×

Application of autoregressive models of wavelet sub-bands for classifying terahertz pulse measurements. (English) Zbl 1146.92321

Summary: This paper presents an approach for automatic classification of pulsed Terahertz (THz), or T-ray, signals highlighting their potential in biomedical, pharmaceutical and security applications. T-ray classification systems supply a wealth of information about test samples and make possible the discrimination of heterogeneous layers within an object. A novel technique involving the use of Auto Regressive (AR) and Auto Regressive Moving Average (ARMA) models on the wavelet transforms of measured T-ray pulse data is presented. Two example applications are examined – the classification of normal human bone (NHB) osteoblasts against human osteosarcoma (HOS) cells and the identification of six different powder samples. A variety of model types and orders are used to generate descriptive features for subsequent classification. Wavelet-based de-noising with soft threshold shrinkage is applied to the measured T-ray signals prior to modeling. For classification, a simple Mahalanobis distance classifier is used. After feature extraction, classification accuracy for cancerous and normal cell types is 93%, whereas for powders, it is 98%.

MSC:

92C55 Biomedical imaging and signal processing
62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
65T60 Numerical methods for wavelets
62H30 Classification and discrimination; cluster analysis (statistical aspects)
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] DOI: 10.1016/S0065-2539(08)60259-0 · doi:10.1016/S0065-2539(08)60259-0
[2] Grundler W., Nanobiology 1 pp 163–
[3] DOI: 10.1109/TMTT.2004.835983 · doi:10.1109/TMTT.2004.835983
[4] DOI: 10.1088/0031-9155/47/21/324 · doi:10.1088/0031-9155/47/21/324
[5] DOI: 10.1109/18.382009 · Zbl 0820.62002 · doi:10.1109/18.382009
[6] Woodward R. M., J. Invest. Dermatol. 120 pp 3853–
[7] DOI: 10.1002/jps.20281 · doi:10.1002/jps.20281
[8] DOI: 10.1211/jpp.59.2.0008 · doi:10.1211/jpp.59.2.0008
[9] DOI: 10.1364/AO.42.005744 · doi:10.1364/AO.42.005744
[10] DOI: 10.1088/0268-1242/20/7/018 · doi:10.1088/0268-1242/20/7/018
[11] DOI: 10.1088/0268-1242/20/7/017 · doi:10.1088/0268-1242/20/7/017
[12] DOI: 10.1016/S0026-2692(01)00093-3 · doi:10.1016/S0026-2692(01)00093-3
[13] DOI: 10.1002/cpa.3160410705 · Zbl 0644.42026 · doi:10.1002/cpa.3160410705
[14] Qian S. E., Time-Frequency and Wavelet Transforms (2002)
[15] Mallat S. G., IEEE Trans. Pattern Anal. Mach. Intell. 14 pp 674–
[16] Mallat S. G., A Wavelet Tour of Signal Processing (1999) · Zbl 0945.68537
[17] Vaidyanathan P. P., Multirate Systems and Filter Banks (1993) · Zbl 0784.93096
[18] Vetterli M., Wavelets and Subband Coding (1995)
[19] Strang G., Wavelets and Filter Banks (1996)
[20] DOI: 10.1109/78.678504 · Zbl 1011.94517 · doi:10.1109/78.678504
[21] DOI: 10.1109/78.869032 · doi:10.1109/78.869032
[22] DOI: 10.5194/npg-14-79-2007 · doi:10.5194/npg-14-79-2007
[23] DOI: 10.1016/j.sigpro.2005.09.021 · Zbl 1172.94466 · doi:10.1016/j.sigpro.2005.09.021
[24] Daubechies I., Ten Lectures on Wavelets (1992) · Zbl 0776.42018
[25] DOI: 10.1007/978-3-642-56702-5 · doi:10.1007/978-3-642-56702-5
[26] DOI: 10.1017/CBO9780511841040 · doi:10.1017/CBO9780511841040
[27] Johnstone M. I., J. Stat. Assoc. 90 pp 1220–
[28] Therrien C. W., Discrete Random Signals and Statistical Signal Processing (1992) · Zbl 0747.94004
[29] Proakis J. G., Digital Signal Processing: Principles, Algorithms, and Applications (1996)
[30] Jain S., IEEE Biotechnol. Bioinf. 12 pp 85–
[31] DOI: 10.1109/78.277805 · Zbl 0825.93748 · doi:10.1109/78.277805
[32] Schürmann J., Pattern Classification: A Unified View of Statistical and Neural Approaches (1996)
[33] Withayachumnankul B., Process of SPIE Photonics: Design, Technology, and Packaging II (2005)
[34] DOI: 10.1109/JSEN.2006.890159 · doi:10.1109/JSEN.2006.890159
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.