The company Kepler Vision Technologies therefore developed Kepler Night Nurse: a system that uses smart sensors to detect the movements of patients very precisely. How did the company go from an innovation to a successful product for the healthcare market? ROM InWest asked CEO Harro Stokman for text and explanation for other digital 'healthcare entrepreneurs'.
“A huge problem is coming our way in healthcare,” says Harro Stokman of Kepler. “A quarter of healthcare personnel will retire in the next five years and the number of elderly people will only continue to increase. Double ageing, in other words. A few months ago, the NRC already spoke of a ‘care infarction’. If we do not intervene, we will really get stuck – perhaps first in elderly care. The sector has more interest in efficiency than ever.”
Image recognition app
In 2010, Harro encountered a modern problem: an endless stream of photos on his iPhone in which he had difficulty finding specific moments. Together with his colleagues from the UvA – where he was working on his PhD in Computer Vision – he developed Impala, an app that can recognize objects and actions in photos. “By distinguishing dogs, cats, a sunset or an environment, we ‘taught’ the app to categorize images,” says Harro. “But developments in AI continued quickly. AI is also starting to recognize human actions. So whether a person is sitting, standing, smoking or eating. The technology fascinated me immensely and I soon realized that we should also do something social with it. The staff shortage in healthcare gave me the idea to use the software for camera images in nursing homes.”
Puzzle pieces
The switch to the healthcare sector brought challenges. Harro: “We still had to put some puzzle pieces in place. For example, in healthcare, cameras with a so-called fisheye-lens. They not only look straight ahead but also to the side, which allows you to capture much more. Because existing computer vision technology did not match this, we first had to adapt it to the technical core. A second problem was the reliability of the AI. When the software once raised the alarm during the holidays, the care staff did not find an injured patient but baby Jesus in the nativity scene. We have now reached the point where the technology analyses the lying position of patients and can draw up reports based on generated data.”
Scalable solution
Harro states that the technology is now rock solid. “Care workers receive notifications based on image analysis not only in the event of an accident, but also to prevent accidents – for example, if a client has been in the bathroom for too long or is sitting on the edge of his bed. Because the sensors recognise the uniform of care workers, the alarm only goes off for clients.” Thanks to the investment by ROM InWest, we can use extra marketing and sales to further market the innovation of Kepler Night Nurse. My goal is for the software to watch over a million clients by 2030. All with the aim of keeping care affordable and at a high level.”
Tips for entrepreneurs
What can other entrepreneurs in digital healthcare solutions learn from this? Harro advises to follow the lead of healthcare personnel. “They know best what they need. And impress upon your target group – healthcare professionals – that they will not be replaced by technology. Smart technological solutions are there to support personnel, so that they can do their work more efficiently and better.”
About Kepler Night Nurse
The already extreme shortage of staff in healthcare will increase by another 25% in the coming five years, while the demand for care is only increasing due to the ageing population. Better detection is crucial to relieve the burden on staff: with the existing sensors in nursing homes, someone often only arrives after five minutes if a client falls out of a chair or bed. Many reports also turn out to be false alarms, for example if someone accidentally presses the alarm button. smart sensor Kepler Night Nurse sends a notification within 10 seconds of a fall and prevents accidents by precisely detecting patient movements – for example, getting up from a chair or sitting on the edge of the bed. This means that staff arrives on site sooner and the number of false alarms with 99% is reduced.
