An automatic method for arterial pulse waveform recognition using KNN and SVM classifiers

AbstractThe measurement and analysis of the arterial pulse waveform (APW) are the means for cardiovascular risk assessment. Optical sensors represent an attractive instrumental solution to APW assessment due to their truly non-contact nature that makes the measurement of the skin surface displacement possible, especially at the carotid artery site. In this work, an automatic method to extract and classify the acquired data of APW signals and noise segments was proposed. Two classifiers were implemented: k-nearest neighbours and support vector machine (SVM), and a comparative study was made, considering widely used performance metrics. This work represents a wide study in feature creation for APW. A pool of 37 features was extracted and split in different subsets: amplitude features, time domain statistics, wavelet features, cross-correlation features and frequency domain statistics. The support vector machine recursive feature elimination was implemented for feature selection in order to identify the most relevant feature. The best result (0.952 accuracy) in discrimination between signals and noise was obtained for the SVM classifier with an optimal feature subset .

[1]  Xiao Hu,et al.  Regression analysis for peak designation in pulsatile pressure signals , 2009, Medical & Biological Engineering & Computing.

[2]  João Cardoso,et al.  Optical Methods for Local Pulse Wave Velocity Assessment , 2011, BIOSIGNALS.

[3]  Vojislav Kecman,et al.  Gene extraction for cancer diagnosis by support vector machines - An improvement , 2005, Artif. Intell. Medicine.

[4]  Lu Wang,et al.  SVM Classification for Discriminating Cardiovascular Disease Patients from Non-cardiovascular Disease Controls Using Pulse Waveform Variability Analysis , 2004, Australian Conference on Artificial Intelligence.

[5]  Hong Yan,et al.  A Machine Learning Approach to Improve Contactless Heart Rate Monitoring Using a Webcam , 2014, IEEE Journal of Biomedical and Health Informatics.

[6]  Laura. Mason Laura,et al.  Signal processing methods for non-invasive respiration monitoring , 2002 .

[7]  M. Crilly,et al.  Indices of cardiovascular function derived from peripheral pulse wave analysis using radial applanation tonometry: a measurement repeatability study , 2007, Vascular medicine.

[8]  Pablo F. Diez,et al.  Application of the empirical mode decomposition to the extraction of features from EEG signals for mental task classification , 2009, 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[9]  H. Struijker‐Boudier,et al.  Expert consensus document on arterial stiffness: methodological issues and clinical applications. , 2006, European heart journal.

[10]  David Ward,et al.  Comparison of statistical methods for classification of ovarian cancer using mass spectrometry data , 2003, Bioinform..

[11]  M Congedo,et al.  A review of classification algorithms for EEG-based brain–computer interfaces , 2007, Journal of neural engineering.

[12]  Vânia Almeida,et al.  Characterization of Optical System for Hemodynamic Multi-Parameter Assessment , 2013 .

[13]  Patrick Segers,et al.  Blood pressure waveform analysis by means of wavelet transform , 2009, Medical & Biological Engineering & Computing.

[14]  Richard Weber,et al.  A wrapper method for feature selection using Support Vector Machines , 2009, Inf. Sci..

[15]  Jon Rigelsford,et al.  Pattern Recognition: Concepts, Methods and Applications , 2002 .

[16]  R. Mark,et al.  An investigation of patterns in hemodynamic data indicative of impending hypotension in intensive care , 2010, Biomedical engineering online.

[17]  A. Dart,et al.  Pulse pressure--a review of mechanisms and clinical relevance. , 2001, Journal of the American College of Cardiology.

[18]  Kanak Saxena,et al.  Performance Evaluation of SVM and K-Nearest Neighbor Algorithm over Medical Data set , 2012 .

[19]  Roberto Hornero,et al.  Multivariate Analysis of Blood Oxygen Saturation Recordings in Obstructive Sleep Apnea Diagnosis , 2010, IEEE Transactions on Biomedical Engineering.

[20]  Jun Tian,et al.  A Gene Selection Method for Cancer Classification , 2012, Comput. Math. Methods Medicine.

[21]  Xiao Hu,et al.  Robust Peak Recognition in Intracranial Pressure Signals , 2010, Biomedical engineering online.

[22]  Yasmin,et al.  Determinants of pulse wave velocity in healthy people and in the presence of cardiovascular risk factors: ‘establishing normal and reference values’ , 2010, European heart journal.

[23]  Stephen R. Alty,et al.  Predicting Arterial Stiffness From the Digital Volume Pulse Waveform , 2007, IEEE Transactions on Biomedical Engineering.

[24]  C. Correia,et al.  Comparison of Low-Cost and Noninvasive Optical Sensors for Cardiovascular Monitoring , 2013, IEEE Sensors Journal.

[25]  J. Victor Marcos,et al.  Feature selection from nocturnal oximetry using genetic algorithms to assist in obstructive sleep apnoea diagnosis. , 2012, Medical engineering & physics.

