Image manipulation and error estimation for MRI analysis

Elena Plante, Anne Uecker, Ava Senkfor, Arthur F. Gmitro

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

Manipulation of MRI images prior to volumetric analysis is a common practice that may unwittingly lead to errors in measurement. In this study, we examine the effects of two types of image manipulation: changes in the total number of slices used to obtain volume estimates (slice sampling rate) and image rotation. A phantom containing two regularly-shaped and two irregularly-shaped regions of interest (ROIs) was scanned using an SPGR sequence and 1-mm slices. Changes in slice sampling rate produced marked effects on volume estimation of irregularly-shaped ROIs. Comparatively little error was associated with changes in slice sampling rates for regularly- shaped ROIs. In addition, there was an interaction between image rotation in non-orthogonal planes and slice sampling rate. The data suggests that the ability to detect anatomical effects may be influenced by an investigator's choices concerning the number of slices included in a region of interest and image rotation when estimating volumes.

Original languageEnglish (US)
Pages (from-to)175-180
Number of pages6
JournalJournal of Neuroscience Methods
Volume85
Issue number2
DOIs
StatePublished - Dec 1 1998

Keywords

  • Cavalieri principle
  • MRI
  • Measurement error
  • Volume estimation

ASJC Scopus subject areas

  • Neuroscience(all)

Fingerprint Dive into the research topics of 'Image manipulation and error estimation for MRI analysis'. Together they form a unique fingerprint.

Cite this