Low dose, non-tomographic estimation of lesion position and trace element concentration in NSECT

Published

Journal Article

Neutron stimulated emission computed tomography (NSECT) is a quantitative imaging method that uses fast-neutron inelastic scatter to identify the elemental composition of diseased tissue in biological organs. Previous NSECT work has shown the ability to quantitatively image liver iron concentrations through tomographic imaging; however, such acquisition imparts considerable radiation dose. To implement NSECT as a low-dose diagnostic tool, we are developing a technique to simultaneously determine the element concentration and position from a single-angle scan of the tissue, thereby eliminating the need for tomography and reducing both scan time and radiation dose. Using known physical factors such as neutron and gamma attenuation that affect the detected gamma signal, a unique equation corresponding to the expected gamma counts can be developed for each detector in the acquisition system, and these equations can be solved iteratively to obtain a simultaneous estimate of the lesion position and iron concentration. As the first step towards the development of this algorithm, we describe here a graphical approach to localize and quantify an iron lesion in the liver without tomographic imaging. The acquisition system with a collimated neutron source, multiple gamma detectors, and a tissue phantom were simulated in GEANT4 and used to generate gamma spectra from 50 different combinations of lesion position and iron concentration: 49 known cases and 1 unknown 'test' case. Surface plots of gamma counts vs. lesion position and iron concentration from the 49 known combinations were generated for each detector. The 'test' lesion signal was overlaid on the surface plots to obtain the best estimate of the unknown lesion concentration and position in the beam. The results showed 100% accurate identification of the concentration and less than 20% error in the identified position. The results validate the approach for non-tomographic determination of these parameters. © 2011 IEEE.

Full Text

Duke Authors

Cited Authors

  • Agasthya, GA; Shah, JP; Harrawood, BP; Kapadia, AJ

Published Date

  • January 1, 2011

Published In

Start / End Page

  • 3796 - 3799

International Standard Serial Number (ISSN)

  • 1095-7863

Digital Object Identifier (DOI)

  • 10.1109/NSSMIC.2011.6153719

Citation Source

  • Scopus