We are all getting older, says Dr. Harro Stokman, CEO and founder of Kepler Vision. Not just on an individual level, but the combination of falling birth rates and longer lifespans mean that the proportion of old people versus young people is higher than it has ever been. While this is a great achievement on the part of the medical and social care industries, this shift comes with its own set of problems.
A growing elderly population puts extra pressure on housing, long-term social care, and income security and healthcare systems. This is compounded by the fact that elderly care is intensive for nurses and other care staff, with many skilled workers ending up overworked. Case in point – in the UK, the care sector has the highest staff turnover rate of all sectors, with one in three workers leaving it every year.
Relieve some pressure
The Internet of Things (IoT) has huge potential to relieve some of this pressure, and to provide care services that can improve the quality of life for both patients and staff – while still balancing the all important priority of preserving the dignity and privacy of those it helps. We can already see plenty of examples of IoT assistance being used for elderly care in both residential homes and increasingly in care-home environments, through systems that employ vital signal sign monitors, smart wearables, audio–video sensors, smart devices, and smart TVs.
However, the sophistication of these systems varies wildly from case to case, as does their ability to ensure privacy. Take for example the systems most commonly used to monitor elderly patients in care homes or medical facilities at night – when they are at their most vulnerable. Before the widespread implementation of domotics, patients would be checked on by care staff in person, perhaps several times a night, which means a lot of work for staff and a disturbed night’s sleep for patients. Conversely, the infrequency of these checks could leave patients experiencing difficulties alone for hours. So what is the optimal solution?
Motion sensors or pressure matts help cut down on the amount of man-hours spent checking on people, but they are prone to producing false alarms. The ability to send alerts to staff in real time doesn’t mean that a motion sensor can tell the difference between someone rolling over in bed and someone falling out of bed. A connected camera system fixes this, but is a huge violation of a patient’s privacy – and having to constantly monitor multiple streams is another unnecessary burden on staff when that time could be spent on actually providing care.
Safety vs privacy
This safety vs privacy debate sits at the very heart of the IoT, and is quite difficult to solve to everyone’s satisfaction. But thankfully the advancement of technologies like computer vision and the ability to integrate it into existing systems, might offer a tidier solution when it comes to elderly care. The technology has made major advances in the last decade in automated and reliable recognition of image or video content, such as face, object and motion and – most specifically in this case – body language recognition.
One or more visual sensors can provide very detailed information about the state of the environment and its inhabitants when combined with the aforementioned pattern recognition and machine learning techniques. Coupling strict data storage with data deletion protocols, computer vision software can provide the accuracy of having a monitored camera set up in a patient’s room without violating privacy.
As we are now at the point where it can identify a whole range of different scenarios (such as falling out of bed, struggling to stand or long unexplained absences) from a live video feed, any camera monitoring systems only needs to activate when a patient needs assistance, without the need for any footage to actually be seen by human eyes, or stored after the fact.
From there it is a simple step to combine this technology with the established Internet of Things. Alerts to staff can be shared as simple texts that assistance is required, and some of the more boring aspects of care work – like the constant report writing – can be safely automated, all while reducing false alarms.
The only drawback, as with many IoT solutions, is in ensuring privacy. Complying to international certification programs which allow the vendor to store and process medical data securely are obviously essential. Importantly, this burden falls on the companies behind the software rather than on the care staff or the patients. In this way staff are freed up to deliver more of the care that is their core purpose, without them needing to worry about complications about data security and privacy.
Balance between protecting lives and dignity
As with so many advances in technology surrounding the IoT – the issues of privacy, security and trust are the most difficult features to navigate. Making sure that only the relevant data for care and medical assistance is shared with care staff is crucial to creating trust in this situation. In our increasingly privacy conscious world, computer vision technology might just offer medical staff the perfect balance between protecting patients’ lives and protecting their dignity as well.
The author is Dr. Harro Stokman, CEO and Founder of Kepler Vision.
Comment on this article below or via Twitter: @IoTNow_OR @jcIoTnow
