Premium
Model‐based and model‐free parametric analysis of breast dynamic‐contrast‐enhanced MRI
Author(s) -
Eyal Erez,
Degani Hadassa
Publication year - 2009
Publication title -
nmr in biomedicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.278
H-Index - 114
eISSN - 1099-1492
pISSN - 0952-3480
DOI - 10.1002/nbm.1221
Subject(s) - contrast (vision) , computer science , dynamic contrast enhanced mri , breast mri , dynamic contrast , parametric statistics , artificial intelligence , pattern recognition (psychology) , machine learning , data mining , magnetic resonance imaging , mammography , breast cancer , medicine , radiology , mathematics , statistics , cancer
A wide range of dynamic‐contrast‐enhanced (DCE) sequences and protocols, image processing methods, and interpretation criteria have been developed and evaluated over the last 20 years. In particular, attempts have been made to better understand the origin of the contrast observed in breast lesions using physiological models that take into account the vascular and tissue‐specific features that influence tracer perfusion. In addition, model‐free algorithms to decompose enhancement patterns in order to segment and classify different breast tissue types have been developed. This review includes a description of the mechanism of contrast enhancement by gadolinium‐based contrast agents, followed by the current status of the physiological models used to analyze breast DCE‐MRI and related critical issues. We further describe more recent unsupervised and supervised methods that use a range of different common algorithms. The model‐based and model‐free methods strive to achieve scientific accuracy and high clinical performance – both important goals yet to be reached. Copyright © 2008 John Wiley & Sons, Ltd.