What is the clinical evidence from this study?

Study design: Observational. Population: Alzheimer's disease and Frontotemporal dementia (n=88). Intervention: Convolutional neural network (CNN) based framework using EEG data vs. Healthy controls. Primary outcome: Classification accuracy for Alzheimer's disease vs. healthy controls using 5-second EEG segments (10-fold cross-validation).

June 4, 2025Open Access

Advancing Alzheimer’s disease detection: a novel convolutional neural network based framework leveraging EEG data and segment length analysis

Key Result

A novel convolutional neural network framework using 5-second EEG segments achieved 97.08% accuracy in detecting Alzheimer's disease compared to healthy controls.

Study Design

Type

Observational (n=88)

Multicenter

Structured PICO

Does a CNN-based framework leveraging EEG data and optimal segment length accurately detect Alzheimer's disease and Frontotemporal dementia?

Population

88 participants from the Department of Neurology of AHEPA General University Hospital of Thessaloniki, divided into Alzheimer's disease (n=36, mean age 66.4±7.9, 13 males), Frontotemporal Dementia (n=23, mean age 63.6±8.2, 14 males), and cognitively normal (n=29, mean age 67.9±5.4, 11 males).

Intervention

Convolutional neural network (CNN) based framework leveraging EEG data with varying segment lengths (5s, 10s, 15s, 20s, 25s, 30s, and 30s with overlap).

Outcome

Classification accuracy of Alzheimer's disease vs. healthy controls and Frontotemporal dementia vs. healthy controls.

A novel CNN-based framework using EEG data achieved high accuracy in detecting Alzheimer's disease and Frontotemporal dementia, demonstrating the importance of optimal segment length in classification performance.

Limitations

Use of a single publicly available dataset limits generalizability across diverse populations and clinical environments.
Small population size and restrictive duration of EEG recordings.
Absence of alternative datasets that include both AD and FTD data for cross-validation purposes.

Abstract

Alzheimer's disease (AD) is a progressive neurodegenerative disorder that primarily affects memory, thinking, and behavior, leading to dementia, a severe cognitive decline. While no cure currently exists, recent advancements in preventive drug trials and therapeutic management have increased interest in developing clinical algorithms for early detection and biomarker identification. Electroencephalography (EEG) is noninvasive, cost-effective, and has high temporal resolution, making it a promising tool for automated AD detection. However, conventional machine learning approaches often fall short in accurately detecting AD due to their limited architectures. We also need to investigate the impact of EEG signal segment length on classification accuracy. To address these issues, a deep learning-based framework is proposed to detect AD using EEG data, focusing on determining the optimal segment length for classification. This framework contains EEG data collection, pre-processing for noise removal, temporal segmentation, convolutional neural network (CNN) model training and classification, and finally, evaluation. We have tested different segment lengths to test the impact on AD detection. We have used both 10-fold and leave-one-out cross-validation techniques and obtained accuracy of 97.08% and 96.90%, respectively, on a publicly available dataset from AHEPA General University Hospital of Thessaloniki. We have also tested the generalizability of the proposed model by testing it to detect frontotemporal dementia and obtained better results than existing studies. Furthermore, we have validated our proposed CNN model using several ablation studies and layer-wise extracted feature visualization. This study will establish a pioneering direction for future researchers and technology experts in the field of neurodiseases.

Mark Helpful

Bookmark

Relay

View Full Paper