%0 Journal Article
%T Estimating myocardial perfusion from dynamic contrast-enhanced CMR with a model-independent deconvolution method
%A Nathan A Pack
%A Edward VR DiBella
%A Thomas C Rust
%A Dan J Kadrmas
%A Christopher J McGann
%A Regan Butterfield
%A Paul E Christian
%A John M Hoffman
%J Journal of Cardiovascular Magnetic Resonance
%D 2008
%I BioMed Central
%R 10.1186/1532-429x-10-52
%X An iterative model-independent analysis method was developed and tested to estimate regional and pixelwise myocardial perfusion in five normal subjects imaged with a saturation recovery turboFLASH sequence at 3 T CMR. Estimates of myocardial perfusion using model-independent analysis are dependent on the choice of the regularization weight parameter, which increases nonlinearly to handle large decreases in the contrast-to-noise ratio of the measured tissue enhancement data. Quantitative perfusion estimates in five subjects imaged with 3 T CMR were 1.1 ¡À 0.8 ml/min/g at rest and 3.1 ¡À 1.7 ml/min/g at adenosine stress. The perfusion estimates correlated with dynamic 13N-ammonia PET (y = 0.90x + 0.24, r = 0.85) and were similar to results from other validated CMR studies.This work shows that a model-independent analysis method that uses iterative minimization and temporal regularization can be used to quantify myocardial perfusion with dynamic contrast-enhanced perfusion CMR. Results from this method are robust to choices in the regularization weight parameter over relatively large ranges in the contrast-to-noise ratio of the tissue enhancement data.Dynamic contrast-enhanced cardiovascular magnetic resonance (DCE-CMR) is a commonly used tool for detecting and quantifying reductions in myocardial blood flow (perfusion) in patients with coronary artery disease (CAD). The early diagnosis of CAD can provide valuable information that may affect interventional strategies in patients with ischemia [1]. In DCE-CMR perfusion studies, a paramagnetic gadolinium (Gd) complex is injected into the patient while at rest and at stress¨Cduring which a pharmacological vasodilator (adenosine) is simultaneously administered to the patient. Once the Gd is injected, it flows through the heart and temporarily distributes in the myocardium before being 'washed out' of the body. The spatiotemporal distribution of Gd within the heart can be measured dynamically and the resultant blood and tissue
%U http://jcmr-online.com/content/10/1/52