Closed-loop Non-invasive Brain Stiumlation

Transcranial magnetic stimulation (TMS) is a widely used tool for exploring brain function in humans (Siebner et al. et al., 2022), which has led to new therapeutics for various psychiatric and neurological disorders (Lefaucheur et al., 2020). However, the open-loop use of this technique has raised questions about its operating principle, due to the high degree of heterogeneity of results and the small to medium observed effect sizes (Zrenner and Ziemann, 2023). To increase the response rate, it has been suggested to individualize stimulation, by adapting the TMS parameters (i.e. delivered dose, target dose, targeting, timing, etc.) to instantaneous estimates of brain brain state. Such an approach, known as closed-loop closed-loop stimulation, is currently one of the main challenges challenges in this field (closed-loop brain state-dependent stimulation). To this end, we are focusing on the combination of robotic TMS and electroencephalography (EEG) (Hernandez-Pavon et al. al., 2023). The closed-loop stimulations using this combination developed to date have two limitations: (i) they are not adaptive and focus focus mainly on calculating the phase of brain oscillations to trigger stimulation and (ii) are limited to central cortical (sensorimotor) areas, where the EEG signal-to-noise ratio is optimal.

This project aims to develop closed-loop TMS-EEG protocols that overcome these two limitations: (i) by incorporating adaptive decision modeling (AutoHS model, Harquel et al. 2017) to optimize several parameters in parallel (coil location, orientation, intensity) while using a wider range of EEG markers (evoked potentials, oscillatory activity strength, connectivity, etc.), and (ii) by integrating real-time EEG pre-processing to access any cortical target (including frontal, temporal and occipital lobes).