Expert Systems with Applications
Available online 16 October 2012
Leila H. Eadie
Centre for Computational Intelligence, De Montfort University, The Gateway, Leicester, UK; Centre for Rural Health, Aberdeen University, Centre for Health Science, Old Perth Road, Inverness, UK
Centre for Computational Intelligence, De Montfort University, The Gateway, Leicester, UK; Centre for Rural Health, Aberdeen University, Centre for Health Science, Old Perth Road, Inverness, UK
Abstract
Terahertz reflection imaging (at frequencies ∼0.1–10 THz/1012Hz) 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. (Physics in Medicine and Biology, 2011, 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.
for full paper see http://www.sciencedirect.com/science/article/pii/S0957417412011335
... One or more normal and diseased samples were
taken from each patient. The samples were imaged using a stand-alone portable THz system, the TPIimaga1000 (TeraViewLtd, Cambridge, UK) in reflection mode. Full details of the system can be found in Wallace et
al. (2004). ...
For more information about the Spectra 3000 visit http://www.teraview.com/products/terahertz-pulsed-spectra-3000/index.html