Detecting and Quantifying T Wave Alternans Using Linear Least Squares Curve Fitting Method

Xiang Kui Wan; Kang Hui Yan; Ming Gui Li

doi:10.4028/www.scientific.net/AMM.239-240.1404

Paper Titles

A Research on Methane Prediction Model Based on Improved BP-GA Network
p.1382

Constrained All-k-Nearest-Neighbor Search
p.1387

Application of on-line Adaptive Weighted Fusion in Mine Gas Measurement System
p.1395

The Quantity Optimization of Integrated Testing Vehicle Based on Genetic Algorithm
p.1399

Detecting and Quantifying T Wave Alternans Using Linear Least Squares Curve Fitting Method
p.1404

Application Analysis of Compressive Sensing/Sampling Theory on Source Localization
p.1409

Time Series Prediction of Landslide Displacement Using SVM Model: Application to Baishuihe Landslide in Three Gorges Reservoir Area, China
p.1413

Simplified Design of Dual Quaternion Strapdown Inertial Navigation Integration Algorithms
p.1421

Research on Maintenance Resources Distribution Based on Queuing Theory
p.1428

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 239-240Detecting and Quantifying T Wave Alternans Using...

Detecting and Quantifying T Wave Alternans Using Linear Least Squares Curve Fitting Method

Abstract:

Evaluatement of T-wave alternans (TWA) is a non-invasive method to identify patients at risk for sudden cardiac death. A novel time-domain algorithm based on linear least squares method (LSM) for TWA analysis is developed in this paper. And the LSM’s ability to TWA analaysis in comparison with the modified moving average (MMA) method was demonstrated. In a simulation study LSM and MMA can both detected TWA with high accuracy, but under lower SNR levels LSM was found more robust, and its evaluated TWA values are closer to true TWA value. The algorithm was subsequently used to analyze the clinical ECGs from T-Wave Alternans Challenge Database, which showed that the results from this method vs. MMA had the correlation coefficients of 0.89.