Abstract
Terahertz reflection imaging (at frequencies ∼0.1–10THz/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. (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.
Highlights
Terahertz (THz) imaging is a novel technique for medical imaging
The parameters identified by the Decision trees (DT) as important (plus an additional parameter that was found to be useful for a 3-way classification task: WP4) were used in a reduced input data set for the neural networks (NN) and support vector machines (SVM) analysis
NN were used to analyze three different data sets: the full data set, a principal component analysis (PCA) reduced input data set, and the reduced input data set suggested by the DT results (8 parameters)
Summary
Terahertz (THz) imaging (imaging at frequencies around 1012 Hz) is a novel technique for medical imaging. It uses non-ionizing radiation and can safely be used for imaging different types of tissue, such as normal cells and tumors; the contrast between tissue types is thought to occur due to differences in water absorption, protein density or cellular structure. THz imaging is still relatively new in the field of medical diagnosis. The optimum methods for using the various data aspects recorded by THz systems are still being determined. This research examines the use of ‘intelligent’ methods of data analysis to optimize THz imaging for cancer diagnosis, and determine which parameters and analysis method are most useful.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
