سامانه همایش‌ها TSTA | ثبت‌نام و ارسال مقاله کنفرانس‌ها و کنگره‌های ملی

صفحه اصلی / سی و دومین کنفرانس ملی و دهمین کنفرانس بین المللی مهندسی زیست پزشکی ایران

Region-Specific EEG Channel-Based Emotion Detection using Bi-directional Deep Neural Networks

نویسندگان :

Mahdi Jafari Asl¹ Sina Shamekhi² Fatemeh Shalchizadeh³

1- دانشکده مهندسی پزشکی، دانشگاه صنعتی سهند، تبریز، ایران 2- دانشکده مهندسی پزشکی، دانشگاه صنعتی سهند، تبریز، ایران 3- دانشکده مهندسی پزشکی، دانشگاه صنعتی سهند، تبریز، ایران

کلمات کلیدی :

Emotion Recognition،region-specific channels،EEG،DEAP،Deep Learning،Hybrid Networks،CNN،Bi-GRU

چکیده :

Emotion recognition using EEG signals has garnered significant attention due to its potential applications in affective computing and brain-computer interface technologies. However, the nonlinear and non-stationary nature of EEG signals presents challenges for designing accurate and efficient models. In this work, we propose a hybrid deep neural network that combines a 1D-CNN with a Bi-GRU. The DEAP dataset is utilized for model evaluation, focusing on region-specific channels, particularly those located in the frontal and occipital lobes, which are physiologically and anatomically known to play critical roles in emotional processing. The frontal lobe regulates and interprets emotions, while the occipital lobe contributes to the visual pathways involved in emotional perception. The proposed model achieved 99.9% accuracy with 14 channels and 99.8% with 8 channels, using 1D-CNN-BiGRU in a four-class setting in the test subset. This study demonstrates that selective utilization of physiologically significant EEG regions can lead to robust and efficient emotion recognition models. The proposed approach enhances the feasibility of real-time and wearable affective systems.