Jorge.

Back

EEG Seizure Detection with CNN–Transformer + Adversarial Learning

01

Overview

  • This work focuses on automatic seizure detection using multi-channel EEG signals through deep learning techniques.
  • A hybrid CNN–Transformer architecture is used to capture both spatial and temporal dependencies in EEG data.
  • The project emphasizes real-world applicability by evaluating models under strict inter-patient conditions.
Overview
Problem
Dataset & Setup
EEG Processing Pipeline
Baseline Model (CNN–Transformer)
Adversarial Learning
Results
Key Takeaways