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

详细信息    查看全文

• 作者：Tânia Pereira ; Joana S. Paiva…
• 关键词：Arterial pulse waveform ; Optical system ; Feature creation ; Recursive feature elimination ; K ; nearest neighbour algorithm ; Support vector machine
• 刊名：Medical and Biological Engineering and Computing
• 出版年：2016
• 出版时间：July 2016
• 年：2016
• 卷：54
• 期：7
• 页码：1049-1059
• 全文大小：1,787 KB
• 参考文献：1.Alty SR, Angarita-jaimes N, Millasseau SC, Chowienczyk PJ (2007) Predicting arterial stiffness from the digital volume pulse waveform. Biomed Eng IEEE Trans 54(12):2268–2275CrossRef
  2.Alvarez D, Member S, Hornero R (2010) Multivariate analysis of blood oxygen saturation recordings in obstructive sleep apnea diagnosis. Biomed Eng IEEE Trans 57(12):2816–2824CrossRef
  3.Álvarez D, Hornero R, Marcos JV, Del Campo F (2012) Feature selection from nocturnal oximetry using genetic algorithms to assist in obstructive sleep apnoea diagnosis. Med Eng Phys 34(8):1049–1057CrossRef PubMed
  4.Angarita-jaimes N, Alty SR, Millasseau SC, Chowienczyk PJ (2006) Classification of aortic stiffness from eigendecomposition of the digital volume pulse waveform. In: 2006 IEEE international conference on acoustics, speech and signal processing, 2006. ICASSP 2006 proceedings, pp 1168–1171
  5.Avolio AP, Butlin M, Walsh A (2010) Arterial blood pressure measurement and pulse wave analysis—their role in enhancing cardiovascular assessment. Physiol Meas 31(1):1–47CrossRef
  6.Bedo J, Sanderson C, Kowalczyk A (2006) An efficient alternative to svm based recursive feature elimination with applications bioinformatics. In: Sattar A, Kang B-H (eds) AI 2006: advances in artificial intelligence. Springer, Berlin, Heidelberg, pp 170–180CrossRef
  7.Blacher J, Asmar R, Djane S, London GM, Safar ME (1999) Aortic pulse wave velocity as a marker of cardiovascular risk in hypertensive patients. Hypertension 33(5):1111–1117CrossRef PubMed
  8.Bombardini T, Gemignani V, Bianchini E, Venneri L, Petersen C, Pasanisi E, Pratali L, Pianelli M, Faita F, Giannoni M, Arpesella G, Picano E (2008) Arterial pressure changes monitoring with a new precordial noninvasive sensor. Cardiovasc Ultrasound 6:41CrossRef PubMed PubMedCentral
  9.Boutouyrie P, Briet M, Collin C, Vermeersch S, Pannier B (2009) Assessment of pulse wave velocity. Artery Res 3(1):3–8CrossRef
  10.Cilla M, Martinez J, Pena E, Martínez MA (2012) Machine learning techniques as a helpful tool toward determination of plaque vulnerability. Biomed Eng IEEE Trans 59(4):1155–1161CrossRef
  11.Crilly M, Coch C, Bruce M, Clark H, Williams D (2007) Indices of cardiovascular function derived from peripheral pulse wave analysis using radial applanation tonometry: a measurement repeatability study. Vasc Med 12(3):189–197CrossRef PubMed
  12.Dart AM, Kingwell BA (2001) Pulse pressure—a review of mechanisms and clinical relevance. J Am Coll Cardiol 37(4):975–984CrossRef PubMed
  13.De Melis M, Morbiducci U, Scalise L, Tomasini EP, Delbeke D, Baets R, Van Bortel LM, Segers P (2008) A preliminary study for the evaluation of large artery stiffness: a non contact approach. Artery Res 2(3):100–101CrossRef
  14.De Melis M, Morbiducci U, Rietzschel ER, De Buyzere M, Qasem A, Van Bortel L, Claessens T, Montevecchi FM, Avolio A, Segers P (2009) Blood pressure waveform analysis by means of wavelet transform. Med Biol Eng Comput 47(2):165–173CrossRef PubMed
