نهمین کنگره نقشه برداری مغز

نمایش مقاله

Title:

Stress Brain Mapping Based on Dynamic and Individualized EEG Microstates

Subject:

Advanced Techniques in Signal Acquisition and Image Processing

Presenter Author:
Name:	Yasamin Baghersad

Author:
Name 1:	Yasamin Baghersad - Presenter
Affiliation 1:	Department of Biomedical Engineering, Faculty of Engineering, Islamic Azad University, Isfahan, Iran

Name 2:	fateme kazemi - Corresponding
Affiliation 2:	d Department of Biomedical Engineering, Raghib Isfahani Institute of Higher Education, Isfahan, Iran

Name 3:	soheil sarvari
Affiliation 3:	d Department of Biomedical Engineering, Raghib Isfahani Institute of Higher Education, Isfahan, Iran

Name 4:
Affiliation 4:

Name 5:
Affiliation 5:

Name 6:
Affiliation 6:

Name 7:
Affiliation 7:

Name 8:
Affiliation 8:

Name 9:
Affiliation 9:

Name 10:
Affiliation 10:

Abstract

Original Research

introduction :

Stress has emerged as a fundamental challenge to mental health in modern societies, exerting significant effects on both physiological functions and cognitive processes. Despite its importance, conventional stress assessment methods, which rely on self-reporting, often lack accuracy and fail to capture real-time fluctuations. In this study, a novel approach for brain-based stress mapping using individualized and dynamic EEG analysis is presented. Raw EEG signals were first preprocessed with advanced techniques, including adaptive filtering, noise removal, and wavelet-based analysis, and were examined in both time and frequency domains. Key indicators, such as signal variation rates, absolute and relative power in delta, theta, alpha, beta, and gamma bands, along with cognitive markers related to stress, were extracted. The derived features were input into a Random Forest machine learning algorithm, successfully distinguishing different stress levels with stable accuracy exceeding 90%. The resulting topographic maps visually and interpretable display the spatial distribution of neural activity, enabling precise, quantitative, and individualized stress assessment. These findings demonstrate that focusing on EEG signal dynamics provides an effective avenue for developing advanced tools for stress monitoring and management in both clinical and research application

Methods:

In this study, EEG signal analysis began with loading the raw data in the standard EDF (European Data Format). During the preprocessing stage, the primary EEG channels were first separated from non-relevant channels, and to enhance data quality, a band-pass filter, a 50 Hz notch filter, and a wavelet-based denoising method were applied.Subsequently, time-domain and frequency-domain analyses were performed, and features such as the signal rate of change, as well as the absolute and relative power of the delta, theta, alpha, beta, and gamma frequency bands were extracted. In addition, key frequency ratios associated with cognitive states and stress levels were computed.Finally, topographic maps of brain activity were generated, and the performance of the classifier was evaluated using accuracy and precision metrics.

Results:

The results indicate that the proposed EEG processing system is capable of effectively processing raw EDF data and extracting meaningful signal features. The preprocessing stage, which included channel separation and noise removal using band-pass filtering, notch filtering, and wavelet transform, significantly improved signal quality. Time-domain and frequency-domain analyses, including the rate of change and the power of the delta, theta, alpha, beta, and gamma bands, as well as statistical indices such as mean and energy, were extracted. The combination of these features enabled the derivation of cognitive indicators such as stress level and provided a clear visualization of topographic maps of brain activity.

Conclusion:

The results demonstrate that dynamic and personalized analysis of EEG microstates provides a reliable framework for distinguishing stress levels. Focusing on temporal variations and frequency-band power across different bands enables the identification of subtle neural markers associated with cognitive and emotional responses. The high accuracy of the Random Forest algorithm confirms the importance of the features extracted from EEG signals. The presented topographic maps enable a clear visual representation of brain activity, facilitating individualized and quantitative assessment of stress levels. These findings highlight the strong potential of dynamic EEG analysis for stress monitoring and management in both clinical and research applications.

Keywords:

EEG Signal, Signal Processing, Brain Mapping, Microstates