Computational neuroscienceDry phantom for magnetoencephalography -Configuration, calibration, and contribution

- We developed a new MEG phantom and its calibration method.
- The calibrated signal source position shifted 0.83 mm, from the designed one.
- The uncertainties of the signal source position can be obtained by the calibration.
- We demonstrated that using a calibrated phantom achieves more appropriate evaluation of the MEG measurements.

BackgroundAn artificial object that imitates human brain activity is called “phantom” and is used for evaluation of magnetoencephalography (MEG) systems. The accuracy of the phantom itself had not been guaranteed in the previous studies, although role of the phantom is to evaluate the accuracy of MEG measurement. The purposes of this paper are to develop a novel MEG phantom that can be calibrated and to demonstrate the advantages of the calibrated phantoms.New methodWe proposed and fabricated a practical dry phantom that is composed of 50 isosceles-triangle coils based on Ilmoniemi's model. This phantom was calibrated based on three-dimensional measurement of the current paths in the phantom and on numerical calculations.ResultsThe calibrated positions of the equivalent current dipoles (ECDs) shifted 0.83 mm, on average, from the designed positions. The uncertainties of the calibrated ECDs were also evaluated, by combining the uncertainties which could reasonably be attributed to them.Comparison with existing method(s)Furthermore, we demonstrated performance of the developed phantom through experimental evaluation of an MEG system. The results of this evaluation differed from those obtained using an uncalibrated phantom. Moreover, the calibrated phantom can provide detailed information regarding the uncertainty of the measurement and also the uncertainty of the phantom itself.ConclusionsA more appropriate evaluation of MEG measurements can be achieved using a calibrated phantom.