Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/43628
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Journal article |
Title: | Application of auto regressive models of wavelet sub-bands for classifying Terahertz pulse measurements |
Author: | Yin, X. Ng, B. Abbott, D. Ferguson, B. Hadjiloucas, S. |
Citation: | Journal of Biological Systems, 2007; 15(4):551-571 |
Publisher: | World Scientific Publ Co Pte Ltd |
Issue Date: | 2007 |
ISSN: | 0218-3390 1793-6470 |
Abstract: | 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. In this paper, 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%. © World Scientific Publishing Company. |
Description: | Copyright © 2007 World Scientific Publishing Company |
DOI: | 10.1142/S0218339007002374 |
Published version: | http://dx.doi.org/10.1142/s0218339007002374 |
Appears in Collections: | Aurora harvest Electrical and Electronic Engineering publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.