Speakers - 2026

Title: Decoding the Mind: AI-Enabled EEG Modeling for Neural & Emotional Understanding

Abstract

Recent advancements in computational neuroscience, coupled with the transformative power of artificial intelligence (AI), have provided deeper insights into the complexities of the human brain. This study explores the integration of electroencephalography (EEG), computational modeling, and AI to better understand cognitive processes. EEG serves as a vital tool for recording electrical brain activity, enabling the analysis of neural patterns associated with various mental states and functions. By employing machine learning and deep neural networks, this research enhances the precision of EEG data analysis, uncovering hidden relationships and features within neural signals. The interdisciplinary framework examines domains such as attention, memory, and emotion, linking EEG-based findings with principles of modern neuroscience. The outcomes offer significant potential for early diagnosis of neurological disorders, personalized cognitive enhancement, and the development of advanced brain–computer interfaces (BCI) that bridge human cognition with technology. Ultimately, this study underscores the synergy between EEG, neuroscience, and AI in revealing the neural mechanisms underlying thought, emotion, and behavior, paving the way toward a new era of neurotechnological innovation.

What will the audience take away from presentation?

• The findings of this study underscore the effectiveness of the proposed hybrid deep learning framework that integrates supervised learning through Convolutional Neural Networks (CNN) and unsupervised learning via Stacked Autoencoders (SAE) for EEG-based emotion recognition. This fusion enables the extraction of more discriminative and meaningful features from EEG signals, leading to superior classification outcomes compared with conventional CNNs and other existing approaches.

• The findings also provides a clear direction for future research, emphasizing the need for transfer learning frameworks and cross-domain calibration techniques to overcome the limitations posed by dataset dependency. Furthermore, integrating multimodal inputs—such as physiological and behavioral signals—may enrich the interpretability and diagnostic utility of EEG-based AI systems. This research presents a promising pathway toward more generalizable, interpretable, and clinically relevant EEG analysis frameworks.

• This analysis and experimentation collectively highlight the dual significance of dataset intelligence and model innovation in advancing EEG research. By aligning standardized data practices with adaptive deep learning frameworks, future work can bridge the gap between computational development and clinical applicability— ultimately contributing to more reliable, interpretable, and patient-centered neuroinformatics solutions.