Bio Digital Twins — Sci-fi or Reality?
A look at the digital twin technology, and its potential use in personalized medicine.
In the quintessential Disney classic, The Parent Trap, Lindsay Lohan stars as Hallie and Annie, two twins who have been separated at birth, and live with different parents. Wanting to meet the other parent, and perhaps influence them to get back together, they interchange places, and act as the other. This essentially allows them to try out new things, while at the same time making sure that their bases are covered and their parents aren’t suspicious.
While this isn’t a perfect analogy, it pretty much sums up the essence of the digital twin technology — a digital copy of a physical object that allows you to simulate potential what-if scenarios, collect information, and use it to make decisions that would impact the physical object itself. In other words, it’s the twin of an existing product or system, that is used to study how it would behave in different situations.
This results in cost-effective simulations of products, to improve the consumer experience and delve into product performance in real time environments. Another possible use would be to monitor equipment from the production floor using data collected from sensors, and thereby increase productivity. Improvements in the Internet of Things technologies, data availability and visualization, computing facilities, and modeling capabilities, have increased the potential of this technology.
Digital twins are now being used for a wide range of purposes — from implementing them in the manufacturing industry to optimize supply chains, to creating a virtual model of the city-state of Singapore to prevent disasters and improve the quality of life of its citizens. Digital twin technology is a major component of the Industry 4.0 technologies, which focus on improving efficiency to increase profitability by creating smart factories with a focus on automation. In a report in 2017, Gartner forecasted that, “Within three to five years, billions of things will be represented by digital twins, a dynamic software model of a physical thing or system.”
Now, I don’t mean to just throw buzzwords around, so let’s get to the interesting part. In the 21st century, data has become the new oil, and this is particularly apparent in the field of biology. With the evolution of genome sequencing and other cutting edge-tools, big data now plays a key role in medicine, for the diagnosis and treatment of diseases. Personalized medicine would take this a step further, integrating research studies, clinical trials, and medical records with omics, physiological, and phenotypic data, to deliver tailored treatment options according to an individual patient’s specifications.
Personalized medicine is already being implemented. The 20th century biologist, Raymond Pearl, worked in the field of constitutional medicine, and attempted to reveal the genomic and environmental factors of disease distribution. Breast cancer patients who overexpress the HER2 protein receive specialized care. Wearable devices and health apps can act as fitness trackers, monitor glucose levels, and even keep track of fertility windows!
However, personalized medicine using digital twin technology is a whole other ballgame. Bio digital twins are the result of the application of the digital twin concept to the life sciences, creating virtual models of the human body. These models are supplied with vast amounts of information, both of the patient, and from other resources and individuals as well. Current patient data would potentially be supplied from sensors placed in the human body. The modularity aspect of the digital twin technology plays a key role here, as we can integrate different modules depicting molecular profiles, environmental factors, and symptoms, to understand disease pathogenesis, and make diagnostic and therapeutic decisions.
Initially, this technology will be applied to patients in critical care, where they would be highly monitored, and data would be more accessible. As technologies evolve, the focus would shift to chronic disease care, and then later on to monitor the overall physical and mental health of a person and make accurate predictions about future conditions through the use of aggregate data, machine learning, analytics, and other algorithms.
Of course, there are a plethora of issues surrounding the collection of personal data, and the use of artificial intelligence. Unauthorized access to personal data has come under fire recently, with companies like Facebook and Cambridge Analytica using user information to influence political elections. The damage done by medical data leaks could potentially be far worse.
In the book Cell by Robin Cook, a smartphone app controlling a microchip that can give diabetic patients their insulin doses eventually murders certain patients by releasing the entire reservoir of insulin at once. The reasoning behind this is that the algorithm used is heuristic (a fancy word for solving problems in a faster and more approximate way), and terminates terminally ill patients to optimize medical resources for people with a better chance of survival.
Of course this is all merely speculative, but it is important to look at the big picture while coming up with any technology that has far-reaching implications.
There are a lot of groups performing research on bio digital twins right now. In the MEI lab of NTT Research, work is being done on creating a cyber alter ego for doctors to try different treatments on, thereby minimizing risk to actual patients. Automation Intelligence uses digital twin technology to improve pharmaceutical operators, and prevent recalls. The June 2020 issue of the IEEE Journal of Biomedical and Health Informatics had sections about the Internet of Medical Things for Health Engineering and other related topics.
There is a lot of work to be done, and ground to cover. The current global pandemic has given us a valuable lesson in terms of the importance of medical research and early warning systems. Diseases manifest in different people differently; it is high time our diagnosis and treatment methods reflect that. The bio digital twin technology, like all technologies, can make great waves of progress, if created with deliberation, attention to detail, and care for the end user.
And who knows, maybe one day we’d have come so far as to use digital models of ourselves to comfort our loved ones, long after we’ve passed on. What could possibly go wrong? As anyone who’s watched Black Mirror can tell you, a lot.
