Abstract
The Transcranial Doppler ultrasound can be used to detect asymptomatic circulating cerebral emboli. Emboli indicate particles that can plug the arterial system. Asymptomatic emboli signals help to discover critical stroke events by taking the embolus activities into consideration. Cerebral emboli detection is searched deeply in the literature. But none of them proposed a polynomial method to generalize the solution of the emboli detection. High Dimensional Model Representation (HDMR) philosophy is an effective way of generating an analytical model for a given multivariate data modeling problem, that is, HDMR can be used in constructing a general polynomial model for detecting embolism. In this study, emboli related data set was collected from \(35\) different patients. HDMR based methods and various data mining techniques were used to detect emboli through that data set. The Euclidean Matrix Based Indexing HDMR method has an important superiority in terms of generating a polynomial model which can then be used in other emboli detection problems without a training process. The method also shows satisfactory results in generalizing the emboli characteristics when compared with the other methods.
Similar content being viewed by others
References
Aydin N, Marvasti F, Markus HS (2004) Embolic doppler ultrasound signal detection using discrete wavelet transform. IEEE Trans Inf Technol Biomed 8(2):182–190
Banerjee I, Ierapetritou MG (2004) Model independent parametric decision making. Ann Oper Res 132(1–4):135–155
Chowdhury R, Rao BN, Prasad A (2008) High dimensional model representation for piece-wise continuous function approximation. Commun Numer Meth Eng 24:1587–1609
Chung E, Fan LK, Degg C, Evans DH (2005) Detection of doppler embolic signals: psychoacoustic considerations. Ultrasound Med Biol 31:1177–1184
Contreras RJ, Vellasco MMBR, Tanscheit R (2011) Hierarchical type-2 neuro-fuzzy bsp model. Inf Sci 181(15):3210–3224
Deitel HM, Deitel PJ, Nieto TR, McPhie DC (2001) How to program perl. Prentice Hall, New Jersey
Demiralp M (2003) High dimensional model representation and its application varieties. Math Res 9:146–159
Felkin M (2007) Comparing classification results between n-ary and binary problems. Stud Comput Intell 43:277–301
Frank E, Witten IH (2005) Data mining: practical machine learning tools and techniques. Morgan Kaufmann, San Francisco
Girault JM, Biard M, Kouame D, Bleuzen A, Tranquart F (2006) Spectral correlation of the embolic blood doppler signal. Acoustics, speech and signal processing (ICASSP 2006) In: Proceedings 2:1200–1203
Girault JM, Kouame D, Tranquart F (2007) Synchronous detection of emboli by wavelet packet decomposition. Acoustics, speech and signal processing (ICASSP 2007) In: Proceedings 1:409–412
Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten IH (2009) The WEKA data mining software: an update. SIGKDD Explor 11(1):10–18
Karahoca A, Kucur T, Aydin N (2007) Data mining usage in emboli detection. In: Proceedings of ECSIS Symposium on bio-inspired, learning, and intelligent systems for security (BLISS 2007), pp 159–162
Karahoca A, Tunga MA (2012) Dosage planning for type 2 diabetes mellitus patients using indexing hdmr. Exp Syst Appl 39(8):7207–7215
Kouame D, Biard M, Girault JM, Bleuzen A (2006) Adaptive ar and neurofuzzy approaches: access to cerebral particle signatures. IEEE Trans Inf Technol Biomed 10(3):559–566
Li GY, Rabitz H, Hu JS, Chen Z, Ju YG (2008) Regularized random-sampling high dimensional model representation (rs-hdmr). J Math Chem 43:1207–1232
Mackinnon AD, Aaslid R, Markus HS (2005) Ambulatory transcranial doppler cerebral embolic signal detection in symptomatic and asymptomatic carotid stenosis. Stroke 36:1726–1730
Marvasti SA, Ghandi M, Marvasti AA, Deb A, Markus HS, Gillies D (2005) Improved detection of embolic signals using multi scale wavelet filtering, ar and ann, for tcd ultrasound. Med Appl Sign Proc 11199:59–64
Miller M, Feng XJ, Li GY, Rabitz H (2009) Nonlinear bionetwork structure inference using the random sampling-high dimensional model representation (rs-hdmr) algorithm. In: Annual International Conference of the IEEE Eengineering in Medicine and Biology Society (EMBC 2009) 1(20):6412–6415
Nebuya S, Noshiro M, Brown BH, Smallwood RH, Milnes P (2004) Estimation of the size of air emboli detectable by electrical impedance measurement. Med Biol Eng Comput 42:142–144
Oevel W, Postel F, Wehmeier S, Gerhard J (2000) The MuPAD tutorial. Springer, New York
Palanchon P, Bouakaz A, Klein J, de Jong N (2005) Multifrequency transducer for microemboli classification and sizing. IEEE Trans Biomed Eng 52:2087–2092
Quinlan JR (1993) C4.5: programs for machine learning. Morgan Kaufmann Publishers Inc., San Francisco
Rodriguez RA, Rubens F, Rodriguez CD, Nathan HJ (2006) Sources of variability in the detection of cerebral emboli with transcranial doppler during cardiac surgery. J Neuroimag 16:126–132
Serpen G, Tekkedil DK, Orra M (2008) A knowledge-based artificial neural network classifier for pulmonary embolism diagnosis. Comput Biol Med 38(2):204–220
Sobol IM (1993) Sensitivity estimates for nonlinear mathematical models. Math Model Comput Exp 1:407–414
Spencer MP (1997) Transcranial doppler monitoring and causes of stroke from carotid endarterectomy. Stroke 28:685–691
Tunga MA, Demiralp M (2003) Data partitioning via generalized high dimensional model representation (ghdmr) and multivariate interpolative applications. Math Res 9:447–462
Tunga MA, Demiralp M (2008) A new approach for data partitioning through high dimensional model representation. Int J Comput Math 85(12):1779–1792
Tunga B, Demiralp M (2009) Constancy maximization based weight optimization in high dimensional model representation. Numer Algor 52(3):435–459
Tunga MA (2011a) An approximation method to model multivariate interpolation problems: indexing hdmr. Math Comput Modell 53(9–10):1970–1982
Tunga MA (2011b) A matrix based indexing hdmr method for multivariate data modelling. J Math Chem 49:1092–1114
Xu D, Wang Y (2007) An automated feature extraction and emboli detection system based on the pca and fuzzy sets. Comput Biol Med 37(6):861–871
Ziehn T, Tomlin AS (2008) A global sensitivity study of sulfur chemistry in a premixed methane flame model using hdmr. Int J Chem Kinet 40:742–753
Ziehn T, Tomlin AS (2009) Gui-hdmr—a software tool for global sensitivity analysis of complex models. Environ Modell Softw 24:775–785
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by M. J. Watts.
Rights and permissions
About this article
Cite this article
Karahoca, A., Tunga, M.A. A polynomial based algorithm for detection of embolism. Soft Comput 19, 167–177 (2015). https://doi.org/10.1007/s00500-014-1240-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-014-1240-x