Using our powerful video analytics, sleeping patterns can be automatically and non-invasively monitored and translated into relevant statistics on nightly movement and activities. Getting in and out of the bed, sitting up, rolling and turning can all be accurately recognized, providing information on quality and duration of sleep. Furthermore, sleeping anomalies, like falling out of the bed and seizures can be detected in real-time and can provide alerts to caretakers, allowing for immediate response. The technology can be used in combination with a regular camera, or with a depth sensor that fully protects privacy. Both the statistics of nightly behaviour, as well as the immediate detection of incidents are especially valuable to the healthcare sector, e.g. in the following applications, which can be both inpatient and outpatient: Monitoring the behavior of post-operative patients as their sedation wears off, Providing sleep studies in people with sleeping issues, sleepwalkers, etc. Immediate detection of falling out of bed (e.g. for elderly, physically/mentally disabled), Immediate detection of people with seizure conditions in case of a nightly seizure, Immediate detection of break-downs and/or self-harm in people with mental conditions, Etc.
The technology consists of a software algorithm that can work with both IP(Internet Protocol)-based cameras and low-cost 3D sensors like the Microsoft Kinect. It has been tested based on regular video footage which led to detection accuracies of 80%-100% on behaviours like rolling/turning, getting in/out of the bed, moving to/from the bed, sitting down, standing up and laying in bed. In separate trials, accurate detections of self-hurt have been provided based on depth sensors to protect the identity of the subjects. The algorithm requires only a low to moderate amount of processing power, and can be run on regular consumer hardware.
This technology is especially valuable in the healthcare sector, including mental healthcare & daycare centres. It also helps in our aging society, allowing the elderly to live at home longer in a safe way, as well as improving safetyin retirement facilities and alleviating the work load of the staff.
The technology provides benefits to the end-user as it allows for an immediate response to dangerous situations (e.g. a night nurse being alerted immediately in case of a seizure). Furthermore, the end-user can track his/her sleeping patterns in a quantitative way, thereby using it as a (self-)diagnostic tool to study e.g. positive and negative contributors to sleep quality. It provides benefits to caregivers by increasing productivity. There will be less need for routine surveillance rounds in wards. Instead, caregivers are alerted to those situations that require their attention, without delay.