Brain Computer Interface Course

Lesson 1 • BCI System Architecture Overview

Introduces the signal acquisition, processing, and output components of a BCI pipeline. Frames the full system view students will build throughout the course.

Lesson 2 • Neuroscience Essentials for BCI

Covers neuronal communication, brain regions, and electrophysiology relevant to BCI design. Connects biological signal sources to measurable electrical activity.

Lesson 3 • BCI Application Domains

Surveys motor restoration, communication, neurorehabilitation, and consumer BCI uses. Motivates technical learning by connecting concepts to real-world impact.

Lesson 1 • Electroencephalography Fundamentals

Explains EEG electrode placement, cap systems, and signal generation. Establishes EEG as the primary non-invasive modality used throughout the course.

Lesson 2 • Hardware Setup and Calibration

Guides practical configuration of acquisition systems for reliable data collection. Connects hardware knowledge to reproducible experimental protocols.

Lesson 3 • Invasive Recording Techniques

Covers electrocorticography and intracortical microelectrode arrays for high-resolution recording. Contrasts invasive signal quality with non-invasive accessibility.

Lesson 4 • Functional Neuroimaging for BCIs

Introduces fMRI, fNIRS, and MEG as complementary acquisition modalities. Highlights spatial resolution advantages and practical deployment constraints.

Lesson 1 • Digital Filtering Techniques

Covers bandpass, notch, and high-pass filters for isolating neural frequency bands. Directly addresses power-line and drift artifacts introduced in Chapter 2.

Lesson 2 • Epoch Segmentation and Rejection

Segments continuous recordings into task-aligned epochs and rejects bad trials. Ensures downstream feature extraction operates on valid data segments.

Lesson 3 • Re-referencing Strategies

Explains average reference, linked mastoids, and Laplacian spatial filters. Shows how reference choice affects signal topography and BCI performance.

Lesson 4 • Baseline Correction and Normalization

Applies pre-stimulus baseline subtraction and z-score normalization across epochs. Standardizes signal amplitude for consistent feature extraction.

Lesson 1 • Spatial Filtering for Feature Enhancement

Uses common spatial patterns and beamforming to maximize class discriminability. Extends re-referencing concepts into supervised spatial optimization.

Lesson 2 • Time-Domain Feature Extraction

Derives amplitude, slope, and event-related potential features from EEG epochs. Establishes the simplest feature set before introducing spectral methods.

Lesson 3 • Spectral and Time-Frequency Analysis

Computes power spectral density and time-frequency representations of neural signals. Captures oscillatory dynamics relevant to motor imagery and attention BCIs.

Lesson 4 • Feature Selection and Dimensionality Reduction

Applies filter, wrapper, and embedding methods to select the most informative features. Prevents overfitting by reducing high-dimensional feature vectors.

Lesson 1 • Performance Evaluation Metrics

Measures accuracy, ITR, kappa, and ROC-AUC for rigorous BCI system evaluation. Connects classifier output to real-world communication throughput.

Lesson 2 • Classical Machine Learning Classifiers

Applies LDA, SVM, and k-NN to BCI feature vectors with cross-validation. Provides baseline decoding performance before introducing advanced models.

Lesson 3 • Deep Learning for BCI Decoding

Implements CNNs, RNNs, and EEGNet architectures for end-to-end BCI decoding. Leverages raw or minimally processed signals to learn spatial-temporal patterns.

Lesson 4 • Adaptive and Online Classification

Updates classifiers in real time to handle non-stationarity and session drift. Prepares students for deployment in live BCI systems.

Lesson 1 • Motor Imagery BCI Paradigm

Designs cue-based motor imagery protocols and integrates CSP-LDA decoding pipelines. Applies feature extraction and classification skills in a complete paradigm.

Lesson 2 • P300 Speller Paradigm

Implements the row-column oddball matrix and xDAWN-LDA decoding for text entry. Demonstrates event-related potential paradigm design and averaging techniques.

Lesson 3 • Hybrid and Passive BCI Paradigms

Combines multiple paradigms or uses passive monitoring to improve robustness. Extends single-paradigm knowledge to multi-modal and implicit BCI designs.

Lesson 4 • Real-Time BCI Software Frameworks

Configures open-source BCI platforms for stimulus delivery, acquisition, and decoding. Integrates all pipeline components into a deployable real-time system.

Lesson 1 • Slow Cortical Potential Training

Trains voluntary control of slow cortical potential shifts for communication and self-regulation. Demonstrates DC-coupled amplification requirements and long-latency feedback.

Lesson 2 • Evaluating Neurofeedback Outcomes

Assesses behavioral, neurophysiological, and clinical outcomes of neurofeedback interventions. Applies rigorous evaluation methods to distinguish true learning from artifacts.

Lesson 3 • Closed-Loop Neurostimulation

Integrates transcranial stimulation triggered by decoded brain states for bidirectional modulation. Extends closed-loop concepts beyond feedback displays to active intervention.

Lesson 4 • Principles of Neurofeedback

Explains operant conditioning of neural oscillations and the feedback loop mechanism. Grounds neurofeedback design in learning theory and neuroplasticity.

Lesson 1 • Regulatory Pathways for BCI Devices

Navigates medical device classification, safety testing, and approval processes for implantable and non-invasive BCIs. Connects technical design to compliance requirements.

Lesson 2 • Ethics, Privacy, and Informed Consent

Examines neuroethical principles, data privacy obligations, and consent frameworks for BCI research and products. Ensures graduates apply ethical reasoning to system design.

Lesson 3 • Long-Term BCI Reliability and Maintenance

Addresses signal drift, electrode degradation, and model recalibration over extended use. Prepares students to sustain BCI performance beyond initial deployment.

Lesson 4 • Handling User Variability and Non-Responders

Analyzes sources of inter-subject variability and strategies to improve BCI performance for all users. Builds on adaptive classification to address the BCI illiteracy problem.