Decoding brain signals to remedy aphasia

Every year in Switzerland, regarding 5,000 people[1] become aphasic. This disorder is characterized by the loss of the ability to find the words to express one’s thoughts. Using brain-machine interfaces, research teams are trying to decode the activity of the brain to restore the words of aphasic people.

A language disorder

At the origin of aphasia, we find lesions affecting the regions of the brain involved in language, located in the left hemisphere in most people. The most common cause is cerebrovascular accident (CVA): 30% of people affected will develop a language communication disorder. A cerebral head trauma, a neurodegenerative disease such as Alzheimer’s or a brain tumor can also be responsible. “In aphasia, it is not speech itself that is affected, but language,” explains Pierre Mégevand, assistant physician at the Neurology Department of the University Hospitals of Geneva (HUG) and assistant professor in the Department of Clinical Neurosciences at the University of Geneva (UNIGE). Everything that goes through words no longer works properly, whether orally or in writing, because it is the part of the brain that manages the link between words and their meaning that is damaged.

Depending on the areas of the brain affected and their extent, the aphasia will be different. It will more or less affect the ability to understand language (read, listen) or produce it (speak, write), with serious consequences on the quality of life of people who suffer from it and their loved ones. “Aphasia often has an immediate impact on their work, but also on social and private relationships. It can cause depression and lead to difficulties in daily tasks for people living alone, such as reading mail, paying bills or shopping, ”notes the specialist.

Decoding the brain to restore speech

Speech passes through motor nerve commands sent by the brain to move the tongue, lips, lungs, vocal cords and everything that allows us to articulate sounds. A research group has also shown that it is possible, using electrodes implanted in the brain and connected to a computer, to decode the electrical activity of the regions that control the articulatory apparatus to reconstruct what a person tries to say through a brain-machine interface. “This research led to the idea that we might restore the speech of people who had lost this ability using a computer that synthesizes thought speech in real time. This interface system was implanted in a first patient suffering from tetraplegia and unable to speak. By refining it and accelerating its functions, it might become very promising,” said the specialist.

At the same time, the language and hearing neuroscience team from the University of Geneva, led by Professor Anne-Lise Giraud, looked at the activity of the brain when a person imagines words or syllables. . “We carried out work with a view to implanting a new type of electrode for this interface,” says the researcher. For this, we recorded the brain activity of people already implanted with electrodes (to treat epilepsy) while they mentally pronounced certain syllables or words. Then we were able to decode that data using algorithms trained to track what the patients thought.” The team was thus able to determine which brain regions were the most interesting for implanting future electrodes and obtaining the best electrical signals. It also established the frequencies of brain activity and their most promising combinations to exploit for the decoding algorithm. In the future, the system may be able to decode syllables one by one to piece together words, a approach considered more reliable and easier than learning to decode all the words in the dictionary or the joint motor sequences. “We also want people to be able to train the algorithm themselves, to fine-tune the signal sent to the decoder. We are going to test this step very soon”, rejoices Anne-Lise Giraud.

______

Published in Le Matin Dimanche on 06/19/2022

[1] Source : https://aphasie.org/wp-content/uploads/2021/02/jahresbericht-2017-f.pdf

Leave a Replay