Premium
The utility of principal component analysis for the image display of brain lesions. A preliminary, comparative study
Author(s) -
Schmiedl Udo,
Ortendahl Douglas A.,
Mark Alexander S.,
Berry Isabelle,
Kaufman Leon
Publication year - 1987
Publication title -
magnetic resonance in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.696
H-Index - 225
eISSN - 1522-2594
pISSN - 0740-3194
DOI - 10.1002/mrm.1910040508
Subject(s) - principal component analysis , component (thermodynamics) , artificial intelligence , computer science , pattern recognition (psychology) , image (mathematics) , psychology , computer vision , physics , thermodynamics
Principal component analysis (PCA), a common tool from multivariate statistical analysis, has been implemented into the computer display system of a MR imaging device. PCA allows the calculation of images in which the information in a defined region of interest inherent in the basic acquired images is condensed. PCA image calculation has been applied to acquired MR studies of 13 patients with brain lesions. The appearance of the brain lesions on the resultant PCA images was scored in comparison to the acquired images before and after administration of Gd‐DTPA as well as to other calculated images including T 1 , T 2 hydrogen density, and contrast‐optimized images. The conspicuity of a lesion and the number of distinguishable components within a lesion were slightly supenor on PCA than on the acquired images. PCA is an analytical tool for MR imaging that should be helpful in revealing information that is inherent in, but not readily visible on, standard acquired MR images. © 1987 Academic Press, Inc.