  15.Diez PF, Mut V, Laciar E, Torres A, Avila E (2009) Application of the empirical mode decomposition to the extraction of features from EEG signals for mental task classification. In: Conference on proceedings of the IEEE engineering in medicine and biology society, vol 2009, pp 2579–2582
  16.Dong S, Boashash B, Azemi G, Lingwood BE, Colditz PB (2014) Automated detection of perinatal hypoxia using time–frequency-based heart rate variability features. Med Biol Eng Comput 52(2):183–191CrossRef PubMed
  17.Elgendi M (2012) On the analysis of fingertip photoplethysmogram signals. Curr Cardiol Rev 8(1):14–25CrossRef PubMed PubMedCentral
  18.Frontzek T, Lal TN, Eckmiller R, Bonn D, Germany FR (2001) Predicting the nonlinear dynamics of biological neurons using support vector machines with di erent kernels. In: International joint conference on neural networks, 2001. proceedings. IJCNN’01, vol 2. pp 1492–1497
  19.He W, Li S, Xiao H, Yu C, Lin H (2012) An arterial elasticity index algorithm based on wavelet transform and curve fitting. J Inf Comput Sci 9(12):3379–3389
  20.Horváth IG, Németh A, Lenkey Z, Alessandri N, Tufano F, Kis P, Gaszner B, Cziráki A (2010) Invasive validation of a new oscillometric device (Arteriograph) for measuring augmentation index, central blood pressure and aortic pulse wave velocity. J Hypertens 28(10):2068–2075CrossRef PubMed
  21.Huang TM, Kecman V (2005) Gene extraction for cancer diagnosis by support vector machines an improvement and comparison with nearest. Artif Intell Med 35(1–2):185–194CrossRef PubMed
  22.Huck CJ, Bronas UG, Williamson EB, Draheim CC, Duprez DA, Dengel DR (2007) Noninvasive measurements of arterial stiffness: repeatability and interrelationships with endothelial function and arterial morphology measures. Vasc Health Risk Manag 3(3):343–349PubMed PubMedCentral
  23.Janney JB, Sruthi SP (2012) Dicrotic notch detection and analysis of arterial pulse by using discrete wavelet. OSIET J Commun Electron 4:93
  24.Jason Weston FS, Elisseeff A, BakIr G The spider. http://​www.​kyb.​tuebingen.​mpg.​de/​bs/​people/​spider
  25.Kim K-A, Choi JY, Yoo TK, Kim SK, Chung KS, Kim DW (2013) Mortality prediction of rats in acute hemorrhagic shock using machine learning techniques. Med Biol Eng Comput 51(9):1059–1067CrossRef PubMed
  26.Kips J, Vanmolkot F, Mahieu D, Vermeersch S, Fabry I, de Hoon J, Van Bortel L, Segers P (2010) The use of diameter distension waveforms as an alternative for tonometric pressure to assess carotid blood pressure. Physiol Meas 31(4):543–553CrossRef PubMed
  27.Kohavi R (1995) A study of cross-validation and bootstrap for accuracy estimation and model selection. In: IJCAI’95 proceedings of the 14th international joint conference on artificial intelligence. pp 1137–1143
  28.Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D, Pannier B, Vlachopoulos C, Wilkinson I, Struijker-Boudier H (2006) Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J 27(21):2588–2605CrossRef PubMed
  29.Lee J, Mark RG (2010) An investigation of patterns in hemodynamic data indicative of impending hypotension in intensive care. Biomed Eng Online 9(1):62CrossRef PubMed PubMedCentral
  30.Liu Y, Zheng YF (2006) FS_SFS: A novel feature selection method for support vector machines. Pattern Recognit 39(7):1333–1345CrossRef
  31.Liu NT, Holcomb JB, Wade CE, Batchinsky AI, Cancio LC, Darrah MI, Salinas J (2014) Development and validation of a machine learning algorithm and hybrid system to predict the need for life-saving interventions in trauma patients. Med Biol Eng Comput 52(2):193–203CrossRef PubMed
