Optimisation of electrode placement for new ambulatory ECG monitoring devices

In this study we aim to determine, from body surface potential map (BSPM) data, the optimal bipolar chest electrode placement for maximum R-wave amplitude. The study data consisted of 117-lead 352-node BSPM data recorded from 229 healthy subjects. The dataset was split into a training set of 172 subjects and a testing set of the remaining 57 subjects. Optimal electrode placement was determined using a lead selection method based on the difference in R-wave amplitude across all 352 nodes for each patient. R-wave values were then extracted and used to create a median BSPM of the training data. From this median BSPM the optimal electrode placement was defined as the location of the minimum and maximum R-wave values. On the testing dataset this new optimal bipolar chest lead (R-lead) was then compared to all of the leads of the Mason-Likar 12-lead ECG and previously described bipolar chest leads, CM₅, CS₅, CC₅ and CB₅. The R-lead showed significant improvement in median R-wave amplitude over the next best lead, CM5 (2562μV vs. 2420μV, Wilcoxon sign ranked test, p< 0.001). Given the improvement in signal strength, an improvement in automated R-wave detection and R-R interval analysis from single lead ECG monitors may be achieved.