Digital Twins — virtual replicas of real things – are already commonplace in manufacturing, industrial and aerospace. There are highly complex digital models of cities, ports and power plants – but what about people?
The idea of the digital doubles was long confined to the realm of science fiction. But one A new book launched at London’s Science Museum last week suggests the concept could come to life.
In virtual youPeter Coveney, Professor of Chemistry and Computer Science at University College London, and Roger Highfield, Director of Science at London’s Science Museum, show how far researchers have come in their quest to create accurate digital simulations of humans.
At the book launch, the authors were joined by leading experts in healthcare digital twins from the University of Oxford, UCL, and the Barcelona Supercomputing Center (BSC). The blackboard discussed the opportunities and challenges in creating a digital twin version of the human body and its implications for medicine.
The BSC has already created virtual models of living cells and entire organs. The most notable example is Alia Reda digital twin of a heart made up of around 100 million virtual cells.
The <3 of the EU tech
The latest rumors from the EU tech scene, a story of our wise founder Boris and some questionable AI artworks. It’s free in your inbox every week. Join Now!
The heart does not beat in people, but inside MareNostrum, one of the most powerful supercomputers in Europe. Working with medical technology company Medtronic, Alya Red’s simulations can help position a pacemaker, fine-tune its electrical stimulus, and model its action.
MareNostrum is located in the Torre Girona Chapel, Barcelona. Photo credit: Karolina Moon Photography.Perhaps one of the most striking examples is Yoon-sun, a 26-year-old Korean whose entire circuit – a 95,000 km network of ships – has been virtually mapped through an international collaboration involving multiple supercomputers. Researchers use the model to learn Blood pressure and the movement of blood clots throughout the vascular system.
These digital twins are not just limited to the lab. Some are already in use and partially approved by the US Food and Drug Administration (FDA).
So far, these models have mainly been used for in silico Trials – when a drug or disease is tested virtually rather than on real human or animal tissue.
These trials allow companies to test their drug on “virtual patients” before testing it on humans. This can help companies identify failures in the making early in the drug development cycle. says François-Henri Boissel, CEO of French Clinical Trial Simulation Platform Novadiscovery. This can result in a significant expenditure of time and money savings for companies conducting clinical trials.
In silico studies also remove the ethical issues associated with animal testing, explained Blanca Rodriguez, professor of computational medicine at the University of Oxford, during the blackboard discussion last Wednesday.
Rodriguez’s team created a digital twin of a heart that is used to simulate the effects of various drugs and diseases on heart function. In a virtual court hearing, her team tested the effects of 66 different drugs on over a thousand different heart cell simulations and was able to predict the risk of cardiac arrhythmias with an accuracy of 89%. Comparable animal research was 75% accurate.
These studies can also help combat the next public health emergency. During the COVID-19 pandemic, supercomputers have been used to simulate almost everything from possible treatments to predicting how the virus might spread, as highlighted in the video below.
And as simulations of human tissues, organs and cells become more sophisticated, they could open up new frontiers for vaccine testing and personalized symptom management, helping doctors study the impact of an infection throughout the body.
Digital twins could also speed up the search for vaccine candidates for the large families of animal viruses that are at risk of spreading to human populations, Highfield said.
Coveney and Highfield believe these advances will lay the groundwork for digital twins of the entire human body.
Computer models of patients would not only look like their human counterparts, but also behave like them.
In order to create a virtual you, enough personal data must be collected and analyzed to provide a realistic representation. This could be from any number of scans of your body and its organs, as well as genomic, biochemical, and wearable devices.
“These digital twins can shed light on what actions a surgeon takes, what medication you are prescribed, or even what type of life you choose,” Coveney said.
Your doctor could run a number of scenarios through your digital twin — like how you might respond to a certain medication or illness — without ever touching you (if you hate doctor visits, take note).
“Virtual patients could potentially save and prolong your life.
Your digital twin could accurately predict your disease risk and recommend medication, diet and lifestyle changes, potentially saving and prolonging your life.
According to the European Commission, around 200,000 people die every year in Europe from the medicines they are prescribed, partly because these therapies are generic and not specially developed for the patient.
The same goes for treating diseases: Doctors are forced to make decisions based on diseases similar but not identical patients in similar but not identical circumstances in the past.
“Modern medicine is like driving a car and looking in the rearview mirror — it’s always looking back to see what’s going on,” Highfield said. “The hope is that digital twins will enable healthcare to become future-oriented, truly personalized and predictive, and take much of the guesswork out of it.”
Perhaps less comforting is the thought that your twin might be used to it predict when you’re going to die, with pretty high accuracy.
So when do I get a virtual me?
Before you get too excited (or petrified) – let’s take stock.
Oxford University professor Denis Noble developed the first model of a beating heart cell in 1959. A few years later his work was extended to the level of cell patches and in the 1990s to whole heart models powered by the earliest supercomputers were beginning to take shape. Now programs like Alya Red allow simulations of almost every part of the human body.
That’s where we’re at right now. To date, a digital twin of an entire human has never been created.
We asked Midjourney’s AI bot to create an image of a “healthcare digital twin” – here’s the result.
There are still “massive technical hurdles” to be overcome, said Conveney.
Simulations of this complexity require access to incredibly powerful computers such as Frontier, the world’s first and fastest exascale supercomputer. These computers are still few and far between and require large amounts of energy run.
Another “big challenge,” he says, is putting all the codes together for each part of the virtual body. Each part of the digital human, like a cell or a heart, is technically a separate simulation. There are also multiple scales for the simulations: one model for one cell and another for the entire organ require different codes and run at different speeds. Loading all these codes at once and at the same speed is not an easy task.
There are also ethical considerations. The ability to predict almost anything about your health is a powerful tool for healthcare professionals, but potentially dangerous in the wrong hands.
Within the confines of current technology, creating your own virtual twin is currently only possible for billionaires, says Conveney.
Even the most powerful computers imaginable in the distant future will not have sufficient capacity to analyze you in all molecular details. You, my friend, are too complicated for even the smartest computer.
But Coveney and Highfield convincingly argue that incomplete digital representations will still be an extremely useful tool to advance medical science and the health of individuals. As the late British statistician George Box put it: “All models are wrong. Some are useful.”
The authors also hope that the computational demands of these twins can be reduced through the use of artificial intelligence.
“AI and machine learning can replicate some code and allow the entire digital twin to load at the speed needed for effective medical decision-making,” said Highfield. AI could allow virtual humans to run on much smaller machines.
While there are many hurdles to overcome and definitely some ethical issues to address, fully functional virtual patients that provide healthcare professionals with insights they can actually act on are not as far off as you might think.
Conveney, one of the world’s leading experts in this field, believes that virtual patients could be available for practical medical purposes in about five years.