We investigate how co-registration errors affect the performance of anatomical constraints in magnetoencephalographic (MEG) source modeling for source-space connectivity analysis. We simulated MEG recordings from pairs of interacting cortical sources with variable locations and with a co-registration error of 0-15 mm. We computed a beamformer-based source estimate with either a free-orientation or constrained-orientation source model and tested four popular connectivity measures: Imaginary part of coherency, amplitude-envelope correlation, phase-slope index and frequency-domain Granger causality. We found that even small co-registration errors (≥2 mm) compromise the performance of source orientation constraints in connectivity estimation.
Influence of Co-Registration Errors on the Performance of Anatomical Constraints in MEG Source Connectivity Analysis*
Chella F.
Primo
;Marzetti L.Secondo
;Basti A.;Pizzella V.Ultimo
2019-01-01
Abstract
We investigate how co-registration errors affect the performance of anatomical constraints in magnetoencephalographic (MEG) source modeling for source-space connectivity analysis. We simulated MEG recordings from pairs of interacting cortical sources with variable locations and with a co-registration error of 0-15 mm. We computed a beamformer-based source estimate with either a free-orientation or constrained-orientation source model and tested four popular connectivity measures: Imaginary part of coherency, amplitude-envelope correlation, phase-slope index and frequency-domain Granger causality. We found that even small co-registration errors (≥2 mm) compromise the performance of source orientation constraints in connectivity estimation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
