Kepler Vision Technologies announced the launch of its Kepler Night Nurse Edge Box. The processor has been designed to integrate Kepler Vision’s Night Nurse software with innovative computer vision technology that will increase patient safety at night while providing healthcare professionals with the tools to provide better quality care.
Using deep learning and computer vision, Kepler Night Nurse can detect falls and physical distress, before automatically alerting care teams when patients need help. Highly effective, eliminating 99% of false alarms, the solution has been developed to replace traditional sensor devices such as bed mats, motion sensors and other sensing devices including necklaces and wristbands.
How does it work?
The Edge Box Night Nurse initially requires a three-month calibration phase during which it processes patient video feeds in-house, ensuring greater privacy for both patients and staff. The Edge Box then analyses the live video feeds through AI, thereby removing the need for human intervention. The only output from the Edge Box is text alerts to staff when patients need help. The solution focuses on both privacy and eliminating the burden of constant human video monitoring, while allowing staff to be able to provide instant help when something goes wrong.
As Dr Harro Stokman, CEO of Kepler Vision Technologies, explains, the new features and flexibility that the Kepler Night Nurse Edge Box brings to patient wellbeing at night is revolutionising the industry, as it combines proven efficiency with reduced installation costs and complexity, while eliminating the need for a constant broadband internet connection to handle multiple streams at the same time. “Given the current state of the care home sector, we are proud to offer a product that reduces the pressure on staff to deliver care more efficiently, without sacrificing patient privacy,” he concluded.
With this solution, the world leader in vision-based human activity recognition software offers the care home industry a unique computer vision video processing solution that allows windowing of video streams without ‘straightening’ the images, which is very difficult for human operators to do. The Edge Box also allows users to run video processing locally rather than in the cloud, eliminating the need to compress video streams. This improves the accuracy and reliability of the computer vision software, as any video processed by the Edge Box can be more closely inspected.
The deep learning-based software also allows for customisation of patient needs. Going beyond generic alerts from the care team, Kepler’s Night Nurse allows caregivers to record different causes of patient concern and send corresponding alerts. If a patient has trouble getting up or seems to be spending an unusual amount of time in the bathroom, a personalized alert is sent to the caregiver using the self-service feature of the software.
Beyond night care, Kepler’s night nurse can also automatically add specific behavioural insights to a patient’s medical record, which can help doctors in the long-term monitoring of patients – by identifying behavioural patterns that may indicate a change in well-being. It therefore provides a complete solution for the care of the elderly.
As the first computer vision-based fall detector to receive medical device status, it has been officially registered as a medical device in accordance with the European Council Directive 93/42/EEC. This registration means that the software has been tested both in-house and in the field to ensure that it meets the highest specifications and that risk assessments and safeguard measures have been met. The software is the subject of eighteen patent applications, of which three have been granted and fifteen more are pending.