Hands-On Data Science Projects with Brain Signals: From Raw EEG Data to Machine Learning Models

Electroencephalography (EEG) is one of the most accessible windows into the

human brain — and one of the richest sources of data for machine learning.

This book is a practical, code-first guide to building real EEG data science

pipelines, from your very first model to the cutting edge of self-supervised

foundation models.

Every chapter pairs clear explanations with a runnable Google Colab notebook,

so you can experiment with real EEG datasets (OpenNeuro, MOABB, and more) as

you read. You'll progress step by step:

- Understand EEG signals, electrodes, and montages, and run your first

end-to-end EEG machine learning application

- Set up a reproducible EEG ML project and build a traditional machine

learning pipeline — features, classifiers, evaluation

- Train deep learning models, including EEGNet and EEGConformer, for motor

imagery, P300, and other BCI paradigms

- Treat EEG as time series with modern libraries such as aeon (ROCKET,

HIVE-COTE)

- Explore self-supervised learning and foundation models for EEG with SelfEEG

(SimCLR, SimSiam)

- Design, prototype, and interpret your own custom EEG model architectures

Whether you're a student, researcher, or engineer moving from tutorials to

real projects, this book gives you a working toolkit — and the intuition

behind it — for turning raw brain signals into machine learning models you

can trust.