A method to measure PET scatter fractions for daily quality control

H W de Jong, M. Lubberink, H Watabe, H. Iida, Adriaan A Lammertsma

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

PURPOSE: Regular monitoring of PET scanner performance is mandatory to assure quality of acquired data. While extensive performance measurements include many scanner characteristics such as resolution, count rate, uniformity, sensitivity, and scatter fraction (SF), most daily QC protocols are limited to uniformity and sensitivity measurements. These measurements may be too insensitive to detect more subtle drifts in detector gains that could lead to reduced detection of primary and increased detection of scattered events. Current methods to measure SF, such as those prescribed by the NEMA protocols (SF-NEMA), however, require specially designed phantoms and are too cumbersome to be performed on a daily basis.

METHODS: In this study, a simple and versatile method to determine SF is described. This method (SF-DAILY) does not require additional measurements, making it suitable for daily QC. The method was validated for four different scanners by comparing results with those obtained with the NEMA 1994 protocol.

RESULTS: For all scanner types and acquisition modes, excellent agreement was found between SF-NEMA and SF-DAILY.

CONCLUSIONS: The proposed method is a very practical and valuable addition to current daily QC protocols. In addition, the method can be used to accurately measure SF in phantoms with other dimensions than the NEMA phantom.

Original languageEnglish
Pages (from-to)4609-15
Number of pages7
JournalMedical Physics
Volume36
Issue number10
DOIs
Publication statusPublished - 2009

Keywords

  • Image Enhancement
  • Internationality
  • Positron-Emission Tomography
  • Quality Assurance, Health Care
  • Reproducibility of Results
  • Scattering, Radiation
  • Sensitivity and Specificity
  • Journal Article

Fingerprint

Dive into the research topics of 'A method to measure PET scatter fractions for daily quality control'. Together they form a unique fingerprint.

Cite this