Corrected Integral Shape Averaging Applied to Obstructive Sleep Apnea Detection from the Electrocardiogram
- PDF / 1,151,886 Bytes
- 12 Pages / 600.03 x 792 pts Page_size
- 76 Downloads / 171 Views
Research Article Corrected Integral Shape Averaging Applied to Obstructive Sleep Apnea Detection from the Electrocardiogram S. Boudaoud,1 H. Rix,1 O. Meste,1 C. Heneghan,2 and C. O’Brien2 1 Laboratoire 2 School
d’Informatique, Signaux et Syst`emes de Sophia Antipolis (I3S), UMR 6070 CNRS, 06903 Sophia Antipolis, France of Electrical, Electronic and Mechanical Engineering, University College Dublin, Belfield, Dublin 4, Ireland
Received 30 April 2006; Revised 31 October 2006; Accepted 1 November 2006 Recommended by William Allan Sandham We present a technique called corrected integral shape averaging (CISA) for quantifying shape and shape differences in a set of signals. CISA can be used to account for signal differences which are purely due to affine time warping (jitter and dilation/compression), and hence provide access to intrinsic shape fluctuations. CISA can also be used to define a distance between shapes which has useful mathematical properties; a mean shape signal for a set of signals can be defined, which minimizes the sum of squared shape distances of the set from the mean. The CISA procedure also allows joint estimation of the affine time parameters. Numerical simulations are presented to support the algorithm for obtaining the CISA mean and parameters. Since CISA provides a well-defined shape distance, it can be used in shape clustering applications based on distance measures such as k-means. We present an application in which CISA shape clustering is applied to P-waves extracted from the electrocardiogram of subjects suffering from sleep apnea. The resulting shape clustering distinguishes ECG segments recorded during apnea from those recorded during normal breathing with a sensitivity of 81% and specificity of 84%. Copyright © 2007 S. Boudaoud et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
1.
INTRODUCTION
In the field of electrocardiogram (ECG) analysis, signal shape (or morphology) analysis is often used intuitively by clinicians for tasks such as beat typing and ischemia detection. However, there are relatively few well-defined analytical tools for automatically quantifying shape and shape differences, which can be used to capture the intuition of clinical practitioners, or to systematically uncover subtle but clinically significant changes in shape. One approach to signal shape quantification assumes a common underlying shape or template which is only subject to time warping [1, 2], and attempts to calculate both the time-warping function and a common template under a restrictive shape hypothesis. However, in practice, variations in shape from an underlying common shape are often the main objective of interest, and quantifying such shape variability is a goal of this work. In order to assess deviation from an underlying “template” shape, we first need to define the concept of an “averaged shape” reference signal. This signal should
Data Loading...
