A digital twin is a real-time digital model of an object or process that incorporates all available data and updates as new data becomes available. Researchers use digital twins to safely and economically test scenarios before trying them in real-life situations or environments.
Digital twins are already commonplace in engineering and manufacturing, but scientists are now looking to apply the same principles to the medical world. Digital twins give researchers the information they need to detect disease patterns and simulate the effects of treatments and identify the most promising paths for further research among real people.
In the past, it was unusual to see digital twins beyond industrial manufacturing because they were prohibitively expensive to build. Prolific and affordable new technologies have lowered the barrier to entry, expanding the use cases for this innovative technology and making digital twins more accessible. Thanks to that continued trend, digital twin technology is transforming how healthcare and life sciences work in order to transform the lives of the people they serve.
Data is key
The use of digital-twin technology enables “patient like me” comparisons across large cohorts of similar medical twins. This can help to identify biological markers for diseases and compare and test treatment options for patients who share similarities in age, gender, ethnicity and even underlying conditions. Such analysis would be impossible for medical professionals to do on the same scale with real-life patients.
Digital-twin technology has the potential to revolutionise the methodology of clinical research. The technology enables users to ask better questions, get better answers and derive actionable insights without risking the health of real-life subjects. Of course, using digital twins needs a data-first mindset. Gathering more data enables more digital twins, resulting in more discoveries and more optimised treatment.
Data is the key to unlocking the future of medicine. And with the advent of “big” health data – electronic health records, digitised medical images, genome sequencing – we are all candidates for digital twinning.
Advantages of digital twins
Digital twins help pool cohorts of comparable patients that would be impossible or prohibitively expensive to assemble in real life. The technology also avoids using real-life subjects with the accompanying risks and consent issues. If used appropriately, digital-twin technology in the healthcare sector enables clinicians to determine optimum therapies, improve patient outcomes and maximise efficiency, leading to hospital cost reductions.
Put simply, digital twins provide a safe and secure environment for testing the impact of change and will be critical to many sectors as we look to recover from the effects of the pandemic.
The pandemic has thrown up some new and unique challenges, and research collaborations have begun leveraging digital twins to help the millions affected. For instance, scientists are grappling with what many see as the next public health crisis – long-haul Covid-19. Estimates suggest that as many as one in 20 people who have had the virus are likely to suffer from Covid-19 symptoms lasting more than eight weeks. Some are still experiencing debilitating symptoms many months down the track.
Many people are too sick to withstand experimental treatment regimes, and it is still unclear why some recover quickly while others are left with long-term consequences. Digital twins have the potential to unlock the mysteries of this previously unknown condition, answering key questions such as: who is most at risk and why, what are the most common symptoms and what treatments are most effective?
We have partnered with i2b2 tranSMART Foundation, a non-profit open source research organisation based in Boston, to better understand and treat the impacts of long-haul Covid-19. Together, we built a data enclave to make this possible and provide the computational, artificial intelligence (AI), machine learning and advanced storage capabilities needed to generate digital twins.
In the data enclave, researchers gather, store and analyse anonymised patient data from around the world scattered across various monitoring systems and electronic health records – and the insights are shared with the broader clinical research community.
Using this technology, researchers use genetic background and medical history, combined with details of the long-term effects experienced, to perform millions of individualised treatment simulations. This means that researchers can identify the best possible therapy option, all with zero risk to actual patients. Patients who provide their data may even benefit directly and immediately if the research produces a relevant breakthrough.
The same technology used to understand long-haul Covid-19 symptoms can also help to create high-resolution, disease-specific medical digital twins that physicians and researchers can use for many other applications in the healthcare sector. AI-driven research and digital twins will support hospitals and research centres globally, allowing the use of technology at scale to advance health, education and economic opportunity for one billion people by 2030.
Importance of technology partnerships
Research of this type and scale requires the mobilisation of colossal amounts of de-identified patient data, including synthetic datasets to help prevent bias in the data and training of algorithms. Researchers must leverage significant computational, storage, AI and machine-learning capability to make these insights actionable – all while maintaining patient privacy and data security.
This is where we in the tech sector can make a genuine impact. By combining the experience, knowledge and skill of those in the scientific community with technology partners’ expertise and infrastructure, we stand our best chance of winning the fight against the pandemic – and digital twins could be a crucial part of our arsenal.
Over the next five years, many more organisations and industries will rely on digital twins to digitise their processes, and they will need a distributed, hybrid, multicloud environment to do so. For that, they will need a technology partner that can provide the right data management hardware, software and integration services, and at the same time protect their data, their simulation and the real-world object, process or person that it represents.
Arash Ghazanfari is UK CTO at Dell Technologies.
