by Jack Gann
Curator Jack Gann outlines the idea of virtual healthcare as explored in our new exhibition, You Choose.
More and more of our lives are lived today in the virtual world. Our digital avatars socialise, work, create, and fall in and out of love, but could they make us healthier? Could we get them to take our medication for us, or trial it at least? That’s the idea behind digital twins for healthcare, a subject that features in our latest exhibition.
But what is a ‘digital twin’? Put simply, it is a working virtual model of something that exists in the real, physical world. This could be an object, a system, or even a person. Crucially, it can’t be a static replica. It has to react and change to stimuli as its real world counterpart would. This allows the digital twin to be used for the same kind of testing that you would put something through in the real world.
Historically the digital twin concept originated with NASA. Although they only coined the term in 2010, the idea of debugging issues with real-time virtual versions of flight components was what helped the Apollo 13 spacecraft return to Earth after announcing to Houston that they had a problem. Other engineering industries took up the concept of the digital twin, with the automotive industry using it to model and improve everything from individual vehicle parts to whole traffic networks. And, if it works for car parts, couldn’t it work for body parts too?
A digital twin of an organ could be used to try out surgical procedures on the computer before carrying them out for real. HeartNavigator, for example, is a software produced by Philips that uses data from CT scan imaging to make digital twin hearts. This is then used to perform ‘virtual aortic valve replacement surgery’ on the digital hearts before doing it for real.
If it all sounds like something from a video game, well, it kind of is just that. Here in Leeds, researchers at the university received a €2.5 million grant from the European Research Council in 2023 to conduct ‘in silico’ (in the computer) trials on virtual prosthetic heart valves with virtual patients. As the University of Leeds press office put it: “Instead of recruiting people to a real-life clinical trial, researchers build digital simulations of patient groups, loosely akin to the way virtual populations are built in The Sims computer game.”
‘In silico’ trials can be run on large numbers of virtual patients simultaneously, producing results far quicker and cheaper than real world medical trials. Trials like these can be used to make sure your medicine is better suited to you too.
Since 2020, the Swedish Digital Twin Consortium have used digital twin technology to develop a personalised medicine strategy based on RNA. A digital twin of an individual patient is created and unlimited copies are made. Each twin is virtually treated with thousands of different drugs. The one with the best effects gets selected for use on the real patient.
This kind of personalisation of drug dosages is one of the subjects that our You Choose exhibition engages with. Something else that we highlight in the exhibition is hybrid partial closed loop automated insulin delivery systems (better known as an artificial pancreas). These allow people with diabetes automatically to control their insulin intake based on tracking their changing blood glucose levels.
An artificial pancreas on display in You Choose. The necessary component parts are a continuous glucose monitor, an insulin pump and an algorithm to communicate between them.
These systems are described as ‘hybrid’ because they still require some manual adjustment. The automated system is unable to handle mealtimes, for example. However, Medtronic, one of the makers of artificial pancreas technology available on the NHS, predict that a digital twin could be the answer to creating a fully closed loop system. Medtronic argue that “incorporating AI through a frequently updated ‘digital twin’ can circumvent daily challenges” in insulin and glucose management.
What that would mean in real terms would be a ‘whole body digital twin’ made from monitoring all of your lifestyle, not just your blood glucose levels. For example, heart rate monitors and other fitness sensors would be used to measure the intensity of any exercise that you do and adjust your insulin levels accordingly before your glucose levels have a chance to drop. Or geolocation on your phone could notice when you visit a favourite restaurant, predict what you tend to order from the menu, and deliver the necessary insulin before you even eat.
Whole body digital twins may be a little while off, but much of this is already possible with today’s technology. So, why aren’t we all testing every health and lifestyle choice on a digital doppelganger already?
A lot of it has to do with the difficulty of big data. In order to replicate real world systems in a useful way, digital twins for health have to rely on huge quantities of personal health data from a wide variety of sources. eHealth records, scans from MRI to xray, or data from wearable devices, none of these are particularly easy to synchronise and cross-reference. Even equivalent types of health data can be stored in a dizzying variety of non-interoperable formats. Equally, if your individual data is fragmented across different hospitals or institutions, it will be hard for your virtual self to be built with a complete picture. Digital twins require complex longitudinal data. The more incomplete the structures that they are built from, the more likely they are to produce inaccurate outputs. Digital twins for health probably won’t take off without the far less glamorous task of standardised health data formats.
There are also ethical concerns around the use of big health data. Your health data is both extremely personal and extremely private. Digital twins for health have to ensure that the data that they use is securely protected against unauthorised access and is used with informed consent. Meanwhile, a dataset that is biased towards certain demographics, whether of patients or medical conditions, risks the digital twins reflecting not just the valuable parts of the real world but also its prejudices. While the ideal is that digital twin technology works as a social equaliser, treating everyone as an individual and creating a more balanced healthcare system, it could instead create a digital divide.
Having a go at the ‘Digital Twin’ interactive in the exhibition
Nevertheless, it is that potential in digital twin technology to create personal health outcomes which excited us when putting together You Choose. The exhibition is an optimistic look forward at potential personal health futures and, as a result, ends with a fun interactive imagining the whole body health digital twin of tomorrow.
We don’t yet know how accurate our imagined twin might be, but digital twins present both opportunities and challenges for personalised health futures.