An improved spectral method of detecting and quantifying T-wave alternans for SCD risk evaluation

Abstract

T-wave alternans (TWA) is widely understood as an important indicator and a predictor of risk of sudden cardiac death (SCD). This research provides an improved spectral method of detecting and quantifying T-Wave alternans. Basically, the algorithm used enhanced TWA spectrum and T-Slope variations to generate two parameters: enhanced alternans ratio (EAR) and range index (RI). Then, the singular value decomposition method was applied to fuse two parameters. The results showed that our method has 82.4% accuracy on MITSCD database and 76.7% accuracy on normal healthy individuals. Moreover, the method achieved the 0.636 final score on the official PhysioNet Challenge 2008 dataset.