundred residents

The focus fell on elderly care. In Dutch nursing homes with a hundred residents, often only three nurses work at night. They respond when a sensor goes off. It frequently turns out to be a false alarm. A pillow falling off the bed gives the same signal as a client who falls. The protocol also requires entering each room a few times per night just to check that everything is okay.

Both scenarios are disruptive. For the resident because of the broken night. For the caregiver who is unnecessarily alerted. And for society, because this way of working demands a lot of manpower that is lacking.

The figures don't lie. The personnel shortage in Dutch healthcare will quadruple in the next ten years, from 66,400 in 2025 to 265,600 in 2034. [1] Specifically in nursing homes, there is already a shortage of 14,200 people. [2] A quarter of the current healthcare staff will retire within five years, while the number of elderly continues to grow.

"They did work in that sector with cameras and sensors, but it was all very outdated and they generated many false alarms."  Harro Stokman [3]

The sensor that understands what it sees

In 2018, Stokman founded Kepler Vision Technologies as a spin-off from the University of Amsterdam. The product is called Kepler Night Nurse. It is a smart sensor that hangs in the room and raises an alarm when something is about to go wrong.

The difference from traditional sensors is not in what the system sees, but in what it understands. The AI recognizes whether someone is lying down, about to fall, walking around, sitting on the ground, or leaving the room. It distinguishes between someone going to the toilet and someone who actually needs help.

And the false alarms? One per sensor every three months. [3]

That sounds like a technical detail. But for a nurse who is disturbed dozens of times during a night shift by alarms that mean nothing, it is the difference between exhaustion and workable nights. Alarm fatigue is a real problem in healthcare. After hundreds of false alarms, responses become slower or non-existent.

No camera, but reassurance

A sensor in your bedroom that is always on. That sounds like a privacy nightmare. But Stokman turned the argument around.

The system does not recognize people, only activities. The sensor does not transmit images, only a text message when something is happening. And most importantly: there is no longer a caregiver entering the room every few hours to check if everything is okay.

For many residents, that is actually an improvement in privacy. No physical intrusion during the night. No flashlight in the face. Only supervision when it is needed.

"The system does not recognize people, only activities. A timely alarm can prevent a lot of misery, but just as important is the peace it brings for both resident and caregiver."  Harro Stokman [3]

From university to nursing home

Kepler is now monitoring more than 14,500 clients and is active in 26 European countries. [4] Revenue was about one million euros in 2023. Stokman aims to triple this annually over the next three years. With an investment of €1.5 million from ROM InWest and previous funding from UvA Ventures and the Ministry of Economic Affairs, the company has now raised more than €9.8 million. [4][5]

Healthcare is often cautious with new technology, Stokman notes. But the fact that there is now competition is a good sign, according to him. It means the market is taking the problem seriously.

The shift from a purely technical company to a healthcare partner called for new faces who speak the market's language. With the arrival of Stephanie van Rosmalen as CMO and Lex Erades as COO, Kepler has professionalized its team for further growth. They must ensure that the technology not only works in the lab but is also embraced on the nursing home's work floor.

Where science meets the work floor

The story of Kepler Vision demonstrates what is possible when academic expertise encounters a concrete problem. Stokman began with the question of how a computer recognizes a chair. Twenty years later, that same knowledge helps a nurse at 3 AM to know when to intervene and when not to.

For the Dutch AI scene, this is a lesson. The most valuable applications are not always in the latest models but in the translation of existing knowledge to places where it truly makes a difference. A spin-off from the UvA helping nursing homes. An algorithm that once recognized chairs and now safeguards human lives.

The complexity is not in the code. It's in finding the right problem.

Willem Blom
Founder Dutchstartup