K-Nearest Neighborhood Approach to Identify Level of Left Ventricular Ejection Fraction From Phonocardiogram

Phonocardiographic data with different left ventricular ejection fraction (LVEFs) levels (5 subjects with reduced LVEF and 3 normal subjects) were acquired from a specialized cardiac hospital. The data were preprocessed, and 50 epochs of complete cardiac cycles were extracted from each subject for frequency spectrum analysis and were classified through the K-nearest neighborhood approach. Two cases of classification were carried out: (1) 3 classes, 2 levels of LVEF (60%-65% and 50%) and normal, and (2) 4 classes, 3 levels of LVEF (63%-65%, 60%, and 50%) and normal. The K-nearest neighborhood classifier presents good classification accuracy with these data (93.5% for 3 classes and 85.5% for 4 classes).

[1]  Yongwan Park,et al.  Separation of Heart Sound Signal from Noise in Joint Cycle Frequency–Time–Frequency Domains Based on Fuzzy Detection , 2010, IEEE Transactions on Biomedical Engineering.

[2]  J. Liszka-Hackzell.  Categorization of Fetal Heart Rate Patterns Using Neural Networks , 2001, Journal of Medical Systems.

[3]  S Ari,et al.  DSP implementation of a heart valve disorder detection system from a phonocardiogram signal , 2008, Journal of medical engineering & technology.

[4]  Rakesh Kumar Sinha,et al.  Backpropagation Artificial Neural Network Classifier to Detect Changes in Heart Sound due to Mitral Valve Regurgitation , 2007, Journal of Medical Systems.

[5]  Andreas Voss,et al.  Diagnosing Aortic Valve Stenosis by Parameter Extraction of Heart Sound Signals , 2005, Annals of Biomedical Engineering.

[6]  G. Cloutier,et al.  Comparison of pattern recognition methods for computer-assisted classification of spectra of heart sounds in patients with a porcine bioprosthetic valve implanted in the mitral position , 1990, IEEE Transactions on Biomedical Engineering.

[7]  V Nigam,et al.  A simplicity-based fuzzy clustering approach for detection and extraction of murmurs from the phonocardiogram , 2008, Physiological measurement.

[8]  C L Feldman,et al.  Assessment of Left Ventricular Ejection Fraction and Volumes by Real-time, Two-dimensional Echocardiography: A Comparison of Cineangiographic and Radionuclide Techniques , 1979, Circulation.

[9]  Efendi N. Nasibov,et al.  Efficiency analysis of KNN and minimum distance-based classifiers in enzyme family prediction , 2009, Comput. Biol. Chem..

[10]  J.T.E. McDonnell,et al.  Time-frequency and time-scale techniques for the classification of native and bioprosthetic heart valve sounds , 1998, IEEE Transactions on Biomedical Engineering.

[11]  J M Moody,et al.  Bedside cardiac examination: constancy in a sea of change. , 2000, Current problems in cardiology.

[12]  R M Rangayyan,et al.  Phonocardiogram signal analysis: a review. , 1987, Critical reviews in biomedical engineering.

[13]  Aubrey Leatham,et al.  Auscultation of the Heart and Phonocardiography , 1971, Journal of the Royal College of Physicians of London.

[14]  Jiebo Luo,et al.  Data Mining. Multimedia, Soft Computing, and Bioinformatics , 2005, IEEE Transactions on Neural Networks.

[15]  D. Berman,et al.  Clinical assessment of left ventricular regional contraction patterns and ejection fraction by high-resolution gated scintigraphy. , 1975, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[16]  J. Hertzberg,et al.  Artificial Neural Network-Based Method of Screening Heart Murmurs in Children , 2001, Circulation.

[17]  Nancy E. Reed,et al.  Heart sound analysis for symptom detection and computer-aided diagnosis , 2004, Simul. Model. Pract. Theory.