AI-based analysis and interpretation of electroencephalography

PhD study program: Applied Mathematics
Akademic year: 2023-2024
Advisor: Ing. Mgr. Roman Rosipal, DrSc. (roman.rosipal@savba.sk)
External educational institution: Institute of Measurement Science SAS
Accepting university: Faculty of Mathematics, Physics and Informatics, Comenius University in Bratislava, Department of Applied Mathematics and Statistics

Annotation:

Electroencephalography (EEG) represents an effective neuroscience research tool for understanding brain functioning. The undemanding nature of EEG recording facilitates the acquisition of a large amount of EEG data collected under different conditions and across different cohorts of subjects. However, specific problems of EEG data associated with phenomena such as high intra- and inter-subject variability, artefact contamination, etc., make physiological inference and learning from EEG data challenging. Artificial intelligence (AI) and machine learning (ML) techniques may be helpful tools to mitigate these problems and provide enlightening solutions and conclusions. The thesis will explore the applicability of AI and ML approaches and models in solving problems associated with EEG data. More importantly, it will investigate and develop tools and methods leading to better interpretability of these models. Simulated and real EEG data collected during the experiments with brain-computer interface protocols will be used.