On the Blind Recovery of Cardiac and Respiratory Sounds

We present a method for smart auscultation by proposing a novel blind recovery of the original cardiac and respiratory sounds from a single observation mixture, in the framework of nonnegative matrix factorization (NMF). The method learns the basis spectra of the mixing sources in unsupervised or semisupervised fashion depending upon the applications. A modified NMF technique is proposed, which enforces the spectral structure of the target sources in mixture factorization, resulting in good separation of target sources, even in the presence of nonstationary noise. Moreover, data is processed in small batches which 1) enables dynamic bases spectra update technique to mitigate the spectral variations of the mixing sources, and 2) reduces computational complexity. The analytical work is verified through simulations using synthetic as well as actual clinical data collected from different subjects in different clinical sittings. The proposed smart auscultation method demonstrates excellent results even in noisy clinical environments.

[1]  Z. Moussavi,et al.  Heart Sounds Separation From Lung Sounds Using Independent Component Analysis , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[2]  G. R. Sinha,et al.  Separation of Lung Sound from PCG Signals Using , 2012 .

[3]  January Gnitecki,et al.  Separating heart sounds from lung sounds. Accurate diagnosis of respiratory disease depends on understanding noises. , 2007, IEEE engineering in medicine and biology magazine : the quarterly magazine of the Engineering in Medicine & Biology Society.

[4]  Hamid Reza Mohseni,et al.  Localizing Heart Sounds in Respiratory Signals Using Singular Spectrum Analysis , 2011, IEEE Transactions on Biomedical Engineering.

[5]  M.G. Christensen,et al.  Structured non-negative matrix factorization with sparsity patterns , 2008, 2008 42nd Asilomar Conference on Signals, Systems and Computers.

[6]  ChingShun Lin,et al.  Blind source separation of heart and lung sounds based on nonnegative matrix factorization , 2013, 2013 International Symposium on Intelligent Signal Processing and Communication Systems.

[7]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[8]  M Nissan,et al.  Spectral characteristics of chest wall breath sounds in normal subjects. , 1995, Thorax.

[9]  Tuomas Virtanen,et al.  Monaural Sound Source Separation by Nonnegative Matrix Factorization With Temporal Continuity and Sparseness Criteria , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[10]  Mark D. Plumbley,et al.  Separating sources from sequentially acquired mixtures of heart signals , 2011, 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[11]  Constantinos B. Papadias,et al.  Blind recovery of cardiac and respiratory sounds using non-negative matrix factorization & time-frequency masking , 2013, 13th IEEE International Conference on BioInformatics and BioEngineering.

[12]  Ivan W. Selesnick,et al.  Translation-invariant shrinkage/thresholding of group sparse signals , 2013, Signal Process..

[13]  T. Rice.  Learning to Listen: Auscultation and the Transmission of Auditory Knowledge , 2010 .

[14]  Ting Li,et al.  Heart sound cancellation from lung sound record using cyclostationarity. , 2013, Medical engineering & physics.

[15]  Mark D. Plumbley,et al.  INVESTIGATING SINGLE-CHANNEL AUDIO SOURCE SEPARATION METHODS BASED ON NON-NEGATIVE MATRIX FACTORIZATION , 2006 .

[16]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[17]  Aimin Zhou,et al.  Non-negative Matrix Factorization: A Short Survey on Methods and Applications , 2012, ISICA.

[18]  Constantinos B. Papadias,et al.  Separation of cardiorespiratory sounds using time-frequency masking and sparsity , 2013, 2013 18th International Conference on Digital Signal Processing (DSP).

[19]  Yu-Jin Zhang,et al.  Nonnegative Matrix Factorization: A Comprehensive Review , 2013, IEEE Transactions on Knowledge and Data Engineering.

[20]  T. E. Ayoob Khan,et al.  Separating Heart Sound from Lung Sound Using LabVIEW , 2010 .

[21]  L.J. Hadjileontiadis,et al.  Adaptive reduction of heart sounds from lung sounds using fourth-order statistics , 1997, IEEE Transactions on Biomedical Engineering.

[22]  M.R. Azimi-Sadjadi,et al.  Reduced order Kalman filtering for the enhancement of respiratory sounds , 1996, IEEE Transactions on Biomedical Engineering.

[23]  Jouni Paulus,et al.  Drum transcription with non-negative spectrogram factorisation , 2005, 2005 13th European Signal Processing Conference.

[24]  Minje Kim,et al.  Nonnegative Matrix Partial Co-Factorization for Spectral and Temporal Drum Source Separation , 2011, IEEE Journal of Selected Topics in Signal Processing.

[25]  Saeid Sanei,et al.  A new time domain convolutive BSS of heart and lung sounds , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).