  32.Lotte F, Congedo M, Lécuyer A, Lamarche F, Arnaldi B (2007) A review of classification algorithms for EEG-based brain-computer interfaces. J Neural Eng 4(2):R1–R13CrossRef PubMed
  33.Maldonado S, Weber R (2009) A wrapper method for feature selection using support vector machines. Inf Sci 179(13):2208–2217CrossRef
  34.Marques JP (2001) Pattern recognition: concepts, methods, and applications, 1st edn. Springer, Berlin, HeidelbergCrossRef
  35.Mason L (2002) Signal processing methods for non-invasive respiration monitoring. University of Oxford, Oxford
  36.Monkaresi H, Calvo RA, Yan H (2014) A machine learning approach to improve contactless heart rate monitoring using a webcam. Biomed Heal Inform IEEE J 18(4):2168–2194
  37.Nayak GS (2012) Classification of ECG signals using ANN with resilient back propagation algorithm. Int J Comput Appl 54(6):20–23
  38.Nayak GS, Davide O (2010) Classification of bio optical signals using k-means clustering for detection of skin pathology. Int J Comput Appl 1(2):92–96
  39.Pachauri A, Bhuyan M (2012) Wavelet transform based arterial blood pressure waveform delineator. Int J Biol Biomed Eng 6(1):16–25
  40.Pereira T, Oliveira T, Cabeleira M, Matos P, Pereira HC, Almeida V, Borges E, Santos H, Pereira T, Cardoso J, Correia C (2011) Signal analysis in a new optical pulse waveform profiler for cardiovascular applications. In: SIPA 2011—proceedings of the IASTED international conference on signal and image processing and applications, no. Sipa. pp 19–25
  41.Pereira T, Cabeleira M, Matos P, Borges E, Cardoso J, Correia C (2011) Optical methods for local pulse wave velocity assessment. In: BIOSIGNALS 2011—4th international conference on bio-inspired systems and signal processing. Rome, Italy, pp 74–81
  42.Pereira T, Cabeleira M, Matos P, Borges E, Almeida V, Pereira HC, Cardoso J, Correia C (2012) Non-contact pulse wave velocity assessment using optical methods. In: Fred A, Filipe J, Gamboa H (eds) Biomedical engineering systems and technologies, vol 273. Springer, Berlin, Heidelberg, pp 246–257. doi:10.​1007/​978-3-642-29752-6_​18 CrossRef
  43.Pereira T, Oliveira T, Cabeleira M, Pereira H, Almeida V, Cardoso J, Correia C (2012) Comparison of low-cost and non-invasive optical sensors for cardiovascular monitoring. IEEE Sens J 13(5):1434–1441. doi:10.​1109/​JSEN.​2012.​2236549 CrossRef
  44.Pereira T, Santos I, Oliveira T, Vaz P, Correia T, Pereira T, Santos H, Pereira H, Almeida V, Cardoso J, Correia C (2013) Characterization of optical system for hemodynamic multi-parameter assessment. Cardiovasc Eng Technol 4(1):87–97CrossRef
  45.Pereira T, Santos I, Oliveira T, Vaz P, Santos H, Pereira H, Almeida V, Cardoso J (2013) Local PWV and other hemodynamic parameters assessment: validation of a new optical technique in an healthy population. In: BIOSIGNALS 2013—6th international conference on bio-inspired systems and signal processing, vol 1. Barcelona, Spain, pp 61–69
  46.Pereira T, Santos I, Santos H, Almeida V, Pereira H, Correia C, Cardoso J (2014) Reproducibility of pulse wave analysis and pulse wave velocity in healthy subjects. In: BIOSIGNALS 2014—7th international conference on bio-inspired systems and signal processing. Angers, France, pp 221–228
