Biophysical modeling to inform performance in motor imagery-based Brain Computer Interfaces
Résumé
Brain-Computer Interface (BCI), translates brain activity into commands. Despite its clinical applications, it fails to detect intents in 30% of the users, due to a poor understanding of the associated mechanisms. Here, we use a mathematical model to identify biophysical changes occurring while controlling a BCI.
We used a motor imagery-based BCI framework [1]. The cohort of 19 subjects was divided into G1 and G2 subgroups, with subjects who performed better or worse than the average.
We inferred four biophysical parameters of a neural mass model from the estimation of the power spectra for each subject during both rest and MI with MEG: two neural gains capturing overall synaptic strength between excitatory and inhibitory neuronal population (g_ei) and among inhibitory neuronal populations (g_ii), time constant of the excitatory neuronal population (tau_e), and time constant of the inhibitory neuronal population (tau_i) [2].
The model parameters show significant condition effects as follows: g_ei in regions involved in visual motion processing in G1 and in regions involved in the default mode network in G2; g_ii in regions involved in decision-making processes in G1 and in areas involved in attention processes in G2; tau_e in areas involved in visual recognition only in G2; tau_i in regions involved in motor imagery performance and in decision making processes in G1 and in areas involved in attention processes in G2.
We observed changes in the excitatory/inhibitory neuronal activity over sensorimotor areas in the most responsive subjects only, eliciting potential markers of BCI performance.
[1]M.-C. Corsi et al., “Functional disconnection of associative cortical areas predicts performance during BCI training,” NeuroImage, vol. 209, p. 116500, Apr. 2020, doi: 10.1016/j.neuroimage.2019.116500.
[2]A. Raj et al., “Spectral graph theory of brain oscillations,” Hum. Brain Mapp., vol. 41, no. 11, pp. 2980–2998, Aug. 2020, doi: 10.1002/hbm.24991.
Origine | Fichiers produits par l'(les) auteur(s) |
---|