08-09-2021 | By Robin Mitchell
Dorset will be trialling a range of sensor systems to see how AI can be used in social care by medically vulnerable people. What challenges does care in hospitals present, what will the trial do, and could it be the future in post-op care?
While many are aware of the costs of medical treatment, the cost of care often goes unnoticed. Caring costs for an individual who has had major surgery or is incapable of performing independent activities can be extremely high. It is not uncommon to see the property sold to help pay for the cost of care homes for the elderly in the last stages of their life.
In the case of hospitals, patients need to be carefully monitored after major surgery to ensure that there are no complications, and doctors cannot release patients until they are confident that no more harm can come to them. From personal experience, I was hospitalised for a severe case of COVID-19 but was not allowed to leave the hospital until I had fully recovered (this included multiple x-rays, frequent consumption of medicines, and blood tests), despite being able to move around, breathe, and function normally.
Regardless of whether a medical industry is public (NHS) or private (US), the cost of such care is always the taxpayer's burden. Sending patients home early is one option for reducing this cost, but doing so could jeopardise patient safety as they may not notice danger signs, including increased urination, tiredness, involuntary movements, and confusion. In fact, the inability to spot such signs could theoretically increase the cost of care by requiring frequent outpatient visits by doctors and nurses and the cost of treatment for secondary complications.
Recently, a company in the UK called Lilli will be trialling 100 outpatients and those who require some degree of social care with home sensors and AI algorithms to help reduce the burden on local hospitals.
The idea is to integrate sensors into everyday items such as kettles and fridges, commonly used at regular intervals. These devices would not only monitor their own usage but some bio signs, too, including temperature and skin conductivity. This data is then fed into an AI to determine if the patient deviates from their usual routine or typical bio signs. If deviations are detected, a doctor or nurse can be sent out to meet with the patient for a checkup to ensure that they are in good health.
The system being developed by Lilli will not only improve patient independence but would also reduce the cost burden on the NHS, and by extension, the taxpayer. According to Lilli, their trial could save local hospitals up to £250,000 by reducing care hours by 780 hours. Such a system would also enable hospitals to focus care on those who need it most.
The use of AI to track patients' health will undoubtedly become popular as it will provide cheaper options to patients and present the opportunity to be independent. In fact, doctors will sometimes encourage movement and return to everyday life as this can help with recovery (of course, this depends on the specific procedure). However, there is an element to patient monitoring that needs to be addressed carefully; privacy. The system developed by Lilli does not integrate cameras for this particular reason but can still monitor data such as toilet usage.
Future systems may integrate camera and microphone technology to provide improved services, but such devices would be ideal for attack by cybercriminals. As such, any and all patient monitoring systems need to ensure patient privacy and confidentiality on any data gathered while remaining impenetrable to any and all attacks.