The abstract is read at the publisher; the summary is JClub's.
Distortion-minimization validation for optically pumped MEG is tolerant to unmodelled sensor errors—under 2 mm per axis, 5° orientation, 10° array displacement, and 10% gain—when simulation and inversion match.
- Why it matters: Because cortical current-flow inversion is ill-posed, head-cast–style validation needs constraints for OP-MEG when key errors (sensor geometry and gain) are not modeled.
- What they did: Simulations of cortical current flow were contaminated with sensor position/orientation errors and gain changes, then distortion was minimized only when simulation and inversion assumptions matched; empirical error magnitudes were quantified in mm relative to the true cortical mesh.
- The result: With no unmodelled errors, distortion minimized under matched assumptions, and empirical OPM datasets showed that reconstructions using minimum norm and beamformer assumptions were sensitive to underlying anatomy.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.