Using Statistical Measures from Causal Information Theory to Characterize Sleep Stages in Human Intracranial Data

The quality of sleep is fundamental for physical and mental health, and understanding its structure is essential for identifying related disorders. This work focuses on the efficient and accurate classification of sleep stages, one of the strategies proposed in the literature to address this issue. In this study, we propose characterizing sleep stages (Wake, N2, N3, and REM) using iEEG (intracranial electroencephalogram) data from 106 patients with focal epilepsy, analyzing only the regions unaffected by the disease.

To achieve this, we employed tools widely established in the literature: Permutation Entropy and Statistical Complexity Measure. The results obtained are significant not only for the characterization of normal iEEG data but also for the classification of sleep stages, which exhibit unique characteristics and are distinctly distributed in the 2D entropy-complexity plane. Furthermore, the experimental data showed consistency with simulations considering transitions between wakefulness and sleep, reinforcing the validity of the proposed approach.

The talk will be broadcast in the following zoom link: https://zoom.us/j/98286706234?pwd=bm1JUFVYcTJkaVl1VU55L0FiWDRIUT09