Matching Items (3)
Filtering by
- All Subjects: Compressed Sensing
- Creators: Kodibagkar, Vikram
- Resource Type: Text
Description
Magnetic Resonance Imaging using spiral trajectories has many advantages in speed, efficiency in data-acquistion and robustness to motion and flow related artifacts. The increase in sampling speed, however, requires high performance of the gradient system. Hardware inaccuracies from system delays and eddy currents can cause spatial and temporal distortions in the encoding gradient waveforms. This causes sampling discrepancies between the actual and the ideal k-space trajectory. Reconstruction assuming an ideal trajectory can result in shading and blurring artifacts in spiral images. Current methods to estimate such hardware errors require many modifications to the pulse sequence, phantom measurements or specialized hardware. This work presents a new method to estimate time-varying system delays for spiral-based trajectories. It requires a minor modification of a conventional stack-of-spirals sequence and analyzes data collected on three orthogonal cylinders. The method is fast, robust to off-resonance effects, requires no phantom measurements or specialized hardware and estimate variable system delays for the three gradient channels over the data-sampling period. The initial results are presented for acquired phantom and in-vivo data, which show a substantial reduction in the artifacts and improvement in the image quality.
ContributorsBhavsar, Payal (Author) / Pipe, James G (Thesis advisor) / Frakes, David (Committee member) / Kodibagkar, Vikram (Committee member) / Arizona State University (Publisher)
Created2013
Description
Compressed sensing magnetic resonance spectroscopic imaging (MRSI) is a noninvasive and in vivo potential diagnostic technique for cancer imaging. This technique undersamples the distribution of specific cancer biomarkers within an MR image as well as changes in the temporal dimension and subsequently reconstructs the missing data. This technique has been shown to retain a high level of fidelity even with an acceleration factor of 5. Currently there exist several different scanner types that each have their separate analytical methods in MATLAB. A graphical user interface (GUI) was created to facilitate a single computing platform for these different scanner types in order to improve the ease and efficiency with which researchers and clinicians interact with this technique. A GUI was successfully created for both prospective and retrospective MRSI data analysis. This GUI retained the original high fidelity of the reconstruction technique and gave the user the ability to load data, load reference images, display intensity maps, display spectra mosaics, generate a mask, display the mask, display kspace and save the corresponding spectra, reconstruction, and mask files. Parallelization of the reconstruction algorithm was explored but implementation was ultimately unsuccessful. Future work could consist of integrating this parallelization method, adding intensity overlay functionality and improving aesthetics.
ContributorsLammers, Luke Michael (Author) / Kodibagkar, Vikram (Thesis director) / Hu, Harry (Committee member) / Harrington Bioengineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
Description
Magnetic resonance spectroscopic imaging (MRSI) is a non-invasive technique that offers a unique ability to provide the spatial distribution of relevant biochemical compounds (metabolites). The ‘spectrum’ of information provided by MRSI is used as biomarkers for the differential diagnosis of several diseases such as cancer or neurological disorders. Treatment responsive brain tumors can appear similar to non-responsive tumors on conventional anatomical MR images, earlier in the therapy, leading to a poor prognosis for many patients. Biomarkers such as lactate are particularly of interest in the oncological studies of solid tumors to determine their energy metabolism, blood flow, and hypoxia. Despite the capability of nearly all clinical MRI scanners to perform MRSI only limited integration of MRSI into routine clinical studies has occurred to date. The major challenges affecting its true potential are the inherently long acquisition time, low signal-to-noise (SNR) of the signals, overlapping of spectral lines, or the presence of artifacts. The goal of this dissertation work is to facilitate MRSI in routine clinical studies without affecting the current patient throughput.
In this work, the Compressed Sensing (CS) strategy was used to accelerate conventional Point RESolved Spectroscopy (PRESS) MRSI by sampling well below the Shannon-Nyquist limit. Two undersampling strategies, namely the pseudo-random variable density and a novel a priori method was developed and implemented on a clinical scanner. Prospectively undersampled MRSI data was acquired from patients with various brain-related concerns. Spatial-spectral post-processing and CS reconstruction pipeline was developed for multi-channel undersampled data. The fidelity of the CS-MRSI method was determined by comparing the CS reconstructed data to the fully sampled data. Statistical results showed that the a priori approach maintained high spectral fidelity compared to the fully sampled reference for an 80% reduction in scan time. Next, an improvement to the CS-MRSI reconstruction was achieved by incorporating coil sensitivity maps as support in the iterative process. Further, a CS-MRSI-based fast lactate spectroscopic imaging method was developed and implemented to achieve complete water and fat suppression for accurate spatial localization and quantification of lactate in tumors. In vitro phantoms were developed, and the sequence was tested to determine the efficacy of CS-MRSI for low SNR signals, the efficacy of the CS acceleration was determined with statistical analysis.
ContributorsBikkamane Jayadev, Nutandev (Author) / Kodibagkar, Vikram (Thesis advisor) / Chang, John (Committee member) / Robison, Ryan (Committee member) / Smith, Barbara (Committee member) / Sohn, Sung-Min (Committee member) / Arizona State University (Publisher)
Created2021