Session S93.5

Are 2 ECG Leads Enough for Multilead Wave Boundary Location and QT Measuring?

R Almeida*, JP Martínez, AP Rocha, P Laguna

Universidad de Zaragoza
Zaragoza, Spain

According to the dipolar hypothesis, the heart’s electrical activity can be approximated by the electrical 3D heart vector (EHV) and the voltage measured at a given lead is the EHV’s projection into the unitary vector defined by the lead axis. Choosing a particular lead for ECG delineation determines a point of view over the cardiac phenomena and different latencies on the wave onsets and ends are found in different leads. Nevertheless, the onset and end of the cardiac electric phenomena are indeed unique, and therefore a global feature for all the leads. Thus combining adequately the information provided by multiple leads is essential for the correct location of lead-independent wave boundaries. A multilead-VCG strategy that locates the onset and the end of the ECG waves, attending to an optimal transformed lead according to the spatial characteristics of the VCG has been previously proposed and validated by out group[R Almeida et al IEEE TBME 2009, in press]. The system constructs a transformed spatial lead obtained from 3 orthogonal leads and optimized for delineation improvement. It then applies a single-lead delineation strategy (SL) based on the Wavelet Transform (WT) to the new synthesized lead. Considering WT loops in a 2D plane instead of in a 3D space is also possible, with some expected loss of performance, allowing to apply this methodology to any 2 ECG orthogonal leads, if the only available, as it is the case of many clinical Holter recordings. The goal of this work was to study and to quantify the performance lost when just 2 leads are used. Combination of 2 orthogonal leads were considered, both using recorded Frank leads (F) and leads synthesised from the standard 12-lead system using Dower transformation (D), and principal component (PC) analysis was used as alternative method for deriving orthogonal leads. The errors in QRS onset and T wave end location and in QT interval measurement were evaluated over the PTB database files (549 files at 1000 Hz, 12 standard leads + 3 Frank leads) as overall mean (m) and standard deviation (s). Considering the 3D approach we obtained m|s ms of -4.7|11.0 using F, -4.5|12.8 using D and -4.4|.6 over 3 PC for QRS onset, 7.3|20.0, 7.1|23.0 and 6.9|24.2, respectively for T wave end, and 11.8|22.1, 12.1|24.9 and 11.3|26.2 for QT interval. With 2D delineation we obtained, respectively, -2.8|12.0, -5.0|9.8 and -3.3|10.1 for QRS onset 8.2|24.1, 7.7|24.3 and 7.6|23.9 for T wave end and 11.0|25.9, 12.6|25.7 and 10.8|26.3 for QT interval. These results indicate that 2 leads properly selected are enough to locate QRS onset but possibly not for T wave end. This can indicate that the EHV changes at the beginning of the QRS are mainly along a single plane, while the T wave end comprehends spatial changes that require a 3D description.

(Abstract Control Number: 4)