Optimized Positioning of ECG Electrodes for WSN Applications - Application and Multidisciplinary Aspects of Wireless Sensor Networks - page 200

I II III

aVR aVL aVF

V1 V2 V3

V4 V5 V6

1 2 3 1 2 3 1 2 t[s]

Fig. 9.15 Target ( black ) and Multivariate Linear Regression (MLR) reconstructed ( red ) 12-lead
electrocardiograms ( ECGs)

9.6 Conclusion

We have shown that the proposed methodology for measuring the electrical activity of
the heart enables improved patient mobility and comfort, requiring only a minimal set
of small wireless electrodes. Two different approaches to the 12-lead ECG recon-
struction from a limited number of leads have been demonstrated, PCA based on
unipolar leads, and MLR based on differential leads. High-resolution multichannel
ECG measurements were used as the input data. From the set of all possible differential
leads only those with closely placed electrodes were taken into account, since they
enable a simple implementation of wireless electrodes that are connected in a WSN.
Based on the two described statistical techniques, together with the use of
different time series similarity factors, optimal leads and a corresponding transfor-
mation matrix have been selected for every patient. The PCA-based methodology
selects four best unipolar reconstruction leads, which makes it applicable for
systems with reduced number of wired electrodes, connected to a gateway that
transmits the measured signals wirelessly, for example, to a personal terminal.