Biometric Identification based on Frequency Analysis of Cardiac Sounds

A multiband approach to human identity verification based on PhonoCardioGram signal analysis

“…Interval between recordings was between 3 and 5 weeks in the 90 % of cases. Since the recordings from the aortic and pulmonic regions exhibit a greater degree of separability between the classes [1], all the PCG sequences considered in this work relate to the pulmonary valve. Each PCG sequence lasts about 10 seconds, so as to have at least 10 cardiac cycles.…”

“…As human verification is performed by analyzing the frequency characteristics of a set of aggregated sub-bands of the sounds S1 and S2, it was necessary to implement a mechanism to identify the bounds of these sounds, in terms of samples, in an audio trace from cardiac auscultation (comprising six cardiac cycles). The details of the S1/S2 detection algorithm are illustrated in [1]. Having determined the time intervals of the S1 and S2 signals, the signals are transformed from the time to the frequency domain [7], [8].…”

“…For this reason, other physiological or behavioural characteristics have recently been taken into consideration. In [1] it has been shown, by means of a preliminary study, that PhonoCardioGram (PCG) signals have specific individual characteristics that can be taken into consideration as a physiological sign used in a biometric system. More recently some other papers have proposed heart sound as biometric [4,] emphasizing the advantages in comparison with ECG signal [5].…”

HSAS: Heart Sound Authentication System

“…The mechanisms that generate cardiac sounds are complex. In particular, the acoustic phenomena include brief vibrations caused by closure of the valves and tensing of the cardiac muscle and other vibrations caused by turbulence in the blood flow through narrow cardiac valves [5].…”

“…These applications include the possibility of estimating the heart rate, automatically detecting the presence of certain pathologies (arrhythmia, systolic murmur, mitral stenosis, etc.) and also of using the intrinsic characteristics of heart sounds for biometric identification purposes [3], [5].…”