The problem of ambulatory (Holter) Automatic Data Classification is faced by means of the "Morphometric Approach". This novel method easily allows to capture the essential shape information of the recorded signals and to reliably compare the results among different realizations of the same experiment (i.e. ECG recordings of different human subjects) on an objective basis. Furthermore, the automatic sensitivity analysis of the different ECG leads and optimal leads selection can be accomplished. Three different human subjects are screened (physiological condition at rest, pace-maker controlled, and atrial fibrillation). The experimental results show dramatic improvement over the previous approaches presented in scientific literature: outstanding potentialities for specific physiological and pathological subject behaviour characterization are envisioned.
The morphometric approach for effective automatic data classification
DACQUINO, GIANFRANCO;R. A. Fiorini
1994-01-01
Abstract
The problem of ambulatory (Holter) Automatic Data Classification is faced by means of the "Morphometric Approach". This novel method easily allows to capture the essential shape information of the recorded signals and to reliably compare the results among different realizations of the same experiment (i.e. ECG recordings of different human subjects) on an objective basis. Furthermore, the automatic sensitivity analysis of the different ECG leads and optimal leads selection can be accomplished. Three different human subjects are screened (physiological condition at rest, pace-maker controlled, and atrial fibrillation). The experimental results show dramatic improvement over the previous approaches presented in scientific literature: outstanding potentialities for specific physiological and pathological subject behaviour characterization are envisioned.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.