DCE-MRI for quantification and segmentation of tumors
MetadataShow full item record
Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has been widely studied as a cancer imaging tool that provides information about blood volume and microvascular permeability by tracking the exchange of contrast agent between the vascular space and extravascular extracellular space. To quantitatively analyze the DCE-MRI data, concentration of the contrast agent in the blood plasma, the so-called arterial input function (AIF), has a very important role in estimating pharmacokinetic parameters accurately. However, the AIF is usually unknown, and it remains very difficult to obtain such information noninvasively. In this study, a reference region (RR) model that does not require information of AIF is used to analyze the kinetic parameters. However, the RR model usually depends on kinetic parameters found in previous studies of a reference region and may generate errors if wrong values are assigned from previous studies for the reference region. In this work, we proposed two pharmacokinetic parameter ratios between the tissue of interest (TOI) and the reference region to overcome these problems. To more accurately analyze DCE-MRI data, an analytical approach is introduced. This analytical method can estimate parameters more accurately than numerical analysis over various SNRs and temporal resolutions. In the studies of tumor segmentation using MR data, partial volume effect (PVE) is one of the major difficulties and may result in inaccurate segmentation results due to inherent low spatial resolution of images. In this study, we introduced the temporal independent component analysis (ICA) to solve partial volume effect (PVE) in tumor segmentation.