Multimodal time series analysis of non-invasive human electrophisiology to characterize brain – heart – gut interactions

Over the past decades, advances in neuroscience, biomedical signal processing, and neural engineering have significantly deepened our understanding of brain function and human physiology. Brain function is now increasingly viewed as an emergent property of large-scale networks, extending beyond isolated regions to include dynamic interactions with peripheral physiological systems. In particular, growing evidence highlights a continuous bidirectional interplay between the brain, the heart, and the gut, which contributes to homeostatic regulation, emotional and cognitive processes. However, despite this progress, robust experimental studies explicitly designed to characterize simultaneous brain–heart–gut interactions remain limited. This thesis, conducted at the Paris Brain Institute (ICM), investigates electrophysiological interactions among brain, heart, and gut in 28 healthy resting-state participants. Multimodal recordings of EEG, ECG, and EGG were processed through a dedicated pipeline to extract time-varying neural, cardiac autonomic, and gastro-intestinal dynamics. Coupling was assessed using spectral coherence, maximal information coefficient, and transfer entropy, accounting for time delays. Statistical significance was evaluated via surrogate-based Monte Carlo testing, and the results were integrated into a network representation of inter-organ interactions. The findings provide preliminary exploratory evidence of triadic functional interactions among brain, heart, and gut activity at rest. This work establishes a methodological foundation for future multi-organ network models and for the identification of biomarkers reflecting dysregulation across interconnected physiological systems.