Quantifying Cerebellar Signal Detectability in MEG and EEG in Epilepsy Using Anatomically Informed Source Modeling.
bioRxiv : the preprint server for biology January 14, 2026 Teppei Matsubara, Abbas Sohrabpour, Seppo Ahlfors et al.
Cerebellar activity is hard to detect with standard magnetoencephalography (MEG) and electroencephalography (EEG) because of the cerebellum's depth, complex folding, and unfavorable source orientations. Analyzing clinical recordings from epilepsy patients, cerebellar signal-to-noise ratio (SNR) was consistently lower than superficial cortical reference levels. Reducing sensor-to-source distance alone, by placing on-scalp optically pumped magnetometers (OPMs) at standard sensor locations, did not improve cerebellar SNR. However, OPM layouts optimized for the posterior fossa produced substantial SNR gains in posterior cerebellar regions. Anatomical depth and geometry, not just sensor proximity, govern cerebellar detectability. This framework advances human brain mapping beyond the cerebrum.