Exploratory data analysis of dynamic cerebral contrast-enhanced perfusion MRI time-series

We compare experimentally four different unsupervised clustering techniques as a tool for the analysis of dynamic cerebral contrast-enhanced perfusion MRI time-series in patients with and without stroke. The goal of the paper is to determine the robustness and reliability of clustering methods in providing a self-organized segmentation of perfusion MRI data sharing common properties of signal dynamics. By using the whole information provided by the dynamic time series, we introduce an extension to the conventional method of analyzing perfusion MRI studies based on the evaluation of a few parameters such as mean transit time (MTT), regional cerebral blood volume (rCBV), and regional cerebral blood flow (rCBF).