[Quantifying the amplitude of microvolt T wave alternans with sparse principal component analysis]

Area of Science:

• Cardiology
• Biomedical Signal Processing
• Computational Medicine

Context:

• T wave alternans (TWA) in electrocardiograms (ECG) are critical indicators for sudden cardiac death risk.
• Accurate quantification of TWA amplitude is clinically essential but challenging due to signal non-stationarity.
• Existing methods for TWA analysis often struggle with precision and reliability.

Purpose:

• To develop and validate a novel method for improving the accuracy of T wave alternans amplitude quantification.
• To address the challenges posed by the non-stationary nature of TWA in ECG signals.
• To compare the proposed method against established TWA quantification techniques.

Summary:

• This study proposes a new method utilizing sparse principal component (SPC) analysis for T wave alternans quantification.
• T wave alternans are identified using principal component analysis (PCA) with statistical testing.
• The identified principal components are further processed with a sparse constraint to enhance amplitude quantification.

Impact:

• The proposed SPC-based method demonstrates superior performance in TWA amplitude quantification compared to four popular existing solutions.
• Results, based on an international benchmark database, show improved rank correlation coefficient and reduced relative error.
• This advancement offers a more reliable tool for clinical risk assessment in cardiology.