Early Alzheimer’s Disease Detection Using EEG Brain Signals

Abstract

Alzheimer\'s Disease (AD) is a progressive neurodegenerative disease, and early detection is essential in order for meaningful interventions to improve patient outcomes. While there are diagnostic modalities such as PET scans that can be effective in detecting AD, they tend to be expensive, invasive, or impractical, illustrating the need for more usable alternatives to monitor the progression of AD. Electroencephalography (EEG) is a non-invasive and inexpensive way to monitor brain activity, and it has the potential of being a useful way to identify early markers of AD before significant cognitive decline as well. However, a reliable EEG detection model still must overcome the two significant challenges of a lack of large- scale labeled datasets sufficient for training, and considerable variability in EEG signals across individuals, which can ultimately greatly affect the generalizability of EEG models for helpful detection of AD through these differences. In this paper, we propose a novel deep learning framework to address these challenges with self-supervised learning. Our framework consists of a pre-training phase from a large, unlabeled corpus of general EEG data to extract the universal features of brain signals without needing to depend on scarce AD-specific labels. We then use a fine-tuning phase to adapt the model to the classification problem using a smaller, labeled AD dataset. The model architecture can acquire both spatial and temporal features from EEG signals, which leads to a more comprehensive method of analysis. The experimental results clearly indicate that this approach offers a substantial improvement in AD detecting performance and that there was an increase in F1 scores at both the sample and the subject level over the current state-of-the-art methods. This research outlines the ability of self- supervised contrastive learning to mitigate the problems of data scarcity and inter-subject variability, and can contribute towards more accurate and scalable AD diagnostic tools.

Introduction

supervised contrastive learning. Alzheimer’s is a progressive neurodegenerative disorder that affects memory, cognition, and daily functioning. Traditional diagnostic methods, such as MRI and cerebrospinal fluid analysis, can be expensive or invasive. EEG offers a promising alternative because it is non-invasive, relatively inexpensive, and capable of recording electrical brain activity.

The proposed system aims to overcome the challenges of noisy EEG signals and limited labelled data by combining sample-level and subject-level contrastive learning to learn useful features. EEG data from public datasets, including ADNI, AIBL, and OASIS, will undergo preprocessing steps such as filtering, artifact removal, normalization, and epoching. Features from temporal and spatial branches will then be combined and fine-tuned to classify individuals as having Alzheimer’s disease or being healthy controls. Majority voting across EEG segments will produce the final subject-level prediction.

The broader system architecture also describes feature extraction in the time, frequency, and wavelet domains, using machine learning and deep learning models such as SVM, Random Forest, CNN, and LSTM. The system is intended to store results securely and provide an interface for visualizing EEG signals, displaying prediction probabilities, and generating reports for clinicians and researchers.

The proposed framework will be evaluated through subject-level cross-validation and cross-dataset testing using accuracy, precision, recall, and area under the ROC curve (AUC). Its main goals are to improve diagnostic performance, generalization across datasets, and potential clinical applicability.

However, the supplied text also contains a separate section on deepfake detection using CNN, VGG16, Xception, and EfficientNet-B7 with attention. Its reported results show approximately 50% accuracy, zero precision and recall for the deepfake class, and signs of overfitting. These findings indicate poor classification performance and a need for improved generalization, class balancing, regularization, and data augmentation. This section appears unrelated to the EEG-based Alzheimer’s project and should be separated or corrected in the final report.

Conclusion

The EEG-based Alzheimer’s detection system provides an efficient and accurate method for early diagnosis of Alzheimer’s disease. By acquiring and preprocessing EEG signals, the system ensures clean and reliable data for analysis. Feature extraction in time, frequency, and wavelet domains captures essential patterns indicative of cognitive decline. Advanced machine learning and deep learning models, including CNN, LSTM, SVM, and Random Forest, enable precise classification of Normal, MCI, and Alzheimer’s cases. Secure storage of patient data and EEG records allows for longitudinal monitoring and research. The intuitive user interface presents results, probability scores, and reports for clinicians and researchers. The system reduces diagnostic time and supports timely intervention. It also provides a scalable framework for integrating new datasets and models. This project demonstrates the potential of combining EEG analysis with AI for neurological disease detection. Overall, it contributes significantly to improving healthcare outcomes through technology-driven solutions.

References

[1] Neural biomarker diagnosis and prediction to mild cognitive impairment and Alzheimer’s disease using EEG technology- Bin Jiao and Rihui Li contributed equally to this work. [2] Smart-Data-Driven System for Alzheimer Disease Detection through [3] Electroencephalographic Signals- Conceptualization, T.A. and P.M.R.; methodology, T.A. [4] A novel method for diagnosing Alzheimer’s disease using deep pyramid CNN based on EEG signals Wei Xia, Ran Zhang, Xiao Zhang, Muhammad Usman School of Medical Informatics and Engineering, Xuzhou Medical University, Xuzhou 221000, China. [5] Early dementia diagnosis, MCI-to-dementia risk prediction, and the role of machine learning methods for feature extraction from integrated biomarkers, in particular for EEG signal analysis. [6] Early Detection of Alzheimer’s Disease With Nonlinear Features of EEG Signal and MRI Images by Convolutional Neural Network - Elias Mazrooei Rad1 ID , Mahdi Azarnoosh1* ID , Majid Ghoshuni1 ID , Mohammad Mahdi Khalilzadeh1 IDB. R. Ranoliya, N. Raghuwanshi and S. Singh, \"Chatbot for university related FAQs,\" 2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Udupi, 2017.

Copyright

Copyright © 2026 Dhruv Manoj Patil, Srushti Raul, Utkarsha Bhoir. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.