β-Divergence Nonnegative Matrix Factorization on Biomedical Blind Source Separation

β-divergence has been studied for years, but it is yet to be discovered thoroughly. In this paper, we proposed the nonnegative matrix factorization (NMF) by using β-divergence in blind source separation (BSS) on biomedical field. The proposed idea is basically aimed at the separation of normal heart sound with normal lung sound. Temporal codes and spectral basis were modelled into a separated source, which is applied to the synthesis and real life data using multiplicative update rules. In the experiment, estimated and original source were compared to evaluate the performance of various source separation algorithms within a general framework, where the original sources and the noise that perturbed the mixture were included.

[1]  DeLiang Wang,et al.  Two-Microphone Separation of Speech Mixtures , 2008, IEEE Transactions on Neural Networks.

[2]  Raul Kompass,et al.  A Generalized Divergence Measure for Nonnegative Matrix Factorization , 2007, Neural Computation.

[3]  Alexey Ozerov,et al.  Multichannel Nonnegative Matrix Factorization in Convolutive Mixtures for Audio Source Separation , 2010, IEEE Transactions on Audio, Speech, and Language Processing.

[4]  M. Tavel,et al.  Spectral analysis of heart sounds: relationships between some physical characteristics and frequency spectra of first and second heart sounds in normals and hypertensives. , 1984, Journal of biomedical engineering.

[5]  V Chernick,et al.  Interference of cardiovascular sounds with phonopneumography in children. , 1985, The American review of respiratory disease.

[6]  W. L. Woo,et al.  Blind Source Separation of Postnonlinear Convolutive Mixture , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

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

[8]  Wai Lok Woo,et al.  Neural network approach to blind signal separation of mono-nonlinearly mixed sources , 2005, IEEE Transactions on Circuits and Systems I: Regular Papers.

[9]  DeLiang Wang,et al.  Monaural speech segregation based on pitch tracking and amplitude modulation , 2002, IEEE Transactions on Neural Networks.

[10]  A. M. Darsono,et al.  Blind Audio Source Separation with Sparse Nonnegative Matrix Factorization , 2014 .

[11]  Wai Lok Woo,et al.  Nonlinear single channel source separation , 2010, 2010 7th International Symposium on Communication Systems, Networks & Digital Signal Processing (CSNDSP 2010).

[12]  A. M. Darsono,et al.  Unsupervised Single Channel Source Separation with Nonnegative Matrix Factorization , 2015, ICIT 2015.

[13]  H. Pasterkamp,et al.  Respiratory sounds. Advances beyond the stethoscope. , 1997, American journal of respiratory and critical care medicine.

[14]  H. Sebastian Seung,et al.  Unsupervised Learning by Convex and Conic Coding , 1996, NIPS.

[15]  Andrzej Cichocki,et al.  Adaptive blind signal and image processing , 2002 .

[16]  Bo Xu,et al.  Monaural Speech Separation Based on Computational Auditory Scene Analysis and Objective Quality Assessment of Speech , 2006, IEEE Trans. Speech Audio Process..

[17]  A. M. Darsono,et al.  β-divergence two-dimensional sparse nonnegative matrix factorization for audio source separation , 2013, 2013 IEEE Conference on Wireless Sensor (ICWISE).

[18]  W. L. Woo,et al.  Single-Channel Source Separation Using EMD-Subband Variable Regularized Sparse Features , 2011, IEEE Transactions on Audio, Speech, and Language Processing.

[19]  Ioannis Pitas,et al.  Nonnegative Matrix Factorization in Polynomial Feature Space , 2008, IEEE Transactions on Neural Networks.