Orthogonal shape modes describing clinical indices of remodeling

Xingyu Zhang, Bharath Ambale-Venkatesh, David A. Bluemke, Brett R. Cowan, J. Paul Finn, William G. Hundley, Alan H. Kadish, Daniel C. Lee, Joao A.C. Lima, Avan Suinesiaputra, Alistair A. Young, Pau Medrano-Gracia

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Quantification of the left ventricle (LV) shape changes (remodeling) is of great importance for therapeutic management of myocardial infarction. Orthogonal shape modes derived from principal component analysis (PCA) often do not describe clinical remodeling indices. We developed a method for deriving orthogonal shape modes directly from any set of clinical indices. Cardiac magnetic resonance images of 1,991 asymptomatic volunteers from the MESA study (age 44−84, mean age 62, 52% women) and 300 patients with myocardial infarction from the DETERMINE study (age 31−86, mean age 63, 20% women) were obtained from the Cardiac Atlas Project. Clinical indices of LV size, sphericity, wall thickness and apical conicity were calculated. For each index, cases outside two standard deviations of the mean, but within one standard deviation for all other indices, were chosen as a representative subgroup. Orthogonal modes were defined sequentially, using the first principal component of each subgroup. At each step, the contribution of the previous mode was removed mathematically from the shape description, similar to Gram–Schmidt orthogonalization. Correlation analysis and logistic regression were performed to show the effectiveness of these features to characterize remodeling due to myocardial infarction.

Original languageEnglish (US)
Title of host publicationFunctional Imaging and Modeling of the Heart - 8th International Conference, FIMH 2015, Proceedings
EditorsHans van Assen, Peter Bovendeerd, Hans van Assen, Peter Bovendeerd, Tammo Delhaas, Tammo Delhaas
PublisherSpringer Verlag
Pages273-281
Number of pages9
ISBN (Print)9783319203089, 9783319203089
DOIs
StatePublished - Jan 1 2015
Event8th International Conference on Functional Imaging and Modeling of the Heart, FIMH 2015 - Maastricht, Netherlands
Duration: Jun 25 2015Jun 27 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9126
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other8th International Conference on Functional Imaging and Modeling of the Heart, FIMH 2015
CountryNetherlands
CityMaastricht
Period6/25/156/27/15

Keywords

  • Cardiac remodelling
  • Magnetic resonance imaging
  • Principal component analysis

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Orthogonal shape modes describing clinical indices of remodeling'. Together they form a unique fingerprint.

Cite this