Fig. 1.

Properties of the metal artefact seen in the MEG of DBS patients. (A) Raw data (high-pass filtered above 5 Hz) from several MEG channels close to the burr hole showing artefact deflections following the peaks of the ECG. The artefact exceeds the typical physiological signal amplitude by several orders of magnitude. (B) QRS-locked average of the MEG (all channels shown together) and the ECG. Numbered arrows point at some of the prominent artefact peaks and the corresponding MEG scalp maps are shown above the trace. The topographies vary greatly between peaks and could not be explained by a single source. (C) Plot of the singular values for decomposition of the QRS-locked average shown in B (circles) shown with the singular values from the same analysis performed on healthy volunteer data with (squares) and without (triangles) percutaneous extension wires glued to the subject's head. The artefact in the healthy subject was similar in magnitude to that of the patient (circles). For both the patient and the healthy subject there were about 6 singular values exceeding the maximal value observed without wires. This reflects the complex spatial topography of the artefact also seen from the scalp maps in B.