Real-ECG extraction and stroke volume from MR-Compatible 12-lead ECGs; testing during stress, in PVC and in AF patients

Background

Due to the Magneto-Hydro-Dynamic (MHD) effect, blood flow within the MRI’s magnetic field (B₀) produces a large voltage during the S-T cardiac segment [1]. The peak MHD voltage (V_MHD) can be comparable, in higher-field MRIs, to the R-wave amplitude of the real Electrocardiogram (ECG_real), so that V_MHD reduces ECG-gating reliability and prevents ischemia-monitoring during cardiac imaging/interventions. We hypothesized that (1) separation of ECG_real and V_MHD from 12-lead ECGs acquired within a 1.5T MRI could be achieved, using adaptive filtering, based on a set of ECG calibration measurements, and (2) a non-invasive beat-to-beat stroke-volume estimation could be achieved from time-integrated systolic V_MHD.

Methods

Fig.1 shows 3 sets of 20-sec breath-held ECGs measured at positions (i), (ii) and (iii), utilizing an MRI-compatible Cardiolab-IT digital ECG-recording system [2]. The adaptive filtering procedure was tested in 5 healthy subjects, and 2 patients with Premature Ventricle Contractions (PVCs) and Atrial Fibrillation (AF). Validation was based on comparing the filter-derived ECG_real with ECGs measured periodically outside the MRI. The data processing block diagram (Fig. 2) includes training of adaptive Least-Mean-Square filters with ECG_real input (i), application of the trained filters to ECGs acquired in (ii) and (iii), which separates the V_MHD from ECG_real.

Figure 1

ECGs measured at 3 positions; outside the field (i) and at isocenter with head-in (ii) and feet-in (iii).

Figure 2

Adaptive filtering diagram used for intermittent PVC patients, with beats separated and then processed independently at abnormal/normal beat filters.

Results

PVC patient’s results (Fig. 3): (a) unprocessed surface-lead V6, (b) extracted ECG_real, and (c) V_MHD. In (b) S-T segment voltage is restored, and the R-wave dominates for gating. Aortic-flow vortices (c) generate oscillating-polarity V_MHD, with V_MHD peaking during S-T segment. Cardiac beat-to-beat stroke volume (d) was estimated from time-integrated systolic V_MHD. PVC beats produce substantially lower stroke volume than during sinus-rhythm. AF patient results (Fig. 4): (c) Irregular V_MHD and (d) irregular stroke volume are due to ventricular-filling differences at varying heart rates (100-140bpm). Athlete subject results (Fig. 5): Filter tracking of rapid heart-rate changes from 44bpm to 87bpm is shown during a treadmill stress test performed inside the MRI. V_MHD (b) and stroke volume (c) increase with heart rate, suggesting that the cardiac output matches higher demand. A stroke-volume comparison of all subjects (Fig 6), derived from time-integrated systolic V_MHDs, demonstrates the measurement’s sensitivity to pathology.

Figure 3

Results from a PVC patient (Ejection Fraction 20-25%, mitral regurgitation.

Figure 4

Results from AF patient (100-150 bpm)

Figure 5

Results from athlete subject during treadmill stress test (44-87 bpm)

Figure 6

Stroke-volume comparison (cardiac cycles n=20 per subject). Athlete (+24%), PVC (-54%) and AF (-23%), relative to average of volunteers.

Conclusions

The filtering extracts ECG_real from measured 12-lead ECG, preserving ECG_real for ischemia monitoring and MRI gating. Stroke volume can be non-invasively derived from the time-integrated systolic V_MHD.