Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 2, 2026Bitlis Eren Üniversitesi Fen Bilimleri DergisiOpen Access

EEG wavelet transform and ensemble learning achieves ~93% accuracy distinguishing AD, FTD, and normal subjects.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Early diagnosis of neurodegenerative diseases such as Alzheimer's Disease and Frontotemporal Dementia is essential for improving patient care and reducing healthcare burden.

Does a machine learning-based framework using Tunable Q-Factor Wavelet Transform (TQWT) accurately classify EEG signals in patients with Alzheimer's Disease, Frontotemporal Dementia, and cognitively normal subjects?

Population

88 participants (36 AD, 23 FTD, and 29 cognitively normal subjects)

Comparison

Classification of resting-state EEG signals using machine learning models

Key result

EEG feature extraction using Tunable Q-Factor Wavelet Transform and an Ensemble Learning classifier achieved 92.7% accuracy in distinguishing AD, FTD, and cognitively normal subjects.

Authors

MVMehmet VuralYAYaman AkbulutAYAbdulkadir Yelman

Discussion

Loading...

Member takes

Overview

Supports EEG-based ML for dementia subtyping; hypothesis-generating and requires prospective validation before clinical use.

Key Points

  • This research aims to classify EEG signals in individuals with Alzheimer’s Disease and Frontotemporal Dementia using the Tunable Q-Factor Wavelet Transform.
  • Analyzed EEG recordings from 88 participants (36 AD, 23 FTD, 29 cognitively normal) under resting-state conditions using 19 EEG channels.
  • Extracted 1881 features from EEG signals using multi-level TQWT and evaluated several machine learning classifiers, including Ensemble Learning and Decision Trees.
  • Applied a machine learning framework for the classification of EEG signals based on the extracted features.
  • Achieved a highest classification accuracy of 92.7% using the Ensemble Learning (Bagged Trees) classifier.
  • Demonstrated that rhythm-based features from TQWT effectively distinguish between AD, FTD, and cognitively normal subjects.
  • Highlighted potential for non-invasive dementia diagnosis through machine learning methodologies.

Study Design

Type

Cross-Sectional (n=88)

Structured PICO

Does a machine learning-based framework using Tunable Q-Factor Wavelet Transform (TQWT) accurately classify EEG signals in patients with Alzheimer's Disease, Frontotemporal Dementia, and cognitively normal subjects?

P
Population
88 participants, including 36 with Alzheimer's Disease, 23 with Frontotemporal Dementia, and 29 cognitively normal subjects, analyzed using resting-state EEG.
E
Exposure
Machine learning-based framework using Tunable Q-Factor Wavelet Transform (TQWT) for EEG signal classification
O
Outcome
Classification accuracy of Alzheimer's Disease, Frontotemporal Dementia, and cognitively normal subjects

TQWT-based EEG feature extraction combined with machine learning algorithms can effectively distinguish Alzheimer's Disease, Frontotemporal Dementia, and cognitively normal subjects with high accuracy.

Cite This Study

Vural et al. (2026) conducted a cross-sectional in Alzheimer's Disease and Frontotemporal Dementia (n=88). Tunable Q-Factor Wavelet Transform (TQWT) based EEG feature extraction and machine learning was evaluated on Classification accuracy. EEG feature extraction using Tunable Q-Factor Wavelet Transform and an Ensemble Learning classifier achieved 92.7% accuracy in distinguishing AD, FTD, and cognitively normal subjects.

synapsesocial.com/papers/6a45ffa29ed134303130ffc1https://doi.org/10.17798/bitlisfen.1772589
View Full Paper
Ask AI
Bookmark
Share