Weighted Time Warping Improves T-Wave Morphology Markers Clinical Significance

Area of Science:

• Cardiology
• Biomedical Engineering
• Signal Processing

Background:

• T-wave (TW) morphology indices derived from time-warping (d_w) offer significant cardiovascular risk stratification value.
• Errors in TW boundary localization can compromise the prognostic accuracy of these indices.
• Investigating methods to enhance the reliability of d_w is crucial for clinical application.

Purpose of the Study:

• To evaluate the efficacy of weighted time-warping functions (WFs) in reducing d_w sensitivity to TW boundary location errors.
• To assess whether WFs can improve the clinical significance and prognostic power of TW morphology indices.
• To compare the performance of different WFs against a standard unweighted approach.

Main Methods:

• Developed WFs proportional to the reference TW (T) and its derivative (D).
• Recalculated d_w using WFs and compared performance against the control (C) in simulated boundary errors, cardiac models, potassium monitoring, and sudden cardiac death (SCD) risk stratification.
• Assessed robustness, physiological information retention, potassium tracking in hemodialysis patients, and SCD risk stratification in chronic heart failure (CHF) patients.

Main Results:

• WFs significantly reduced d_w sensitivity to boundary location errors (median R=0.19-0.22 for WFs vs. 0.35 for C).
• WFs preserved the physiological relationship between d_w and ventricular repolarization dispersion.
• While no improvement in potassium tracking was observed, WF T significantly enhanced SCD risk stratification in CHF patients (Hazard Ratio=2.80).

Conclusions:

• Proposed WFs, particularly WF T, enhance the robustness of time-warping based cardiovascular risk markers against location errors.
• These WFs maintain physiological information content and improve SCD risk stratification capabilities.
• The findings support the clinical utility of WF T for deriving d_w in future applications.