Optimizing multi-dimensional terahertz imaging analysis for colon cancer diagnosis

Leila H Eadie, Caroline B Reid, Anthony J Fitzgerald, Vincent P Wallace

Research output: Contribution to journalArticle

33 Citations (Scopus)
3 Downloads (Pure)

Abstract

Terahertz reflection imaging (at frequencies ~0.1–10 THz/1012 Hz) is non-ionizing and has potential as a medical imaging technique; however, there is currently no consensus on the optimum imaging parameters to use and the procedure for data analysis. This may be holding back the progress of the technique. This article describes the use of various intelligent analysis methods to choose relevant imaging parameters and optimize the processing of terahertz data in the diagnosis of ex vivo colon cancer samples. Decision trees were used to find important parameters, and neural networks and support vector machines were used to classify the terahertz data as indicating normal or abnormal samples. This work reanalyzes the data described in Reid et al. (2011) (Physics in Medicine and Biology, 56, 4333–4353), and improves on their reported diagnostic accuracy, finding sensitivities of 90–100% and specificities of 86–90%. This optimization of the analysis of terahertz data allows certain recommendations to be suggested concerning terahertz reflection imaging of colon cancer samples.
Original languageEnglish
Pages (from-to)2043-2050
Number of pages8
JournalExpert Systems with Applications
Volume40
Issue number6
DOIs
Publication statusPublished - May 2013

Fingerprint

Imaging techniques
Medical imaging
Decision trees
Medicine
Support vector machines
Physics
Neural networks
Processing

Keywords

  • terahertz
  • optimization
  • neural networks
  • support vector machines
  • decision tree
  • colon cancer

Cite this

Optimizing multi-dimensional terahertz imaging analysis for colon cancer diagnosis. / Eadie, Leila H; Reid, Caroline B; Fitzgerald, Anthony J; Wallace, Vincent P.

In: Expert Systems with Applications, Vol. 40, No. 6, 05.2013, p. 2043-2050.

Research output: Contribution to journalArticle

Eadie, Leila H ; Reid, Caroline B ; Fitzgerald, Anthony J ; Wallace, Vincent P. / Optimizing multi-dimensional terahertz imaging analysis for colon cancer diagnosis. In: Expert Systems with Applications. 2013 ; Vol. 40, No. 6. pp. 2043-2050.
@article{335afcb446614c448f3e5bb885071dcd,
title = "Optimizing multi-dimensional terahertz imaging analysis for colon cancer diagnosis",
abstract = "Terahertz reflection imaging (at frequencies ~0.1–10 THz/1012 Hz) is non-ionizing and has potential as a medical imaging technique; however, there is currently no consensus on the optimum imaging parameters to use and the procedure for data analysis. This may be holding back the progress of the technique. This article describes the use of various intelligent analysis methods to choose relevant imaging parameters and optimize the processing of terahertz data in the diagnosis of ex vivo colon cancer samples. Decision trees were used to find important parameters, and neural networks and support vector machines were used to classify the terahertz data as indicating normal or abnormal samples. This work reanalyzes the data described in Reid et al. (2011) (Physics in Medicine and Biology, 56, 4333–4353), and improves on their reported diagnostic accuracy, finding sensitivities of 90–100{\%} and specificities of 86–90{\%}. This optimization of the analysis of terahertz data allows certain recommendations to be suggested concerning terahertz reflection imaging of colon cancer samples.",
keywords = "terahertz, optimization, neural networks, support vector machines, decision tree, colon cancer",
author = "Eadie, {Leila H} and Reid, {Caroline B} and Fitzgerald, {Anthony J} and Wallace, {Vincent P}",
year = "2013",
month = "5",
doi = "10.1016/j.eswa.2012.10.019",
language = "English",
volume = "40",
pages = "2043--2050",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "PERGAMON-ELSEVIER SCIENCE LTD",
number = "6",

}

TY - JOUR

T1 - Optimizing multi-dimensional terahertz imaging analysis for colon cancer diagnosis

AU - Eadie, Leila H

AU - Reid, Caroline B

AU - Fitzgerald, Anthony J

AU - Wallace, Vincent P

PY - 2013/5

Y1 - 2013/5

N2 - Terahertz reflection imaging (at frequencies ~0.1–10 THz/1012 Hz) is non-ionizing and has potential as a medical imaging technique; however, there is currently no consensus on the optimum imaging parameters to use and the procedure for data analysis. This may be holding back the progress of the technique. This article describes the use of various intelligent analysis methods to choose relevant imaging parameters and optimize the processing of terahertz data in the diagnosis of ex vivo colon cancer samples. Decision trees were used to find important parameters, and neural networks and support vector machines were used to classify the terahertz data as indicating normal or abnormal samples. This work reanalyzes the data described in Reid et al. (2011) (Physics in Medicine and Biology, 56, 4333–4353), and improves on their reported diagnostic accuracy, finding sensitivities of 90–100% and specificities of 86–90%. This optimization of the analysis of terahertz data allows certain recommendations to be suggested concerning terahertz reflection imaging of colon cancer samples.

AB - Terahertz reflection imaging (at frequencies ~0.1–10 THz/1012 Hz) is non-ionizing and has potential as a medical imaging technique; however, there is currently no consensus on the optimum imaging parameters to use and the procedure for data analysis. This may be holding back the progress of the technique. This article describes the use of various intelligent analysis methods to choose relevant imaging parameters and optimize the processing of terahertz data in the diagnosis of ex vivo colon cancer samples. Decision trees were used to find important parameters, and neural networks and support vector machines were used to classify the terahertz data as indicating normal or abnormal samples. This work reanalyzes the data described in Reid et al. (2011) (Physics in Medicine and Biology, 56, 4333–4353), and improves on their reported diagnostic accuracy, finding sensitivities of 90–100% and specificities of 86–90%. This optimization of the analysis of terahertz data allows certain recommendations to be suggested concerning terahertz reflection imaging of colon cancer samples.

KW - terahertz

KW - optimization

KW - neural networks

KW - support vector machines

KW - decision tree

KW - colon cancer

U2 - 10.1016/j.eswa.2012.10.019

DO - 10.1016/j.eswa.2012.10.019

M3 - Article

VL - 40

SP - 2043

EP - 2050

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 6

ER -