Related Experiment Videos
Automated detection of trigeminal neuralgia using multi-domain EEG feature analysis and CNN-attention architecture: a
Sina Samieirad1, Sajjad Rezvani Boroujeni2, Shayan Rokhva3
1Department of Otolaryngology-Head and Neck Surgery, Sinus and Surgical Endoscopic Research Center, Mashhad University of Medical Sciences, Mashhad, Iran.
Scientific Reports
|July 16, 2026
Summary
This study developed an electroencephalography (EEG)-based automated detection framework for trigeminal neuralgia (TN). A novel Convolutional Neural Network with Attention model showed promising accuracy, identifying key neurophysiological signatures in TN patients.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Computational Neuroscience
Background:
- Trigeminal neuralgia (TN) diagnosis relies on subjective reporting, leading to delays and misdiagnosis.
- Objective, non-invasive diagnostic tools are needed to improve TN detection and management.
Purpose of the Study:
- To develop and evaluate a proof-of-concept electroencephalography (EEG)-based automated detection framework for TN.
- To compare the performance of various machine learning architectures for TN classification using multi-domain EEG features.
Main Methods:
- Resting-state EEG data were collected from 36 TN patients and 36 healthy controls.
- EEG signals were preprocessed, decomposed into frequency bands, and 3,800 features were extracted.
- A two-stage feature selection identified 28 optimal features; six classification architectures were evaluated using cross-validation.
Main Results:
- The Gamma band and frontal regions showed the highest discriminative power for TN detection.
- A novel Convolutional Neural Network with Attention model achieved high preliminary accuracy (96.44%) and AUC (0.996).
- Identified neurophysiological signatures include Gamma-band entropy alterations and frontal complexity measures.
Conclusions:
- EEG-based automated detection shows promise as a non-invasive approach for TN diagnosis.
- The CNN with Attention model demonstrated encouraging performance, warranting further investigation.
- External validation on larger, independent cohorts is crucial to establish generalizability and clinical readiness.