Weighted Time Warping Improves T-Wave Morphology Markers Clinical Significance

Background: T-wave (TW) morphology indices based on time-warping ($d_{w}$) have shown significant cardiovascular risk stratification value. However, errors in the location of TW boundaries may impact their prognostic power. Our aim was to test the hypothesis that a weighted time-warping function (WF) would reduce the sensitivity of $d_{w}$ to these errors and improve their clinical significance. Methods: The WFs were proportional to (i) the reference TW ($\mathcal {T}$), and (ii) the absolute value of its derivative ($\mathcal {D}$). The index $d_{w}$ was recalculated using these WFs, and its performance was compared to the unweighted control case ($\mathcal {C}$) in four different scenarios: 1) robustness against simulated TW boundaries location errors; 2) ability to retain physiological information in an electrophysiological cardiac model; 3) ability to monitor blood potassium concentration changes ($\Delta [K^+]$) in 29 hemodialysis (HD) patients; 4) and the sudden cardiac death (SCD) risk stratification value of the TW morphology restitution (TMR) index, derived from $d_{w}$, in 651 chronic heart failure (CHF) patients. Results and Discussion: The WFs led to a reduced sensitivity ($\mathcal {R}$) of $d_{w}$ to TW boundary location errors as compared to $\mathcal {C}$ (median $\mathcal {R}$=0.19 and 0.22 and 0.35 for $\mathcal {T}$, $\mathcal {D}$ and $\mathcal {C}$, respectively). They also preserved the physiological relationship between $d_{w}$ and repolarization dispersion changes at ventricular level. No improvements in $\Delta [K^+]$ tracking were observed for the HD patients (Pearson’s median correlation [$r$] between $\Delta [K^+]$ and $d_{w}$ was $0.86\leq r \leq 0.90$ for $\mathcal {T}$, $\mathcal {D}$ and $\mathcal {C}$). In CHF patients, the SCD risk stratification value of TMR was improved by applying $\mathcal {T}$ (hazard ratio, HAR, of 2.80), followed by $\mathcal {D}$ (HAR=2.32) and $\mathcal {C}$ (HAR=2.23). Conclusions and Significance: The proposed WFs, with $\mathcal {T}$ showing the best performance, increased the robustness of time-warping based markers against TW location errors preserving their physiological information content and boosting their SCD risk stratification value. Results from this work support the use of $\mathcal {T}$ when deriving $d_{w}$ for future clinical applications.

I. Introduction

The electrocardiogram (ECG) is a non-invasive diagnostic tool that has substantial clinical impact on investigating and monitoring the severity of cardiovascular (CV) diseases [1]–[3]. Several T wave (TW)-derived indices, reflecting the repolarization of the ventricles [4], have been proposed for risk stratification, like the TW width [5], the distance from the peak to the end of the TW (T interval) [6], the dynamics of the T interval [7], the QT adaptation time [8], and the TW alternans [9]. However, these markers do not capture variations of the whole TW morphology, which has demonstrated improved capacity for risk stratification [10]–[12].