English | EN 
A Brain-Computer Interface (or BCI) allows a person to control a machine using only their brain activity, without any physical movement. This technology falls under human-machine interfaces, but in this case, it focuses exclusively on brain signals.
For example, a person with paralysis can use a BCI to move a cursor on a screen simply by thinking about an action. How does it work? The system captures brain waves using electrodes placed on the head, then translates them into commands that a computer can understand.
Before it can work in real time (known as the online phase), the BCI needs a training period (offline phase). During this phase, the computer learns to recognize patterns of brain activity linked to specific thoughts or intentions. It is a bit like teaching an AI to recognize words in a foreign language.
A BCI usually works in three main steps:
- Signal acquisition: Electrodes record brain activity while the person imagines an action or mental image.
- Signal processing: The data is cleaned and analyzed, and a model, often based on artificial intelligence, identifies which command the signal corresponds to.
- Action and feedback: If the system correctly recognizes the intention, it executes the command (like moving a cursor). In some cases, it also sends feedback to the user to help them adjust their mental activity.
This kind of technology opens up incredible possibilities, not only for people with disabilities, but also for future applications in video games, communication, and even smart home systems.