Correlation in sequential frame analysis of coronary angiogram: Validation and clinical considerations

D. M. Herrington, F. C.P. Yin, Gary D. Walford

Research output: Contribution to journalConference articlepeer-review


Automated measurement of cardiac cycle dependent changes in coronary artery dimension requires successful registration of points on vessel edges in sequential frames of a cineangiogram. This process is complicated by nonlinearities in the X-ray/image acquisition chain and spatial translations of the vessel. To determine if an implementation of a digital correlation described by Ranson and Vachon can follow unique points on a vessel edge, the authors used angiographic images of injected postmortem hearts and standard clinical angiograms to measure the pixel distance between operator selected edge point and the corresponding autocorrelated points while controlling for several sources of variability. The method allows changes in the diameter profile and location of the edges of a vessel to be displayed in an animated fashion which can then be used for a qualitative and quantitative assessment of mechanical properties of vessel segments.

Original languageEnglish (US)
Pages (from-to)199-200
Number of pages2
JournalComputers in cardiology
StatePublished - Dec 1 1987

ASJC Scopus subject areas

  • Computer Science Applications
  • Cardiology and Cardiovascular Medicine


Dive into the research topics of 'Correlation in sequential frame analysis of coronary angiogram: Validation and clinical considerations'. Together they form a unique fingerprint.

Cite this