Page 19 - Doing Data Together by The Scotsman

ACRC: AGILE RESEARCH FOR PUBLIC GOOD
I The aim of the ACRC is to
deliver “high-quality data-driven, personalised and affordable care that supports the independence, dignity and quality of life of people living in their own homes and in supported care environments”.
I More than 20 staff will be appointed and around 35 PhD students enrolled by 2025. “This will make our research very agile,” says Ian Underwood. “We can seed new projects and add depth to existing projects where needed.”
I The ACRC’s research is not proprietary or “owned” by anyone. “Legal & General has a clear interest in the ageing population, but this is
very much an open source public benefit project. We will learn a lot, but others can too,” says the firm’s John Godfrey.
I The New Technologies of Care work package will include “exploring and developing sensor-specific data- driven ‘internet of things’ platforms to produce accurate data about instant events” – vital signs and serious incidents like falls, plus short-term activities and long-term pursuits (physical and mental activities over weeks and months). This can extract predictive information and patterns to be used, among other things, to try to prevent adverse events, design care pathways and identify effective interventions.
is also a realisation that research use of health and social care data is constrained by a reliance on data recorded in very structured and specific ways, when a wide range of important information surrounding care is often not so structured.
Natural language processing – where artificial intelligence and other elements of computing science analyse human language – will be used to enhance existing well struc- tured data by adding additional data from “free-text” clinical records.
Professor Ian Underwood, an expert in sensors and director of the ACRC Academy, says: “As Pro- fessor Bruce Guthrie [the Univer- sity’s ACRC lead] puts it, we have ‘hard’ data about when people enter or leave hospital and what happens inside. But we don’t have the same level of data for care, which is more about general frailty and degenera- tion, often from several ill-defined conditions. Using NLP will allow us to take a more nuanced, subtle approach.”
Godfrey believes joining up the data is vital to delivering on the ambi- tions of the centre. He says: “When care sector data is not fully joined-up with healthcare data, it’s hard to get the full picture of an individual. We
need that if we want to ensure peo- ple are supported to live better lives for longer in their own homes, then enjoy high-quality care when they need to take that step.”
Wilkinson agrees: “Most of my work has been around people who are excluded and how to connect them with each other, with their fam- ilies and with society.
“The work package I’m leading is focused on that – drilling down into individual experiences to get a better view and explore where connections are broken or blocked.”
Part of this will involve traditional data gathering at regular intervals from just over 200 people in Scotland and north-east England (the Univer- sity of Newcastle is also involved in the project) from a range of rural, urban, suburban, coastal and post- industrial settings.
Wilkinson says: “Individual projects will then be able to draw down on a fascinating mix of data from that cohort in different ways and glean different things from it.
Bringing traditional
data together with anecdotal information from the person being cared for can help in tailoring support to
their individual needs. Picture: Shutterstock
“I’m very interested in that sense of space and place. One way of collect- ing data is ‘walk-along interviews’, to show how elements of the envi- ronment are relevant to a person’s care, health and well-being. Data is fundamental to everything we are doing at the ACRC; it’s about bring- ing all kinds of data together.”
It’s also about bringing all kinds of expertise together – not just data sci- entists and experts in robotics, infor- matics and AI, but engineers, built environment experts and a wide range of medical experts.
Underwood says that collabora- tion across university departments is essential to tackle big, real-life challenges.
He says: “As an engineer, I can help with connectivity. Engineering has been about connectivity through history – building roads, then ships, cars and planes and then TV, phones, computers, the internet. Now it’s all about data – connecting up data, curating it and analysing it, then putting it to good use. But I’m just part of the big picture.
“Diverse teams generally operate more effectively than teams of peo- ple who are very similar; it’s about people with different perspectives who bring different tools to prob-
lem-solving. You have to learn the languages of other disciplines and understand the skills and tools they bring, because you are part of a common goal and facing common challenges together.
Godfrey says this holistic approach is vital to avoiding narrow conver- sations and making real change. He says: “The debate is always about the financing of care, and politi- cians get stuck there. Wider debate grinds to a halt. Financing is impor- tant but we need to talk about care in a much broader sense, including the way buildings and care packages are designed.
“We know there isn’t a single answer to make care better – but we’re setting the bar very high and hoping for the emergence of whole new methods of care, based on sus- tainable healthcare and financial models.
“Data is fundamental; any changes must be strongly evidence-based. If we don’t know the full picture, at an individual level or population level, we cannot decide how to improve things. We want people to live longer with better health and greater inde- pendence – and technology and data should make it easier to achieve real change.”
If we don’t know the full picture, at an individual level or population level, we cannot decide how to improve things
19