Quartiles and Mel Frequency Cepstral Coefficients vectors in Hidden Markov-Gaussian Mixture Models classification of merged heart sounds and lung sounds signals

This paper presents integrated Hidden Markov and Gaussian Mixture Models (HMM-GMM) to classify lung sounds (LS) and heart sounds (HS) characteristics. In order to optimize the models' size, several methodologies encompassing dendrograms, silhouettes and the Bayesian Information Criterion (BIC) were applied. The experiments were carried out extracting features from the LS and HS with MFCC (Mel-Frequency Cepstral Coefficients) vectors and Quantile vectors, specifically Quartiles. The merged HMM-GMM architecture for the signals using Quartiles, overall offered consistent classification results. In both types of vectors, a high degree of classification efficiency was obtained reaching up to 96% for the studied sets of signals. For MFCC the classification results were not conclusive. An assessment of the number of clusters using dendrograms, silhouettes, and BIC linked with the models' size. Consequently this allows to enhance efficiency of merged HMM-GMM models in diagnostic classification of cardiopulmonary acoustic signals.

[1]  C. Druzgalski,et al.  GMM y LDA aplicado a la detección de enfermedades pulmonares , 2013 .

[2]  Zahra Moussavi,et al.  Diagnostic potential in state space parameters of lung sounds , 2007, Medical & Biological Engineering & Computing.

[3]  C. Druzgalski,et al.  The HMM diagnostic models of respiratory sounds , 2014, 2014 Pan American Health Care Exchanges (PAHCE).

[4]  孙家广,et al.  Extracting Silhouette Curves of NURBS Surfaces by Tracing Silhouette Points , 1998 .

[5]  Andrzej Cichocki,et al.  A New Learning Algorithm for Blind Signal Separation , 1995, NIPS.

[6]  J. Vega,et al.  Modelos Acústicos HMM Multimodales para Sonidos Cardiacos y Pulmonares , 2014 .

[7]  R. Unbehauen,et al.  Estimation of image noise variance-Vision, Image and Signal Processing, IEE Proceedings- , 2004 .

[8]  Z. Moussavi.  Respiratory sound analysis [Introduction for the Special Issue] , 2007 .

[9]  Mayorga-Ortiz Pedro,et al.  Quantile Acoustic Vectors vs. MFCC Applied to Speaker Verification , 2014 .

[10]  Jianhua Zhao.  Efficient Model Selection for Mixtures of Probabilistic PCA Via Hierarchical BIC , 2014, IEEE Transactions on Cybernetics.

[11]  Younès Bennani,et al.  Dendogram based SVM for multi-class classification , 2006, 28th International Conference on Information Technology Interfaces, 2006..

[12]  Cepal. Comisión Económica para América Latina y el Caribe.  Panorama social de América Latina , 2012 .

[13]  Andrew R. Webb,et al.  Statistical Pattern Recognition , 1999 .

[14]  P. Mayorga,et al.  Expanded quantitative models for assessment of respiratory diseases and monitoring , 2011, 2011 Pan American Health Care Exchanges.

[15]  Ben P. Milner,et al.  Robust speech recognition over mobile and IP networks in burst-like packet loss , 2006, IEEE Transactions on Audio, Speech, and Language Processing.

[16]  Taeg Keun Whangbo,et al.  Vector Silhouette Extraction for Generating Blueprint , 2007, 2007 IEEE International Conference on Automation and Logistics.

[17]  Biing-Hwang Juang,et al.  Fundamentals of speech recognition , 1993, Prentice Hall signal processing series.

[18]  Sun Jiaguang,et al.  Extracting silhouette curves of NURBS surfaces by tracing silhouette points , 1998 .

[19]  Mohamed Chetouani,et al.  Advances in Nonlinear Speech Processing , 2011, Lecture Notes in Computer Science.

[20]  Jean-François Serignat,et al.  Audio packet loss over IP and speech recognition , 2003, 2003 IEEE Workshop on Automatic Speech Recognition and Understanding (IEEE Cat. No.03EX721).

[21]  Douglas A. Reynolds,et al.  A Gaussian mixture modeling approach to text-independent speaker identification , 1992 .

[22]  Jean-François Bonastre,et al.  Overview of compression and packet loss effects in speech biometrics , 2003 .

[23]  Ben P. Milner,et al.  Robust speech recognition over IP networks , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[24]  C. Druzgalski,et al.  Multimodal classification of heart sounds attributes , 2014, 2014 Pan American Health Care Exchanges (PAHCE).

[25]  Dan Istrate.  Détection et Reconnaissance des Sons pour la Surveillance Médicale. (Sound Detection and Classification for medical telemonitoring) , 2003 .

[26]  J. Wade Davis,et al.  Statistical Pattern Recognition , 2003, Technometrics.

[27]  P. Mayorga,et al.  Quantitative models for assessment of respiratory diseases , 2010, 2010 Pan American Health Care Exchanges.

[28]  Pedro Mayorga,et al.  Modified classification of normal Lung Sounds applying Quantile Vectors , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[29]  Sonia Charleston-Villalobos,et al.  Assessment of multichannel lung sounds parameterization for two-class classification in interstitial lung disease patients , 2011, Comput. Biol. Medicine.

[30]  Chafic Mokbel,et al.  Behavior of a Bayesian adaptation method for incremental enrollment in speaker verification , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).