Using statistical distances to detect changes in the normal behavior of ECG-Holter signals

Julio Cesar Bastos de Figueiredo; Sergio Shiguemi Furuie

doi:10.1117/12.428236

30 May 2001 Using statistical distances to detect changes in the normal behavior of ECG-Holter signals

Julio Cesar Bastos de Figueiredo, Sergio Shiguemi Furuie

Author Affiliations +

Proceedings Volume 4325, Medical Imaging 2001: Ultrasonic Imaging and Signal Processing; (2001) https://doi.org/10.1117/12.428236
Event: Medical Imaging 2001, 2001, San Diego, CA, United States

Abstract

One of the main problems in the study of complex systems is to define a good metric that can distinguish between different dynamical behaviors in a nonlinear system. In this work we describe a method to detect different types of behaviors in a long term ECG-Holter using short portions of the Holter signal. This method is based on the calculation of the statistical distance between two distributions in a phase-space of a dynamical system. A short portion of an ECG-Holter signal with normal behavior is used to reconstruct the trajectory of an attractor in low dimensional phase-space. The points in this trajectory are interpreted as statistical distributions in the phase-space and assumed to represent the normal dynamical behavior of the ECG recording in this space. A fast algorithm is then used to compute the statistical distance between this attractor and all other attractors that are built using a sliding temporal window over the signal. For normal cases the distance stayed almost constant and below a threshold. For cases with abnormal transients, on the abnormal portion of ECG, the distance increased consistently with morphological changes.

Citation Download Citation

Julio Cesar Bastos de Figueiredo and Sergio Shiguemi Furuie "Using statistical distances to detect changes in the normal behavior of ECG-Holter signals", Proc. SPIE 4325, Medical Imaging 2001: Ultrasonic Imaging and Signal Processing, (30 May 2001); https://doi.org/10.1117/12.428236

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE