The watch is a normal watch, but very unique. It is water resistant, shows the time precisely, has an alarm button and alarms automatically if you should fall.
Inside the watch there is a processor, a sensor and Bluetooth. BellPal has developed a software algorithm which is installed in the processor itself. There, the user’s movements are constantly analyzed to identify possible falling movements. The sensor provides information to the processor and the algorithm then analyzes the sensor data. In the event of an incident, the algorithm sends an alarm, which in turn is forwarded via Bluetooth to the Smartphone.
What is unique is that BellPal has tested the algorithm with real case studies of falls with seniors. BellPal documented this visually and with sensors on the seniors’ bodies. The documented cases are based on what healthcare shows are the most common fall patterns. When we tested the algorithm, the fall cases were done by a large group of individuals, where the majority were over 70 years old, in order to get as close as possible to the target group. The purpose of the algorithm is to analyze a fall, with as much precision as possible, in order to exclude the movements that are not real cases.
BellPal’s algorithm has excellent precision, which makes it one of the foremost fall algorithms. The method of developing the algorithm involved both manual and machine-learning. That meant that thorough and iterative testing was done to ensure the quality of the product.
Artificial Intelligence is a big concept, in summary it is about the algorithm being able to make decisions that are adequately based on the situation, and tested to analyze the differences identified.
At the beginning of the development work, it was decided that the battery life must be at least 6 months for operational effectiveness. The user should not need to think about charging the watch or be worried if the watch is on or off – it should work and do its job day in and day out. It is in itself an engineering masterpiece.