Skip to main contentOpen Accessibility MenuAccessibility Menu

Can AI and BCI revolutionise brain injury rehabilitation?

By Natalia Masztalerz (Assistant Psychologist), Sara Da Silva Ramos (Senior Research Fellow), & George Bergmann (Application Support Analyst)

Can AI and BCI revolutionise brain injury rehabilitation?

This month we dive into recent advances in artificial intelligence (AI), brain-computer interfaces (BCI) and their combined potential for use in rehabilitation.  

AI can be defined as a set of approaches used by a computer to learn based on data and predict outcomes. It can be seen pretty much anywhere we look in the modern world, Think of your Netflix suggestions or social media feed. But researchers are also beginning to explore how AI might be used to help people with brain injury: including diagnosis using AI-generated models in the interpretation of various types of non-invasive scanning; advanced behavioural analysis through capture and processing of interactive individual data; and the use of supportive technologies in rehabilitation, including combining AI with BCI to reduce physical disabilities, and using AI-controlled artificial environments to provide learning interactivity in daily tasks within a safe environment, while monitoring and analysing progress, to enable progress at a suitable rate for each person’s learning ability [1,2].  

The discovery of relationships between factors that influence the physical structures of our brains is pivotal to unlocking our understanding of how the brain behaves under stress and how it heals [3]. Recently proposed “Constitutive Neural Networks” use reverse-engineered building blocks from established models to predict impact of different forces on brain tissue [3]. The implications of this could fast-track our capacity for accurate diagnosis and treatment planning for people with brain injuries. BCI technology is designed to form a direct line of communication between our brain and the external world [1].

A study by Kennedy and colleagues [4] was one of the early developments of this technology. The single participant in this study learned to use it to type and produce synthetic speech. Similar advances are now being pursued by private companies such as Neuralink which has recently began human trials in the Precise Robotically Implanted Brain-Computer Interface – PRIME, a study which aims to evaluate the safety of their brain implant and surgical robot and assess the initial functionality for enabling people with paralysis to control external devices with their thoughts. [5].

When combined with AI, such as in Neuralink, BCIs can learn and adapt to the individual, which could be life-changing for people with severe disabilities.  

Despite its promise, there are some challenges in this area which require attention and further development. Previous studies found that up to 80% of people with loss of upper-limb function after stroke did not find BCI interventions helpful [6], so acceptance may be limited for some. The use of AI and BCI in cognitive training is still in its infancy as AI cannot, at least yet, accurately account for human emotional and mental states [7]. Also, where the use of BCI is concerned, there are still inaccuracies in the ability of decoding a person’s intentions based on their brain’s electrical activity alone [1].  

Developments in this field also pose some important ethical questions. These include concerns regarding the high costs, particularly of BCIs, which could be unaffordable for some people, potentially those who may need it most. Another dilemma is the risk for exploitation of our interests (conscious and unconscious), emotional reactions, memories, and intentions through the use of neural data [2]. This issue of data security and exhaustive risk assessment is especially pertinent when considering vulnerable populations.

Overall, since Kennedy and colleagues’ study in 2000, the use of technology in brain science has taken gigantic strides. AI and BCI technology is predicted to fuse more and more with everyday life and eventually with practice within healthcare settings in the not-too-distant future.

Employing these technologies in healthcare has the potential to support burnt-out staff and stretched services, while working together for the good of patients and communities. So, get ready, because revolution is on the way!

If you are curious to learn more about recent developments in AI, get in touch with us at research@brainkind.org to check out our recently published article in the Neuropsychologist exploring this area in further detail [8].  

References 

  1. Zhang, X., Ma, Z., Zheng, H., Li, T., Chen, K., Wang, X., Liu, C., Xu, L., Wu, X., Lin, D., Lin, H., Wang, X., Liu, C., Xu, L., Wu, X., & Lin, D. (2020). The combination of brain-computer interfaces and artificial intelligence: applications and challenges. Annals of Translational Medicine, 8(11), 712–712. https://doi.org/10.21037/ATM.2019.11.109
  2. de Filippi, E., Wolter, M., Melo, B. R. P., Tierra-Criollo, C. J., Bortolini, T., Deco, G., & Moll, J. (2021). Classification of Complex Emotions Using EEG and Virtual Environment: Proof of Concept and Therapeutic Implication. Frontiers in Human Neuroscience, 15, 483. https://doi.org/10.3389/FNHUM.2021.711279/BIBTEX
  3. Linka, K., st. Pierre, S. R., & Kuhl, E. (2023). Automated model discovery for human brain using Constitutive Artificial Neural Networks. Acta Biomaterialia, 160, 134–151. https://doi.org/10.1016/j.actbio.2023.01.055
  4. Kennedy, P. R., Bakay, R. A. E., Moore, M. M., Adams, K., & Goldwaithe, J. (2000). Direct control of a computer from the human central nervous system. IEEE Transactions on Rehabilitation Engineering, 8(2), 198–202. https://doi.org/10.1109/86.847815
  5. Neuralink. (2023) The Precise Robotically Implanted Brain-Comupter Interface (PRIME) study. https://neuralink.com/blog/first-clinical-trial-open-for-recruitment/
  6. Hendricks, H. T., Limbeek, J. van, Geurts, A. C., & Zwarts, M. J. (2002). Motor recovery after stroke: A systematic review of the literature. Archives of Physical Medicine and Rehabilitation, 83(11), 1629–1637. https://doi.org/10.1053/APMR.2002.35473 
  7. Zhao, J., Wu, M., Zhou, L., Wang, X., & Jia, J. (2022). Cognitive psychology-based artificial intelligence review. Frontiers in Neuroscience, 16, 1699. https://doi.org/10.3389/FNINS.2022.1024316/BIBTEX
  8. Masztalerz, N., & da Silva Ramos, S. (2024). What are the benefits and challenges of using artificial intelligence (AI) in neurorehabilitation? A very rapid review of the literature. The Neuropsychologist, 1(17), 21–32. https://doi.org/10.53841/bpsneur.2024.1.17.21
Pattern used for background spacing