DeepTokenEEG Enhancing Mild Cognitive Impairment and Alzheimers Classification via Tokenized EEG Features

2026-05-14

AI Summary

This study addresses the challenges of early Alzheimer’s disease (AD) diagnosis—namely, limited EEG data availability, insufficient accuracy of existing deep learning models, and low efficiency in expert interpretation—by proposing DeepTokenEEG, a lightweight end-to-end model with only 0.29 million parameters. The method introduces a novel spatiotemporal tokenizer that integrates time-frequency domain features to extract AD-related biomarkers effectively. Evaluated on a combined dataset of 274 subjects, DeepTokenEEG achieves 100% classification accuracy within specific frequency bands and outperforms current state-of-the-art approaches by 1.41–15.35% in overall performance. These results demonstrate a compelling balance between high diagnostic accuracy and clinical deployability, highlighting its strong potential for early AD screening.

Abstract

The detection of Alzheimers disease (AD) is considered crucial, as timely intervention can improve patient outcomes. Electroencephalogram (EEG)-based diagnosis has been recognized as a non-invasive, accessible, and cost-effective approach for AD detection; however, it faces challenges related to data availability, accuracy of modern deep learning methods, and the time-consuming nature of expert-based interpretation. In this study, a novel lightweight and high-performance model, DeepTokenEEG, was designed for the diagnosis of AD and the classification of EEG signals from AD patients, individuals with other neurological conditions, and healthy subjects. Unlike traditional heavy-weight models, DeepTokenEEG ultilizes spatial and temporal tokenizer that effectively captures AD-related biomarkers in both temporal and frequency domain with only 0.29 million paramaters. Trained in a combined dataset of 274 subjects, including 180 AD cases, and 94 healthy controls, the proposed method achieves a maximum recorded accuracy of 100% on specific frequency bands, representing an improvement of 1.41-15.35% over state-of-the-art methods on the same dataset. These results indicate the potential of DeepTokenEEG for early detection and screening of AD, with promising applicability for deployment due to its compact size.

Problem

• Alzheimer's disease
• Mild Cognitive Impairment
• EEG classification
• early detection
• diagnostic accuracy

Innovation

• DeepTokenEEG
• tokenized EEG features
• lightweight model
• Alzheimer's disease classification
• spatial-temporal tokenizer