Digital twins are already being used in areas such as aircraft and automotive design. So what do they have to do with mental health? Can we make a digital model of a living molecular system?
By Naveesha Karunanayaka and Isabella von Holstein
A digital twin is a computer model of a system, such as an engine. They are updated in real time from sensor data, they model the whole lifecycle of the system, and they use artificial intelligence (AI) or machine learning (ML) to enable decisions (Figure 1). They are now being researched for use in the medical sector, creating a digital genetic and biometric profile of an individual. If that person gets ill, drugs and therapies can be tested in the digital ‘self’ before being administered to the real self.
Figure 1 – Possible digital twin pathways (Force Technology, n.d.)
Digital twin creation
The process of creating the digital twin involves a vast amount of data collection, using sensors and modelling, to build a parallel digital twin to the physical one. Then a computer program runs simulations on the individual’s genetic, biometric and behavioural information to detect the mind and body’s performance. This could allow the digital twin to detect issues before symptoms occur, saving lives, or at least preventing suffering. However, for this to occur, mental health researchers have to be able to show real evidence of causal links between genetic code and/or specific biometric markers, and health outcome for each condition.
Medical digital twins
Despite digital twins being a relatively new idea in the medical sector, progress has been fast. The 3D experience firm, Dassault Systems, have released the world’s first realistic computer model of the human heart called “The Living Heart”. This software can turn a 2D image into a 3D model of the individual’s heart. Applying this technology to the brain will be much more difficult, due to the brain’s complexity, and the relative lack of knowledge about the intricacies of the brain compared to the heart.
Gawel and colleagues at Linköping University Hospital created advanced computer models of 13 autoimmune, metabolic and malignant conditions using high-resolution data from individuals in a successful seven-year study. These digital twins can be ‘treated’ with a variety of drugs to discover which is most effective for each individual patient. Linköping University is hoping to stage clinical trials for cancer within the next few years. The twins will have regular updates and constant testing for signs of any changes in health. This is hoped to be implemented globally, paired with increased education about health and mechanisms of different diseases.
For patients with scoliosis, a condition in which the spine is curved sideways, a digital twin can predict changes in the cells in the spine, and suggest which physical therapy exercises would be most beneficial.
Digital twins in mental health
This idea will be revolutionary if it can be applied to mental illnesses. One of the pioneers in this area, Mikael Benson, believes “there’s far too much trial and error involved in diagnosing and treating people today”. There is already research and innovation in mental health treatment using AI and ML together with speech recognition. These methods use an individual’s changes in speech, activities or facial expressions to warn about potential illness and diagnose it. Modern technology enables collection over one million data points per patient per day via smartphone apps. The development of 5G networks adds additional potential for accurate real time data collection and access.
The healthcare system has not yet felt the advantages of using data analysis systems and AI for diagnosis and prognosis tools. Even an improvement in a small proportion of patient outcomes will greatly impact the healthcare industry, making the system more time- and cost-efficient. The use of digital twins will mean more personalised healthcare: rather than treatment based on the “average person”, it will be for the actual person.
Ethics of digital twins
There is still need for clarification on what a digital twin is for the public. A key to overcome this barrier is to use standard language, as many outside the professional field don’t yet understand the concept. It is hard to predict public response to this technology, as it has a several different ethical implications.
On the plus side, the use of a digital twin would vastly reduce the need for animal testing during clinical trials. Furthermore, side effects of drugs could be detected early and avoided by prior testing on the digital twin.
However, the creation of a digital twin could potentially breach a patient’s right to data privacy, as the twin is created with personal data and is shared with people who are not medical professionals. There is a risk that the capability to predict future health could be used to discriminate against people when applying for insurance policies or jobs. Patient data therefore needs to be strongly protected by law. To tackle issues such as this, Benson suggests anonymity of the data where necessary and that patients should own their own data.
The use of digital twin technology will bring with it a new era of personalised medicine in healthcare, including in mental health. Research developments in this field are currently very rapid. Digital twins could improve patient outcomes and satisfaction. However the technology needs to be implemented ethically.
This blog concludes our blog series on molecular science in Mental Health, through which key, prominent conditions that affect society today, including depression, anxiety and schizophrenia, were explored through a molecular science, engineering and technological lens. As well as discussing key diagnostic tools and treatment methods, we looked into the future of mental health and the healthcare industry.
What do you think this is going to look like? How can you see molecular science contributing to good mental health for us all in the future?
Duarte, S. (2019). Digital twins: needs, challenges and understanding. Available from: https://theodi.org/article/digital-twins-user-research/ [Accessed20th June 2021]
Erol, T., Mendi, A.F., and Doğan, D. (2020). The Digital Twin Revolution in Healthcare. In: 4th International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT), October 2020, Turkey [Online]. Turkey: ISMSIT, pp. 1-3. [Accessed 22nd June 2021]. http://dx.doi.org/10.1109/ISMSIT50672.2020.9255249
Force Technology (n.d.). Digital Twins. Available from: https://forcetechnology.com/en/services/digital-twins [Accessed 6th August 2021].
Gawel, D.R., Serra-Musach, J., Lilja, S. et al. A validated single-cell-based strategy to identify diagnostic and therapeutic targets in complex diseases. Genome Med 11, 47 (2019). https://doi.org/10.1186/s13073-019-0657-3.
Godard, B., Raeburn, S., Pembrey, M. et al. Genetic information and testing in insurance and employment: technical, social and ethical issues. European Journal of Human Genetics. 11, S123–S142 (2003). https://doi.org/10.1038/sj.ejhg.5201117
James, L. (2021), ‘Digital twins will revolutionise healthcare’. E&T, 16(2): 50-53. https://ieeexplore.ieee.org/document/9450624. [Accessed 8th April 2022]..
Medical Futurist Institute. (2020), Digital Twins and the Promise of Personalized Medicine. Available from: https://medicalfuturist.com/digital-twin-and-the-promise-of-personalized-medicine/ [Accessed 22nd June 2021]
Miskinis, C. (2018). Improving healthcare using medical digital twin technology. Available from: https://www.challenge.org/insights/digital-twin-in-healthcare/ [Accessed 23rd June 2021].
Roberts, L., Chan, S. and Torous, J. (2018). New tests, new tools: mobile and connected technologies in advancing psychiatric diagnosis. npj Digital Med 1(20176): 1-4. doi: https://doi.org/10.1038/s41746-017-0006-0.
Semic, R.F. (n.d.). Available from: https://www.semic.de/en/ai/semic-health [Accessed 6th August 2021].