Single-Channel EEG Tokenization Through Time-Frequency Modeling

2025-02-22

AI Summary

To address the challenges of high temporal continuity, substantial noise, and poor suitability for direct sequence modeling in single-channel EEG signals, this paper proposes TFM-Tokenizer—the first time-frequency-aware discrete tokenization method tailored for single-channel EEG. Departing from multi-channel dependencies and continuous embedding paradigms, it leverages time-frequency analysis (CWT/STFT) to extract interpretable time-frequency motifs, which are then encoded into discrete tokens exhibiting class discriminability, frequency-domain enhancement, and explicit time-frequency structural interpretability. Integrated with a single-channel self-supervised masked prediction objective and a Transformer encoder, it establishes a dedicated EEG pretraining–fine-tuning pipeline. Evaluated on four standard EEG benchmarks, TFM-Tokenizer consistently surpasses state-of-the-art methods, achieving a 5% improvement in both balanced accuracy and Cohen’s Kappa on TUEV. It further enables robust cross-dataset transfer and facilitates NLP-style sequence modeling of EEG.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Machine Learning: Multimodal Learning

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

We introduce TFM-Tokenizer, a novel tokenization framework tailored for EEG analysis that transforms continuous, noisy brain signals into a sequence of discrete, well-represented tokens for various EEG tasks. Conventional approaches typically rely on continuous embeddings and inter-channel dependencies, which are limited in capturing inherent EEG features such as temporally unpredictable patterns and diverse oscillatory waveforms. In contrast, we hypothesize that critical time-frequency features can be effectively captured from a single channel. By learning tokens that encapsulate these intrinsic patterns within a single channel, our approach yields a scalable tokenizer adaptable across diverse EEG settings. We integrate the TFM-Tokenizer with a transformer-based TFM-Encoder, leveraging established pretraining techniques from natural language processing, such as masked token prediction, followed by downstream fine-tuning for various EEG tasks. Experiments across four EEG datasets show that TFM-Token outperforms state-of-the-art methods. On TUEV, our approach improves balanced accuracy and Cohen's Kappa by 5% over baselines. Comprehensive analysis of the learned tokens demonstrates their ability to capture class-distinctive features, enhance frequency representation, and ability to encode time-frequency motifs into distinct tokens, improving interpretability.

Problem

• Transforms noisy EEG into discrete tokens
• Captures time-frequency features from single channel
• Enhances interpretability with distinct token encoding

Innovation

• Single-channel EEG tokenization
• Time-frequency feature capture
• Transformer-based TFM-Encoder integration