[26]  G.Subramanya Nayak,et al.  Classification of Bio Optical signals using K- Means Clustering for Detection of Skin Pathology , 2010 .

[27]  Conrad Sanderson,et al.  An Efficient Alternative to SVM Based Recursive Feature Elimination with Applications in Natural Language Processing and Bioinformatics , 2006, Australian Conference on Artificial Intelligence.

[28]  João Cardoso,et al.  Pulse pressure waveform estimation using distension profiling with contactless optical probe. , 2014, Medical engineering & physics.

[29]  K. Kawecka-Jaszcz,et al.  Comparison of aortic pulse wave velocity measured by three techniques: Complior, SphygmoCor and Arteriograph , 2008, Journal of hypertension.

[30]  Alan Murray,et al.  Comparative reproducibility of dermal microvascular blood flow changes in response to acetylcholine iontophoresis, hyperthermia and reactive hyperaemia , 2010, Physiological measurement.

[31]  Thomas Weber,et al.  Arterial Stiffness, Wave Reflections, and the Risk of Coronary Artery Disease , 2004, Circulation.

[32]  Hui-yan Wang,et al.  A model for automatic identification of human pulse signals , 2008 .

[33]  Vânia Almeida,et al.  Signal Analysis in a New Optical Pulse Waveform Profiler for Cardiovascular Applications , 2011 .

[34]  Mark Butlin,et al.  Arterial blood pressure measurement and pulse wave analysis—their role in enhancing cardiovascular assessment , 2010, Physiological measurement.

[35]  João Cardoso,et al.  Non-contact Pulse Wave Velocity Assessment Using Optical Methods , 2011, BIOSTEC.

[36]  Boualem Boashash,et al.  Automated detection of perinatal hypoxia using time–frequency-based heart rate variability features , 2013, Medical & Biological Engineering & Computing.

[37]  B. Gaszner,et al.  Invasive validation of a new oscillometric device (Arteriograph) for measuring augmentation index, central blood pressure and aortic pulse wave velocity , 2010, Journal of hypertension.

[38]  Thakker Bhaskar,et al.  Support Vector Machine for Abnormal Pulse Classification , 2011 .

[39]  J. Blacher,et al.  Aortic pulse wave velocity as a marker of cardiovascular risk in hypertensive patients. , 1999, Hypertension.

[40]  E. P. Tomasini,et al.  A preliminary study for the evaluation of large artery stiffness: a non contact approach , 2008 .

[41]  João Cardoso,et al.  Local PWV and Other Hemodynamic Parameters Assessment - Validation of a New Optical Technique in an Healthy Population , 2013, BIOSIGNALS.

[42]  Stephen R. Alty,et al.  Classification of Aortic Stiffness From Eigendecomposition of the Digital Volume Pulse Waveform , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[43]  M. Elgendi.  On the Analysis of Fingertip Photoplethysmogram Signals , 2012, Current cardiology reviews.

[44]  M. Bhuyan,et al.  The Wavelet Transform Based Arterial Blood Pressure Waveform Delineator , 2012 .

[45]  Javier Martinez,et al.  Machine Learning Techniques as a Helpful Tool Toward Determination of Plaque Vulnerability , 2012, IEEE Transactions on Biomedical Engineering.

[46]  P. Boutouyrie,et al.  Assessment of pulse wave velocity , 2008 .

[47]  R. Eckmiller,et al.  Predicting the nonlinear dynamics of biological neurons using support vector machines with different kernels , 2001, IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222).

[48]  D. Duprez,et al.  Noninvasive measurements of arterial stiffness: Repeatability and interrelationships with endothelial function and arterial morphology measures , 2007, Vascular health and risk management.

[49]  Kyung-Ah Kim,et al.  Mortality prediction of rats in acute hemorrhagic shock using machine learning techniques , 2013, Medical & Biological Engineering & Computing.

[50]  F. Pessana,et al.  Arterial pressure and diameter waveforms analysis by means of wavelet transform: Application to artery de-endothelization , 2010, 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology.

[51]  Yi Liu,et al.  FS_SFS: A novel feature selection method for support vector machines , 2006, Pattern Recognit..

[52]  Francesco Faita,et al.  Arterial pressure changes monitoring with a new precordial noninvasive sensor , 2008, Cardiovascular ultrasound.

[53]  João Cardoso,et al.  Reproducibility of Pulse Wave Analysis and Pulse Wave Velocity in Healthy Subjects , 2014, BIOSIGNALS.

[54]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[55]  G.Subramanya Nayak,et al.  Classification of ECG Signals using ANN with Resilient Back Propagation Algorithm , 2012 .

[56]  Patrick Segers,et al.  The use of diameter distension waveforms as an alternative for tonometric pressure to assess carotid blood pressure , 2010, Physiological measurement.

[57]  Leopoldo C. Cancio,et al.  Development and validation of a machine learning algorithm and hybrid system to predict the need for life-saving interventions in trauma patients , 2013, Medical & Biological Engineering & Computing.