the promise of brain implants
A Dutch man with chronic quadriplegia has been able to walk again thanks to a new brain-machine interface developed by a team of scientists in Switzerland. A therapeutic breakthrough followed by the invention of the whimsical Elon Musk and his company Neurolink intended to “increase the human”.
He thinks therefore he walks. Gert-Jan Oskam, a 38-year-old Dutchman, felt he would never be able to stand up and walk again after a serious accident more than 10 years ago. But an implant in his brain has allowed this chronic quadriplegic to regain the use of his legs, details the Swiss team of scientists behind this therapeutic breakthrough, in an article published Wednesday, May 24 in the journal Nature.
Gert-Jan Oskam was able to move several times, including climbing stairs, thanks to the device designed by these researchers. “I was able to get up for the first time in 10 years to go have a drink with my friends. It’s pretty cool”, he admitted to the daily The Guardian.
A “digital bridge”
“Our concept of a ‘digital bridge’ between the brain and the spinal cord announces a new era in the treatment of motor deficits caused by neurological disorders”, enthuse these scientists led by the Swiss Jocelyne Bloch and the French Grégoire Courtine, two leaders in their field.
This “digital bridge” consists of two implants: one placed on the surface of the brain and the other at the level of the spinal cord “under the lesion”, specifies Henri Lorach, head of the brain-spinal cord interface project at the Federal Polytechnic School of Lausanne and member of the team that carried out the operation.
These two boxes are connected via a wireless connection and can communicate with each other directly. “In the event of a spinal cord injury, the natural connection with the brain is severed, and the purpose of this bridge is to restore it by measuring the activity of the brain to then transmit the stimulations to the spinal cord”, summarizes Henri Lorach.
Easier said than done. The implant must first recognize the electrical impulses from the brain that correspond to the commands to walk, it must then decode them, and then transmit the correct information to the spinal cord.
The team of Jocelyne Bloch and Grégoire Courtine has been working on these brain-machine interactions for therapeutic purposes for several years and they “are recognized as precursors in this field”, assures Camille Jeunet-Kelway, specialist in brain-machine interfaces at the Institute of Cognitive and Integrative Neurosciences of Aquitaine (INCIA), a joint research laboratory of the CNRS and the University of Bordeaux.
These Swiss-based neuroscientists had already made a big leap forward in 2018. They had then identified the right electrical impulses which corresponded, at the level of the brain, to the commands to walk and had succeeded in creating a program capable of reproducing them in order to transmit the necessary stimuli to the spinal cord.
Gert-Jan Oskam was able to benefit from the improved version of this device. The big difference is that “everything is done this time in real time”, emphasizes Henri Lorach. The patient wants to walk and the device understands this and complies. “It’s a big step forward in the field of natural walking for these individuals,” admits Camille Jeunet-Kelway.
A pinch of AI
How does the implant succeed in “reading” thoughts? It is the result of a training phase. “They used a statistical method to accumulate evidence during the calibration phase,” explains Camille Jeunet-Kelway. Concretely, for several weeks, Gert-Jan Oskam had to think about moving so that the algorithm used to assist the patient correctly isolates the electrical impulses corresponding to the different movements.
After this training, the AI used will, thanks to what it has learned, “predict which cervical data to decode and transmit”, explains Henri Lorach.
The result is impressive in more ways than one. The device “decodes reliably and there are few false positives”, notes Camille Jeunet-Kelway. The implant not only decodes the correct signals, but it also correctly interprets the intensity of the order to be transmitted. Otherwise, the legs would risk making too large or too discreet movements to move forward or climb the stairs.
Finally, the scientists found that this work with the implants had also had a beneficial effect on the patient’s healing…once the device was removed. “Victims of this kind of neurological damage generally experience a therapeutic plateau after a certain period of rehabilitation. In this case, the patient seemed to have reached it, but after using the implant, his general condition improved and he exceeded the plateau. So the implant not only seems to make it possible to compensate for the handicap but also helps in healing”, notes Camille Jeunet-Kelway.
>> To read also: Elon Musk’s brain implants, science or science fiction?
It is therefore a very promising advance, but the results of which still need to be confirmed. The experiment, although crowned with success, involved only one individual. “The spinal cord injuries were severe but only partial, and we don’t yet know if the device will work with patients with deeper injuries or in other locations,” the authors of the Nature report acknowledge.
There are also limits related to “the autonomy of the device”, recognizes Henri Lorach. If the device breaks down while its user is in full operation, the consequences could be serious.
The effort to walk again after years of forced immobility is also intense. As a result, “we must also take into account the fatigue which imposes sessions limited in time”, notes Henri Lorach. One of the avenues is to succeed in miniaturizing the entire device – in particular the laptop computer which decodes the signals from the brain – in order to make it less bulky.
Very different from Elon Musk’s implant
In any case, brain-machine interfaces are on the rise. The day after the publication of the article by these neuroscientists, the whimsical multi-billionaire Elon Musk announced that the American authorities had given him the green light to test his implant – developed by his company Neurolink – on humans.
In 2019, the wealthy boss of Tesla, SpaceX, Twitter and Neurolink clarified that his implant would help millions of people with neurological disorders.
Will the scientists at the École polytechnique de Lausanne have to face competition from one of the richest men in the world? The two projects actually have little in common. The Neurolink implant must be inserted directly into the cortex as close as possible to the neurons. “Our method is less invasive because the electrodes are placed on the surface of the brain [mais tout de même sous la boîte crânienne, NDLR]. It’s more acceptable for the patient”, assures Heni Lorach.
Next, Neurolink considers that its mission goes beyond simply therapeutic goals. Elon Musk thinks that his implant will also make it possible to learn languages directly, to control his smartphone by thought, etc. “In fact, we don’t yet know what it should be used for,” sums up Camille Jeunet-Kelway. “There is a difference in philosophy, Elon Musk proposes to increase even healthy humans, while our purpose is only to help patients with neurological disorders”, summarizes Henri Lorach.
A more limited field of action but which nevertheless opens many doors. Walking is just the first step. The next objective “is to restore the movements of the arms in those who have lost the use of them”, affirms Henri Lorach.
And that’s another story. “The number of degrees of freedom [de mouvements] to take into account is much more important than for the legs, where it is essentially hip, knees, feet. Especially if we take into account the hands and fingers”, explains Camille Jeunet-Kelway.
In addition, much greater precision is needed in the analysis of signals and information. For example, picking up a drink and drinking requires the right range of motion to aim accurately, and enough but not too much pressure to hold the glass without breaking it…
This implant, which helps the patient directly, can also be useful beyond the case of tetraplegic or paraplegic patients. “From a medical point of view, when it comes to rehabilitation of motor functions, this type of implant can be useful,” assures Camille Jeunet-Kelway. The accompaniment of other diseases that lead to a reduction in motor skills – such as Parkinson’s disease – could also benefit from these implants.