EEGDecoder-x: an explainable deep learning framework for cross-subject EEG-based detection of Alzheimer's and Creutzfeldt–Jakob disease

Interpretable self-supervised transformers for resting-state EEG analysis in Alzheimer's disease

Existing EEG-based methods have been constrained by limited availability of labeled data, hand-crafted features, poor spatio-temporal modeling, sub-optimal cross-hardware performance, and lack of interpretability due to being expensive and intrusive. To address these limitations, this study introduces innovative neural signal decoding techniques for cognitive state modeling in order to enhance the potential of AI-integrated models for early detection of Alzheimer's disease. This research introduces a self-supervised spatio-temporal transformer (STT-EEG) for early detection of Alzheimer's disease from resting-state EEG. This framework includes four key components: first, self-supervised pretraining on 111 healthy controls using masked auto-encoding and temporal order prediction to learn robust generalisable representations. Second, a novel spatial attention module (SAM) that explicitly captures both long-range temporal dependencies and channel interactions, reflecting the distributed network pathology of AD; third, cross-dataset transfer learning from 64-channel BioSemi to 19-channel Nihon Kohden systems, which showed strong hardware generalization; and finally, analyses of attention rollout and channel perturbation for clinically interpretable insights. The model was trained in a subject-wise 5-fold cross-validation fashion on the SRM dataset and fine-tuned on the OpenNeuro dataset (ds004504) consisting of 36 AD and 29 CN. On the same dataset, the accuracy of STT-EEG was 96.42% for AD vs. CN classification. The most significant improvement +7.08% was made with the help of self-supervised pretraining, followed by data augmentation +6.30% and the SAM +4.86%, as was confirmed in the ablation studies. For continuous prediction of MMSE scores, the Pearson correlation of the model was 0.872 and the mean absolute error (MAE) was 2.34 points. Regions of T3–T6, P3–Pz–P4 and O1–O2 were identified as areas of attention-based interpretability, which were consistent with the known neuropathology of AD that involved the temporoparietal lobe. STT-EEG strengths include the high interpretability of the framework, its spatio-temporal attention, and the fact that STT-EEG is a self-supervised learning method and can be used in an efficient and generalizable way.

Interpretable AI for neural signal decoding in dementia: an EEG ensemble approach to differential diagnosis

Introduction The differential diagnosis between Alzheimer’s disease (AD) and frontotemporal dementia (FTD) presents a significant clinical challenge due to overlapping early-stage symptom profiles. Conventional resting-state EEG provides limited sensitivity to the impaired neural plasticity and lateralized cortical degeneration frequently observed in FTD. Methods We developed a domain-informed heterogeneous ensemble framework incorporating dynamic neural reactivity and hemispheric asymmetry metrics from 19-channel EEG recordings acquired from 88 participants (36 AD, 23 FTD, 29 cognitively normal controls) during resting-state and photic stimulation paradigms. A neural reactivity vector (V_diff) was derived to quantify state-dependent spectral transitions. The 1,014-dimensional feature space was reduced to 200 features via recursive feature elimination, prioritizing spectral power distributions, hemispheric asymmetry indices (HAI), and stimulation-induced reactivity parameters. A weighted ensemble of Extreme Gradient Boosting (XGBoost) and Random Forest classifiers was evaluated using a subject-aware 90/10 holdout split with internal five-fold cross-validation. Results The optimized model achieved a multi-class segment-level accuracy of 95.63% on an independently held-out test partition of 1,281 segments, with an internal five-fold cross-validation mean of 0.9846 ± 0.003. FTD-specific precision reached 0.9907. SHAP analysis identified beta-band hemispheric asymmetry and alpha-band reactivity as the principal contributors to class separation. Discussion These findings indicate that the integration of dynamic state-transition measures with structural asymmetry proxies enhances electrophysiological discrimination between dementia subtypes. The framework provides a computationally efficient and biologically interpretable alternative to deep learning–based methodologies for EEG-driven dementia classification.

Leveraging a Foundation Model for the EEG-Based Diagnosis of Alzheimer's Disease

It is demonstrated that deep latent representations extract clinically relevant signatures from noisy signals, enabling precise, rapid, and data-efficient diagnosis.

A deep learning approach to mild cognitive impairment detection from electroencephalogram signals

ABSTRACT. Mild cognitive impairment (MCI) is an important condition that may progress to Alzheimer disease (AD) or other types of dementia. If MCI can be detected early, timely interventions may be implemented. Traditional diagnostic methods rely on neuropsychological tests and imaging studies but have limitations in terms of efficiency and accessibility. Therefore, electroencephalogram (EEG)-based deep learning approaches represent promising developments that may offer new frameworks for the early diagnosis of MCI. Objective: This study aimed to develop a long short-term memory (LSTM)-based architecture for classifying MCI using EEG signals. The temporal characteristics of EEG signals may reveal meaningful information that improves classification performance. Methods: A publicly available EEG dataset comprising 27 subjects (11 MCI; 16 normal) was used. Raw EEG signals were preprocessed using band-pass filtering, Independent Component Analysis, and segmentation. The 64-node LSTM model was trained using processed EEG segments for binary classification. The model was evaluated using standard performance metrics. Results: The proposed LSTM-based model achieved an accuracy of 98.14% for MCI versus normal classification, outperforming conventional algorithms such as K-Nearest Neighbors (KNN) and Support Vector Machine (SVM). The model also demonstrated high precision (99.21%), recall (98.69%), and F1-score (98.95%). The confusion matrix showed 4,774 normal and 3,257 MCI segments correctly classified, with very few misclassifications, highlighting its strong discriminative capability. Conclusion: This study highlights the potential of LSTM networks for early MCI detection. The findings suggest that deep learning-driven EEG analysis may be a valuable tool for noninvasive and scalable cognitive health assessments.

Machine Learning and Deep Learning for EEG-Based Dementia Prediction: A Comprehensive Survey

Dementia-related disorders, particularly Alzheimer’s disease (AD), represent a growing global health challenge, increasing the need for early and reliable detection. Electroencephalography (EEG), a non-invasive and cost-effective neurophysiological modality, has emerged as a promising tool for identifying neural signatures associated with cognitive decline. Recent advances in machine learning (ML) and deep learning (DL) have enabled more effective analysis of complex EEG signals for automated dementia prediction. This survey provides a comprehensive synthesis of EEG-based dementia studies published between 2020 and 2025, with a primary focus on Alzheimer’s disease (AD), frontotemporal dementia (FTD), mild cognitive impairment (MCI), and related dementia disorders. Unlike previous reviews that emphasize multimodal approaches or specific methodologies, this work exclusively focuses on EEG and presents a systematic comparison of ML and DL approaches, preprocessing pipelines, feature extraction techniques, publicly available datasets, validation strategies, performance metrics, and explainable methods. We further examine current research trends, identify methodological limitations such as small dataset sizes, subject-level data leakage, class imbalance, inconsistent preprocessing protocols, and limited multicenter validation, and discuss their implications for model generalizability and clinical adoption. We conclude by outlining future research directions toward developing robust, interpretable, and scalable EEG-based dementia prediction systems suitable for real-world clinical applications.

Explainable multimodal learning for early differential diagnosis oF Alzheimer’S disease and frontotemporal dementia