  47.Pereira T, Santos I, Oliveira T, Vaz P, Pereira T, Santos H, Pereira H, Correia C, Cardoso J (2014) Pulse pressure waveform estimation using distension profiling with contactless optical probe. Med Eng Phys 36(11):1515–1520CrossRef PubMed
  48.Raikwal JS, Saxena K (2012) Performance evaluation of SVM and k-nearest neighbor algorithm over medical data set. Int J Comput Appl 50(14):35–39
  49.Rajzer MW, Wojciechowska W, Klocek M, Palka I, Brzozowska-Kiszka M, Kawecka-Jaszcz K (2008) Comparison of aortic pulse wave velocity measured by three techniques: Complior, SphygmoCor and Arteriograph. J Hypertens 26(10):2001–2007CrossRef PubMed
  50.Scalzo F, Xu P, Asgari S, Bergsneider M, Hu X (2009) Regression analysis for peak designation in pulsatile pressure signals. Med Biol Eng Comput 47(9):967–977CrossRef PubMed PubMedCentral
  51.Scalzo F, Asgari S, Kim S, Bergsneider M, Hu X (2010) Robust peak recognition in intracranial pressure signals. Biomed Eng Online 9(1):61CrossRef PubMed PubMedCentral
  52.Schlesinger MI, Hlavac V Statistical pattern recognition toolbox. http://​cmp.​felk.​cvut.​cz/​cmp/​software/​stprtool/​
  53.Thakker B, Lal Vyas A (2011) Support vector machine for abnormal pulse classification. Int J Comput Appl 22(7):13–19
  54.Vermeersch SJ, Dynamics B, Society L (2010) Determinants of pulse wave velocity in healthy people and in the presence of cardiovascular risk factors: ‘establishing normal and reference values’. Eur Heart J 31(19):2338–2350CrossRef
  55.Wang X, Tian J (2012) A gene selection method for cancer classification. Comput Math Methods Med 2012:586246PubMed PubMedCentral
  56.Wang H, Zhang P (2008) A model for automatic identification of human pulse signals. J Zhejiang Univ Sci A 9(10):1382–1389CrossRef
  57.Wang K, Wang L, Wang D, Xu L (2004) SVM classification for discriminating cardiovascular disease patients from non-cardiovascular disease controls using pulse waveform variability analysis. In: Webb GI, Yu X (eds) AI 2004: advances in artificial intelligence. Springer, Berlin Heidelberg, pp 109–119CrossRef
  58.Weber T, Auer J, O’Rourke MF, Kvas E, Lassnig E, Berent R, Eber B (2004) Arterial stiffness, wave reflections, and the risk of coronary artery disease. Circulation 109(2):184–189CrossRef PubMed
  59.Wu B, Abbott T, Fishman D, McMurray W, Mor G, Stone K, Ward D, Williams K, Zhao H (2003) Comparison of statistical methods for classification of ovarian cancer using mass spectrometry data. Bioinformatics 19(13):1636–1643CrossRef PubMed
  60.Zajarevich N, Bia D, Pessana F, Codnia J, Armentano R (2010) Arterial pressure and diameter waveforms analysis by means of wavelet transform: application to artery de-endothelization. In: Conference on proceedings of the IEEE engineering in medicine and biology society, vol 2010. pp 4550–4553
  
• 作者单位：Tânia Pereira (1)
  Joana S. Paiva (2)
  Carlos Correia (1)
  João Cardoso (1)
  
  1. Physics Department, Instrumentation Center, University of Coimbra, Rua Larga, 3004-516, Coimbra, Portugal
  2. Physics Department, University of Coimbra, Coimbra, Portugal
  
• 刊物类别：Engineering
• 刊物主题：Biomedical Engineering
  Human Physiology
  Imaging and Radiology
  Computer Applications
  Neurosciences
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1741-0444
• 卷排序：54

文摘

The 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 . Keywords Arterial pulse waveform Optical system Feature creation Recursive feature elimination K-nearest neighbour algorithm Support vector machine