Pranav Bhagirath

172 Chapter 9 Every BSP recording was immediately followed by an MRI scan in order to register the electrode positions to the anatomy of the volunteer. Magnetic Resonance Imaging After each BSPM recording, MRI markers were applied to replace all torso electrodes. These markers were used to locate the electrode positions on the MRI images, thereby minimizing the systematic error in the inverse procedure. Axial, coronal and sagittal anatomical images were obtained using a Turbo Spin Echo (black blood) sequence during breath hold (slice thickness 6 mm, no gap between slices). MRI was performed on a Siemens Aera 1.5 Tesla MRI scanner (Siemens Healthcare, Erlangen, Germany). For patients with an implanted pacemaker, pacing thresholds, P- and R-wave amplitude and lead impedance were determined before entering theMRI roomand the pacemaker system was programmed into MRI SureScan® mode [17]. These parameters were again determined after the examination and compared to the initial values. Finally, original programming of the pacemaker was restored. Inverse reconstruction of recorded ECG data From the MRI images, a 3D thorax model was constructed comprising the epicardial surface and the thorax volume conductor, accounting for lungs, liver and spleen. Epicardial potentials were calculated from the recorded BSP (P bs ) using P epi = (T T T + λ 2 I) -1 T T P bs , whereT is the forward transfermatrix and λ is the regularisation strength. Following each BSP recording, epicardial activation sequences were inversely reconstructed and visualised Evaluation of the quality of the inverse results To evaluate the quality of the results, ECGs were reconstructed from the inverse by forward transformation. The correlation between the recorded ECG potentials and the computed ECG potentials was subsequently determined. Note that while an a priori activation model was used for simulations to optimise the electrode positioning, no such model was used to perform the inverse reconstruction from recorded human BSP.