HSAS: Heart Sound Authentication System

In this paper, we study the probability of using heart sound as a biometric for human authentication. The most significant contribution of using heart sound as a biometric is that it cannot be easily replicated as compared to other conventional biometrics. The proposed Heart Sound Authentication System (HSAS) consists of five main phases, namely, Heart Sound Capturing, Pre-processing, Feature Extraction, Training, and Classification and Authentication phases. The proposed biometric system comprises a digital electronic stethoscope, a computer equipped with a sound card and heart sound capturing software application. In this work, two classifiers were used, which are Mean Square Error (MSE) and K-Nearest Neighbor (KNN). Results indicated that the proposed system has attained recall 82.4% and precision 80.7% for MSE classifier and has attained recall 94.5% and precision 93% for KNN classifier for a database of 400 heart sounds that were recorded from 40 participants by 10 heart sound recordings for each participant.

[1]  W. Marsden.  I and J , 2012 .

[2]  Tran Huy Dat,et al.  Heart sound as a biometric , 2008, Pattern Recognit..

[3]  Tran Huy Dat,et al.  Human identification using heart sound , 2006 .

[4]  George Tzanetakis,et al.  Audio Analysis using the Discrete Wavelet Transform , 2001 .

[5]  S. Mallat.  A wavelet tour of signal processing , 1998 .

[6]  P. Laguna,et al.  Signal Processing , 2002, Yearbook of Medical Informatics.

[7]  M.G. Bellanger,et al.  Digital processing of speech signals , 1980, Proceedings of the IEEE.

[8]  Tohru Takagi,et al.  A method for pitch extraction of speech signals using autocorrelation functions through multiple window lengths , 2000 .

[9]  Michel Vacher,et al.  Sound detection and classification through transient models usingwavelet coefficient trees , 2004, 2004 12th European Signal Processing Conference.

[10]  P.H. King.  Noninvasive instrumentation and measurement medical diagnosis , 2002, IEEE Engineering in Medicine and Biology Magazine.

[11]  Thierry Dutoit,et al.  Complex cepstrum-based decomposition of speech for glottal source estimation , 2009, INTERSPEECH.

[12]  Francesco Beritelli,et al.  Biometric Identification Based on Frequency Analysis of Cardiac Sounds , 2007, IEEE Transactions on Information Forensics and Security.

[13]  William Francis Ganong,et al.  Review of Medical Physiology , 1969 .

[14]  A C Dornhorst,et al.  Review of Medical Physiology. , 1966 .

[15]  D. Reynolds,et al.  Authentication gets personal with biometrics , 2004, IEEE Signal Processing Magazine.

[16]  Robert B. Northrop.  Noninvasive Instrumentation and Measurement in Medical Diagnosis , 2001 .