Machine learning models for proper rehabilitation strategies

Digital twins’ techs have come a long way. The term digital twin in the present context was first described and used by John Vickers of NASA in 2002. But digging a bit deeper will show that the concept itself is much older and complicated than this. (1)It was first applied in the 1970s during the Apollo XIII space program. NASA scientists needed to work with devices in outer space where it’s almost impossible to be physically present and these virtual representations were extremely helpful. (2)

Digital twins sound like science fiction, but they are very much real. And they’re making headway in the healthcare sector. In short, this technology allows us to create a virtual representation of a physical object or system. (3) The options have expanded during recent years to include huge items such as buildings, factories, and even people. Tech companies describe digital twins as the virtual representation of a physical object or system across its life cycle. It uses real-time data and other sources to enable learning, reasoning, and dynamically recalibrating for improved decision making. In short, they are highly complex digital models that are the counterpart, or twin, of a physical thing. (4) These ‘things’ could be a car, a tunnel, a bridge, or an athlete. Connected sensors on the physical asset collect data that can be mapped onto the virtual model. Anyone looking at the digital twin can now see crucial information about how the physical thing is doing out there in the real world. (5)

So the definition is that “A digital twin is a virtual representation of an object or system that spans its lifecycle is updated from real-time data and uses simulation, machine learning, and reasoning to help decision-making.”  (6)

Emergent technology like digital twin which we define here as a living replica of a physical system (human or non-human). A digital twin combines various emerging technologies such as AI, the Internet of Things, big data, and robotics, each component bringing its socio-ethical issues to the resulting artefacts. The question thus arises which of these socio-ethical themes surface in the process and how they are perceived by stakeholders in the field. A digital twin is a living model of the physical asset or system, which continually adapts to operational changes based on the collected online data and information, and can forecast the future of the corresponding physical counterpart .(7)

Digital twins help sports teams and accomplish a great deal, like:

• Visualizing treatments and training methods in use, by real users, in real-time
• Building a digital thread (injury?), connecting disparate systems, and promoting traceability
• Refining assumptions with predictive analytics
• Better decision making for return to play after injury
• Managing complexities and linkage within systems-of-systems like human health.

How can we apply Digital Twin in medicine and more precisely in sports medicine decision-making for return to play or injury prevention programs??

The use of digital twins in the healthcare industry is revolutionizing clinical processes and hospital management by enhancing medical care with digital tracking and advancing modelling of the human body. (8) Creating real-world scenarios, virtually a doctor testing an athlete, for example, would run a computer simulation to understand how the system would perform in various real-world scenarios. This method has the advantage of being a lot quicker and cheaper than avoiding a re-injury before testing in the pitch. But there are still some shortcomings.

First, computer simulations like the one described above are limited to current real-world events and environments. They can’t predict how the athlete will react to future scenarios and changing circumstances. Second, humans are more than muscles and bones. They’re also comprised of millions of lines of code because they are complex systems non-linear organisms. (11)

The technology can also be used for modelling an individual’s genetic makeup, physiological characteristics, and lifestyle to create personalized medicine. It has a more individualized focus than precision medicine which typically focuses on larger sample groups. (12)The goal of digitizing the human body and creating fully functioning replicas of its internal systems is to enhance medical care and patient treatment. (8)

Ideally, data scientists and developers will be able to leverage the partnership to quickly and easily build, train, and deploy machine learning models, to develop capabilities that may ultimately help predict and prevent injury to athletes. (9)

Sports injury is a complex emergent phenomenon and needs to be seen through a ‘lens of complexity. In this case, we should seek to identify features that are present in complex systems: (1) the pattern of relationships (interactions) between units (determinants); (2) the regularities (profiles) that simultaneously characterize and constrain the phenomenon and (3) the emerging pattern that arises from the complex web of determinants. (13)

Since injury is a complex phenomenon characterized by uncertainties and inherent non-linearity, an ACL injury will emerge when a specific pattern of interaction happens in the presence of an inciting event of a given value. Thus, the best manner to predict an injury is by understanding the interactions among the web of determinants and not the determinants themselves. (18)

Return to play and Digital Twin

One of the biggest challenges in sports medicine today is Return to Play decision making.The re-injuries rates are increasing the past 28 years for muscle injuries mostly and the need of an accurate way to decide how,and when a player is ready to return to the pitch is crucial.(19)This failure is most likely due to the following: 1) an over-reliance on treating the symptoms of injury, such as subjective measures of “pain”, with drugs and interventions; 2) the risk factors investigated for hamstring injuries have not been related to the actual movements that cause hamstring injuries i.e. not function. The experts reached consensus on the following criteria to support the RTP decision: medical staff clearance, absence of pain on palpation, absence of pain during strength and flexibility testing, absence of pain during/after functional testing, similar hamstring flexibility, performance on field testing, and psychological readiness..The modern sports teams produces tons of data regarding performance and injuries and the technology can help with the use of the Big Data and sports analytics to analyze them and to create novel decision making strategies in the SEM field.

