iEEG Based Brain-Computer Interfaces for Neural Encoding of Dynamic Natural Vision

This research delves into the complex task of interpreting visual stimuli based on intracranial electroencephalogram (iEEG) recordings, leveraging the power of advanced machine learning techniques, specifically Convolutional Neural Networks (CNNs). The study meticulously tackles the challenges encountered in the preprocessing phase of the data and the intricate process required to map the vast iEEG datasets accurately to corresponding visual stimuli. Despite facing these significant hurdles, the research provides profound insights into how brain activity is linked to visual perception. It significantly contributes to the understanding and development of brain-computer interfaces (BCIs) and advances the field of neuroscience. By exploring the potential of CNNs to decode visual information from iEEG data, this study not only pushes the boundaries of current technological capabilities but also opens new avenues for exploring the cognitive processes underlying human vision.