Anatomical localization for pet using MR imaging

Carolyn Cidis Meltzer, R. Nick Bryan, Henry H. Holcomb, Allyn W. Kimball, Helen S. Mayberg, Bernard Sadzot, Jeffrey P. Leal, Henry N. Wagner, J. James Frost

Research output: Contribution to journalArticlepeer-review


Accurate localization of an imaging plane of interest is often needed prior to a positron emission tomographic (PET) study. We have developed a simple method for accurate and reproducible selection of an imaging plane for PET using magnetic resonance (MR) imaging. This method is useful when optimal sampling of specific brain structures, such as small subcortical nuclei, or when a specific imaging angle is required for the PET study. An external localizing device, consisting of a series of tubes visible on MR, is affixed to an individually fitted thermoplastic mask. This mask system is worn by the patient during both the MR and PET studies. A plane of interest is planned from the sagittal MR image and defined by its relation to the localizing device and to the MR scanner’s “landmark” or reference position. This plane is transferred to the mask by means of a calibrated alignment laser. The coplanar acquisition of MR and PET images allows individualized analysis of brain structure-function relationships. Phantom studies demonstrated the accuracy and reproducibility of imaging plane selection by this method to be within 1 mm and 1°. Application of the localization protocol in a human subject is also presented.

Original languageEnglish (US)
Pages (from-to)418-426
Number of pages9
JournalJournal of computer assisted tomography
Issue number3
StatePublished - May 1990
Externally publishedYes


  • Anatomy
  • Atlas and atlases
  • Brain
  • Emission computed tomography
  • Emission computed tomography
  • Localization
  • Magnetic resonance imaging Techniques
  • Techniques

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Anatomical localization for pet using MR imaging'. Together they form a unique fingerprint.

Cite this