NATIONAL PHD IN ARTIFICIAL INTELLIGENCE

Thesis title: Accessible and Explainable AI for EEG Decoding in Brain-Computer Interfaces

The field of Brain Computer Interface (BCI) is increasingly emerging as a highly relevant domain of research and experimentation, with growing implications for human well-being and activity. Within this context, the Electroencephalogram (EEG) signal, as a representation of brain activity, constitutes a crucial tool that enables both the investigation of neural mechanisms still largely unexplored and interventions across multiple application areas, including motor rehabilitation, emotional regulation, and mental health promotion. In recent years, technological development has accelerated remarkably, driven in particular by the exponential growth of models and methodologies in the domain of Artificial Intelligence (AI). The adoption of Machine Learning (ML) and Deep Learning (DL) techniques now makes it the interpretation and decoding of highly complex patterns in EEG signals, whichargely inaccessible through traditional analytical methods. At the same time, AI is increasingly recognized as a transformative force in healthcare, offering unprecedented opportunities to advance diagnosis, clinical management, and the development of innovative solutions within the domain of BCI. This doctoral thesis is situated at the intersection of neuroscience and AI and is structured primarily around two complementary areas of research. First, it addresses the development and integration of advanced EEG preprocessing techniques designed to function in both offline analysis and real time applications. Second, it investigates the application of state of the art ML and DL models for the decoding and analysis of EEG patterns. The practical contributions presented in this thesis introduce tools and frameworks that aim to bridge the gap between AI research and its translation into real world applications. By embedding the principles of Explainable Artificial Intelligence (XAI), the models developed in subsequent work are designed to be transparent and interpretable. Each contribution is described in detail in its respective chapter, providing a comprehensive overview of the research conducted and its impact on the fields of artificial intelligence and computational neuroscience.