MySense - a data lifeline for the care sector

Last week I visited Imperial College’s Living Lab and spent a fascinating day contributing data about my every move. It is one of a number of projects where academics are trying to develop technology to monitor people living with neurological conditions. What I did not realise until I got an email about my piece was that this is an idea already being deployed in the real world.

The email was from MySense, which describes itself as a “wellbeing analytics platform powered by artificial intelligence.” That did not mean much to me but it all became a lot clearer when I got on a call with the company’s Chief Executive Lucie Glenday. She explained that MySense was, in effect, installing “living labs” in thousands of homes on behalf of local authority social care services.

MySense sensors

The aim is to monitor people with dementia or other neurological conditions, cut down on hospital visits and allow them to live at home longer. The company installs eight sensors in each home - one is a wearable heart monitor, another a strap which goes across the mattress to measure sleep, but the rest are used to detect movement and activities at the front door, and in the kitchen and bathroom.

They give the MySense number crunchers an extraordinary 15,000 data points on each person every day. I wondered what use it is knowing that someone has cleaned their teeth or opened the fridge door. Lucie explained:

“We're not particularly interested in someone opening the fridge door, for example, or someone opening the front door. We want to know how long it took to get to that door. We want to know the mobility of someone getting to that door. We want to know how difficult they find it getting from sitting to standing. We want to know the impact of sleep beforehand.”

What they then do with this ocean of data is add more to contextualise it - demographic and environmental information, health data: “We build essentially this pattern of normality for this individual, understanding the contextual levers and dials for somebody's ability to remain independent and survive in their own environment.”

By knowing what “normal” looks like, MySense can spot variations in the data from the sensors that might be warning signals. Starting off in Surrey, the technology has now been deployed by two NHS Trusts in the Midlands and will soon be installed in 2,000 homes in Greater Manchester.

It is very early days but the company and its customers seem to think it is having a positive impact. South Warwickshire NHS Foundation Trust reported a 46% reduction in unplanned hospital admissions for patients using MySense. University Hospital of North Midlands NHS Trust (UHNM) reported a similar fall in hospital visits and an average cost reduction per patient of £16,800.

It seems the monitoring service proved very popular with the patients and their families: “I feel safer in my own home with the sensors as I know the ladies at the office are always monitoring me,’ one is quoted as saying .

A word of caution - it is early days and it seems to me that part of the reduction in hospital visits may have been the result of a greater focus on the needs of these patients during the trial. Nevertheless it seems obvious that this is an area where data analytics has great potential to provide better care at a lower cost.

Lucie Glenday - a self-confessed maths geek - trained as a lawyer but her career has included working on the Government Digital Services team which created gov.uk and becoming chief digital data officer at Surrey County Council, one of the first local authorities to have such a role.

But it was a personal tragedy which set her on the path to building this company. When she was in her 20s and working in Japan, her sister was diagnosed with Motor Neurone Disease, and died just nine months later. Ever since, she’s wondered whether she and her family could have done more: “The one thing that really frustrated me is that we weren't really able to do anything that gave her joy during those nine months because we were so focused on trying to fix the problem because it was such a rare form of the disease.”

Hence Lucie’s interest in helping people with neurological disorders - whether that be MND, dementia or Parkinson’s - live better lives. As far as what can be gleaned from the vast amounts of data that will be collected from the MySense sensors she says it should give far better predictions of disease progression:

“Fast forward 18 months time, we'll be in a position where we can say okay, we've got enough people on the platform that for example, have Parkinson's or enough people that have MND, and actually we've seen this pattern of behaviour a thousand times before and therefore we're in a place where we can fundamentally understand what's going to happen.” And that should mean that care can be tailored better to each patient.

We are hearing a lot at the moment about the dangers of AI and the need to regulate it and perhaps take a pause in its development. But it would be a shame if the brakes were applied to attempts to use the technology to tackle one of the biggest health challenges we face.

stop and think about its development. But here is an example of where machine learning (which after all is what people mean when they say AI) could make a huge contribution to improving lives.