Session S63.4

Stability Analysis of the 12-Lead ECG Morphology in Different Physiological Conditions: Of Interest for Biometric Applications

F Porée, JY Bansard, G Kervio, G Carrault*

Université de Rennes
Rennes, France

Studies concerning biometry based on human ECG signals have been recently proposed in the literature. Most of them have as an objective, the proposal of a decision systems based on the extraction of a set of features from the ECG. Such approaches require the detection of fiducial points, which is generally difficult. In order to overcome such difficulties, recent approaches perform pattern recognition by comparing shapes. Most of these works hypothesize the stability of ECG in supine rest conditions and few of them consider the long-term stability. The objective of this paper is to evaluate the robustness of biometric approaches based on ECG recordings and two questions are addressed: 1) Is it required to record the ECG in supine rest as it is classically performed? 2) Is it necessary to compare only shapes recorded in the same condition?
In this study, recordings of a 12-lead ECG, from 14 normal subjects, in three experimental conditions (supine rest, standing and exercise), have been carried out and repeated 4 times. The correlation coefficient, computed on various interval lengths, is used for shape comparison and quality of the separation between genuine and impostors is evaluated with the well-known (in biometry) Discrimination Coefficient (DC) that must be greater than one. Results show that comparing shapes recorded in the same condition, leads to similar values of DC (with 1.2 (Abstract Control Number: 237)