作者: Hui Xue , Sven Zuehlsdorff , Peter Kellman , Andrew Arai , Sonia Nielles-Vallespin
DOI: 10.1007/978-3-642-04271-3_90
关键词: Proton density 、 Perfusion 、 Cardiac perfusion 、 Motion compensation 、 Mr images 、 Computer vision 、 Artificial intelligence 、 Nuclear medicine 、 Perfusion magnetic resonance imaging 、 Computer science
摘要: In this paper we first discuss the technical challenges preventing an automated analysis of cardiac perfusion MR images and subsequently present a fully unsupervised workflow to address problems. The proposed solution consists key-frame detection, consecutive motion compensation, surface coil inhomogeneity correction using proton density robust generation pixel-wise parameter maps. entire processing chain has been implemented on clinical systems achieve inline MRI. Validation results are reported for 260 time series, demonstrating feasibility approach.