VGU RESEARCH REPOSITORY
Please use this identifier to cite or link to this item:
https://epub.vgu.edu.vn/handle/dlibvgu/541| Title: | Dynamic contrast-ehanced magnetic resonance imaging - sparse sampling & reconstruction | Authors: | Le Hoang Nam | Keywords: | Dynamic contrast;Magnetic resonance imaging;Sparse sampling;Reconstruction | Issue Date: | 2017 | Publisher: | Vietnamese-German University | Abstract: | The purpose of this thesis was to apply and compare different acquisition and reconstruction methods to three breast DCE-MRI data and a phantom data. Specifically, the study revisited the generalized series (GS) method as well as its variation, called generalized series with temporal constraint (GenSeT) and applied them to the experimental data. The results were then compared to those of the discrete Fourier transform (DEC), the time-resolved imaging with stochastic trajectories (TWIST) and the generalized series with TWIST sequence (twistGS) methods, The GS model, developed by Liang and Lauterbur. uses a high resolution as reference for reconstruction. This method is based on the observation that in many imaging applications, the high-resolution image morphology does not change over time. Specifically. the reference image is incorporated in to the GS basis functions, the subsequent dynamic images can be reconstructed with fewer spatial encodings, which allows a faster and more efficient acquisition process. In this study, the experimental data were sparsely sampled along a predefined trajectory. The undersampled data were reconstructed using the above reconstruction schemes. Experimental results showed that the GS and GenSeT reconstructions provided better spatial resolution and more accurate dynamic information than the DFT method. However, the those methods showed consistently inferior performance compared to the TWIST and twistGS schemes. The results implied that the GS-based methods need to be improved mathematically in order to be applied effectively in DCE-MRI. Additionally, choices of sampling trajectories affected reconstruction results considerably. Finally, because the breast data, which were used as the groud truths, were acquired using the TWIST sequence, experimental results might have been biased in favor of the TW1ST-based methods. |
URI(1): | http://epub.vgu.edu.vn/handle/dlibvgu/541 | Rights: | Attribution-NonCommercial 4.0 International |
| Appears in Collections: | Electrical Engineering and Information Technology (EEIT) |
Files in This Item:
| File | Description | Size | Format | Existing users please Login |
|---|---|---|---|---|
| Dynamic contrast-ehanced magnetic resonance imaging - sparse sampling & reconstruction.pdf | 285.15 MB | Adobe PDF |
Page view(s)
36
checked on Aug 23, 2025
Download(s)
8
checked on Aug 23, 2025
Google ScholarTM
Check
This item is licensed under a Creative Commons License