Estimation of Tissue Permeability in MRI Images by Empirical Mode Decomposition Method.

Abstract

The precise evaluation of tissue permeability using the Magnetic Resonance Imaging (MRI) method requires high-quality images. Due to noisy acquired dynamic MRI images, some methods of processing are required to obtain the imaging detail of interest. This study aimed to implement Empirical Mode Decomposition (EMD) to the Lock-Locker (LL) images to improve the permeability of the normal tissue and tumor region, evaluated by the Quantitative Autoradiography (QAR) method. In this experimental and analytical study, the EMD method was used to improve the tissue permeability from the LL- MRI images of the rat brain. The EMD components were extracted from LL images, and the resulting components were combined using different weighting factors. The tissue permeability was derived by extracting the information of each pixel from the LL image series and fitting curves to the data. The optimum weighted combination factors images were 0.7 for the middle and low-frequency components and 1 for the high-frequency component. The calculated tissue permeability was between 0.0023-0.0043 (ml.min-1.g-1) for abnormal tissue. The estimation of the permeability of tumors in the rat brain with the LL images and the processed LL images by the EMD method shows that the EMD method and the weighted combination of frequency components can improve the permeability calculation in the LL images for the rat brain. The results of permeability estimation by EMD due to noise reduction of LL images are closer to the values obtained from the Quantitative Autoradiography (QAR).