Browse All Terms

Variability or measurements taken within the same subject across different times or conditions.

A BCI system in which neural activities are recorded using a surgically implanted device. It involves placing sensors directly into the brain tissue to capture high-resolution signals from individual neurons.

A simple classification method based on measuring the distance between neural feature vectors.

A recursive algorithm for estimating and predicting noisy time-series signals, suitable for smooth BCI control.

A standardized protocol and software framework for synchronized acquisition, processing, and sharing of multimodal brain imaging data streams.

A piece of information attached to a data sample that identifies its class or category, providing the ground truth for machine learning algorithms to learn from and use for future predictions.

The process of identifying raw data (such as neural signals or images) and adding one or more relevant labels to provide context, enabling supervised machine learning models to learn from the data and make accurate predictions.

The aggregate electrical signal generated by the summed synaptic and other electrical activity of a local population of neurons in a small volume of brain tissue.[20] LFPs reflect the synchronized input to a neural area and are a primary signal source measured by both intracortical and ECoG electrodes.[28]

A measure of brain activity recorded from a small group of neurons in a specific area of the brain, capturing the collective electric potentials generated by local neuronal activity.

A type of RNN capable of capturing long-term dependencies in neural signals, such as EEG time series.

Study design in which the same subjects are followed and measured across multiple sessions or points in time.

A filter that allows frequencies below a certain cutoff to pass and attenuates higher frequencies.

A subfield of AI focused on enabling machines to learn from data and improve performance on specific tasks without explicit programming.

A medical imaging technique that uses strong magnetic fields and radio waves to create detailed images of organs and tissues within the body, especially the brain, without using ionizing radiation (like X-rays).

A topographic or graphic record, as a function of time, of magnetic fields associated with the electrical activity of the brain.

The imagination of a specific scenario or action by the subject, without any actual physical execution.

The amount of cognitive resources used when the subject is performing a task.

Protecting the confidentiality of personal brain data and ensuring unauthorized parties cannot access or use it.

Learning the ability to rapidly adapt to new users or contexts using minimal training data.

An invasive device composed of numerous microscopic electrodes designed to be implanted directly into the brain's cortex.[2, 29] MEAs offer the highest spatial resolution of any BCI modality, capable of recording the action potentials (spikes) of individual neurons or small neuronal populations, which provides the fine-grained information necessary for complex control tasks like operating a multi-fingered robotic hand.[13, 30]