3D position of radiation sources using an automated gamma camera and ML algorithm with energy-dependent response functions

Wonho Lee, David K. Wehe

Research output: Contribution to journalConference articlepeer-review

2 Citations (Scopus)

Abstract

In this work, the 3D position information of a radiation source is determined by a compact gamma ray imaging system. 2D gamma ray images were obtained from different positions by the gamma camera and the third dimension, the distance between the detector and the radiation source, was calculated using triangulation. Additionally, a CCD camera is attached to the top of the gamma camera and provides coincident 2D visual information. The inferred distances from the center of the two measurement points and a radiation source had less than a 5% error within a range of 3m. From the measured distances and camera intrinsic efficiencies ε(θ,φ) from MCNP simulations, the activity of the source was normally determined within 80% of the true value depending upon source position. The parallax between the two visual images was corrected using the inferred distance between the detector and the radiation source. The radiation image from gamma camera and the visual image from CCD camera are superimposed into one combined image using a maximum-likelihood (ML) algorithm to make the image alignment more precise. Energy dependent response functions were found to be better than a fixed energy response function for ML image processing.

Original languageEnglish
Article numberN26-65
Pages (from-to)737-741
Number of pages5
JournalIEEE Nuclear Science Symposium Conference Record
Volume2
Publication statusPublished - 2003
Externally publishedYes
Event2003 IEEE Nuclear Science Symposium Conference Record - Nuclear Science Symposium, Medical Imaging Conference - Portland, OR, United States
Duration: 2003 Oct 192003 Oct 25

ASJC Scopus subject areas

  • Radiation
  • Nuclear and High Energy Physics
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of '3D position of radiation sources using an automated gamma camera and ML algorithm with energy-dependent response functions'. Together they form a unique fingerprint.

Cite this