Conclusion

The present study has shown that the digital twin is still, at the current technology level, almost indistinguishable from already-existing efforts to digitalize the healthcare and sports medicine process. The digital twin might of course in theory different from the (less emerging, more established) digital model or digital shadow. But until the connection between the athlete and the model is brought to life, the two worlds will share socio-ethical benefits and socio-ethical risks.

Author: Georgios Kakavas PT OMT ATC MSc(Phd cand.)

References:

1. M. Grieves. (2015). Digital Twin: Manufacturing Excellence through Virtual Factory Replication. [Online]. Available: https://research.fit. edu/media/site-specific/researchfitedu/camid/documents/1411.0 _Digital_Twin_White_Paper_Dr_Grieves.pdf
2. Q. Qi and F. Tao, ”Digital twin and big data towards smart manufacturing and industry 4.0: 360-degree comparison,” IEEE Access, vol. 6, pp. 3585–3593, 2018, doi: 10.1109/access.2018.2793265.
3. F. Tao and M. Zhang, ”Digital twin shop-floor: A new shop-floor paradigm towards smart manufacturing,” IEEE Access, vol. 5, pp. 20418–20427, 2017, doi: 10.1109/access.2017.2756069.
4. B. R. Barricelli, E. Casiraghi, and D. Fogli, ”A survey on digital twin: definitions, characteristics, applications, and design implications,” IEEE Access, vol. 7, pp. 167653–167671, 2019, doi: 10.1109/access. 2019.2953499.
5. J. Vachàlek, L. Bartalsky, O. Rovny, D. Sìsmìsovà, M. Morhàç, and M. Loksk, ”The digital twin of an industrial production line within the industry 4.0 concept,” in Proc. 21st Int. Conf. Process Control (PC), Bratislava, Slovakia, 2017, pp. 258–262.
6. B. R. Barricelli and S. Valtolina, ”A visual language and interactive system for end-user development of Internet of Things ecosystems,” J. Vis. Lang. Comput., vol. 40, pp. 1–19, Jun. 2017, doi: 10.1016/ j.jvlc.2017.01.004.
7. L. Zhu, P. Mussio, B. R. Barricelli, and C. Iacob, ”A habitable space for supporting creative collaboration,” in Proc. Int. Symp. Collabo- rative Technol. Syst. (CTS), 2010, pp. 617–622, doi: 10.1109/CTS. 2010.5478455.
8. L. Zhu, B. R. Barricelli, and C. Iacob, ”A meta-design model for creative distributed collaborative design,” Int. J. Distrib. Syst. Technol., vol. 2, no. 4, pp. 1–16, 2011, doi: 10.4018/jdst. 2011100101.
9. Björnsson, B., et al. 2020. Digital twins to personalize medicine. Genome Medicine 12 (1): 1–4.
10. 10.Brey, P. 2012. Anticipatory ethics for emerging technologies. NanoEthics 6 (1): 1–13 https://doi.org/10.1007/s11569-012-0141-7.
11. Bruynseels, K. 2020. When nature goes digital: routes for responsible innovation. Journal of Responsible Innovation 7 (3): 342–360 https://doi.org/10.1080/23299460.2020.1771144.
12. Bruynseels, K., F. Santoni de Sio, and J. van den Hoven. 2018. Digital twins in health care: Ethical implications of an emerging engineering paradigm. Frontiers in Genetics 9 (31) https://doi.org/10.3389/fgene.2018.00031.
13. Bittencourt NFN, Meeuwisse WH, Mendonça LD, et alComplex systems approach for sports injuries: moving from risk factor identification to injury pattern recognition—narrative review and new conceptBritish Journal of Sports Medicine 2016;50:1309-1314.
14. Coffey DS. Self-organization, complexity and chaos: the new biology for medicine.Nat Med 1998;4:882–85.
15. Rickles D, Hawe P, Shiell A. A simple guide to chaos and complexity. J EpidemiolCommunity Health 2007;61:933–7.
16. Philippe P, Mansi O. Nonlinearity in the epidemiology of complex health anddisease processes. Theor Med Bioeth 1998; 19:591–607.
17. Meeuwisse WH, Tyreman H, Hagel B, et al. A dynamic model of etiology in sportinjury: the recursive nature of risk and causation. Clin J Sport Med2007; 17:215–19.
18. Kakavas G, Malliaropoulos N, Pruna R, Maffulli N. Artificial intelligence: A tool for sports trauma prediction. Injury. 2020 Aug; 51 Suppl 3:S63-S65. doi: 10.1016/j.injury.2019.08.033. Epub 2019 Aug 19. PMID: 31472985.
19. Ekstrand J, Waldén M, Hägglund M. Hamstring injuries have increased by 4% annually in men’s professional football, since 2001: a 13-year longitudinal analysis of the UEFA Elite Club injury study. Br J Sports Med 2016;50:731–7.