Artificial intelligence (AI) machine learning is making a difference in assistive technology to help restore movement for the paralyzed. A new study in the American Institute of Physics journal APL Bioengineering shows how AI has the potential to restore lower-limb functions in those with severe spinal cord injuries (SCIs) by identifying patterns in brain signals captured noninvasively via electroencephalography (EEG).
“These findings establish a baseline for EEG decoding of lower-limb motor attempts in severely paralyzed individuals and pave the way for the development of brain-controlled neuroprosthetic systems,” wrote cor…
Artificial intelligence (AI) machine learning is making a difference in assistive technology to help restore movement for the paralyzed. A new study in the American Institute of Physics journal APL Bioengineering shows how AI has the potential to restore lower-limb functions in those with severe spinal cord injuries (SCIs) by identifying patterns in brain signals captured noninvasively via electroencephalography (EEG).
“These findings establish a baseline for EEG decoding of lower-limb motor attempts in severely paralyzed individuals and pave the way for the development of brain-controlled neuroprosthetic systems,” wrote corresponding author Silvestro Micera along with co-authors Laura Toni, Valeria De Seta, Luigi Albano, Daniele Emedoli, Aiden Xu, Vincent Mendez, Filippo Agnesi, Sandro Iannaccone, Pietro Mortini, and Simone Romeni.
The means to harness technology to improve the quality of daily living for the paralyzed is a worthy cause. An estimated 15 million people worldwide have spinal cord injury according to the World Health Organization. In the U.S., there are over 308,600 Americans living with traumatic spinal cord injury, and since 2015 78% of new cases are male according to the National Spinal Cord Injury Statistical Center. The lifetime cost for a 25-year-old with high-level quadriplegia, also known as high tetraplegia, can cost over USD 4.7 million according to the Christopher & Dana Reeve Foundation.
The researchers for this study are affiliated with Ecole Polytechnique Federale de Lausanne (EPFL) in Lausanne, Switzerland, the University Hospital Lausanne in Switzerland, Scuola Superiore Sant’Anna in Pisa, Italy, and the IRCCS Ospedale San Raffaele in Milan, Italy.
Four male patients with spinal cord injury ranging in age from 18 to 33 years old participated in this proof-of-concept study. Each participant performed four experimental sessions consisting of multiple trials over intervals of weeks. While seated in a wheelchair with at 64-channel EEG cap by ANT-Neuro, each participant was asked to visualize movements without attempting it and later attempt performing the movements. The four different movements were bending the left and right hips and extending the left and right knees.
After statical data analysis of the brain activity recordings, an AI machine learning classifier XGBoost (eXtreme Gradient Boosting) was used to analyze all of the sessions for each participant, as well as Monte Carlo iterative sampling.
“Our results suggest that EEG signals can often differentiate lower-limb movement attempts from rest, whereas decoding of left vs right and hip vs knee movements was more elusive,” the researchers reported.
With this proof-of-concept, the scientists have demonstrated that AI can decode brain activity recorded noninvasively with an EEG cap and tell the difference between intended lower-limb movement attempts from rest in severely paralyzed patients. This favorable feasibility assessment is the starting point for future novel noninvasive neuroprosthetics to help the paralyzed to move again using just their thoughts.
Copyright © 2026 Cami Rosso All rights reserved.