Visual analysis of cardiac 4D MRI blood flow using line predicates

Silvia Born*, Matthias Pfeifle, Michael Markl, Matthias Gutberlet, Gerik Scheuermann

*Corresponding author for this work

Research output: Contribution to journalArticle

24 Scopus citations

Abstract

Four-dimensional MRI is an in vivo flow imaging modality that is expected to significantly enhance the understanding of cardiovascular diseases. Among other fields, 4D MRI provides valuable data for the research of cardiac blood flow and with that the development, diagnosis, and treatment of various cardiac pathologies. However, to gain insights from larger research studies or to apply 4D MRI in the clinical routine later on, analysis techniques become necessary that allow to robustly identify important flow characteristics without demanding too much time and expert knowledge. Heart muscle contractions and the particular complexity of the flow in the heart imply further challenges when analyzing cardiac blood flow. Working toward the goal of simplifying the analysis of 4D MRI heart data, we present a visual analysis method using line predicates. With line predicates precalculated integral lines are sorted into bundles with similar flow properties, such as velocity, vorticity, or flow paths. The user can combine the line predicates flexibly and by that carve out interesting flow features helping to gain overview. We applied our analysis technique to 4D MRI data of healthy and pathological hearts and present several flow aspects that could not be shown with current methods. Three 4D MRI experts gave feedback and confirmed the additional benefit of our method for their understanding of cardiac blood flow.

Original languageEnglish (US)
Article number6365632
Pages (from-to)900-912
Number of pages13
JournalIEEE Transactions on Visualization and Computer Graphics
Volume19
Issue number6
DOIs
StatePublished - Apr 17 2013

Keywords

  • 4D MRI
  • Flow visualization
  • line predicates
  • medical visualization

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Computer Graphics and Computer-Aided Design

Fingerprint Dive into the research topics of 'Visual analysis of cardiac 4D MRI blood flow using line predicates'. Together they form a unique fingerprint.

  • Cite this