| Magnetic Resonance Angiography (MRA) has been widely used as a noninvasive approach to aid diagnosis of vascular diseases. Quantification of vessel dimension and signal to noise ration (SNR) from high resolution three-dimensional (3D) MRA images has been hindered by a lack of a robust frame of reference and degradation of image quality due to motion artifacts.; In the first part of this dissertation, an automatic centerline/bifurcation extraction algorithm is presented to produce well-centered vessel skeletons for simulated 3D vessels and two real MRA image volumes. For the simulated data, the root mean square error for centerline detection is about half a voxel. For the human intracranial MRA data, the best case in bifurcation detection gave a sensitivity of 91.4%, a positive predictive value of 91.4%, and a position RMS error of 1.7 voxels.; Motion artifacts can seriously degrade the quality of MRA images. In the second part, a novel average-specific phase encoding (PE) ordering scheme is applied for the turbo-spin-echo (TSE) sequence to automatically detect motion-corrupted PE views and suppress motion artifacts. Application of this method to phantom and human studies has demonstrated significant reduction of motion artifacts in the reconstructed images. Simulation has shown that for motions in the frequency encoding direction, if the number of corrupted echo trains is less than 7, the detection failure rate can be kept below 10%. For motions in the PE direction, the corrupted trains should be less than 4 to have a failure rate less than 10%.; T2-weighted carotid artery images acquired using the TSE sequence often suffer from motion artifacts caused by vessel and neck skin pulsation. In the third part, a fast, half-Fourier single-shot turbo-spin-echo (HASTE) sequence is used to acquire carotid artery images. The combination of HASTE with parallel acquisition, T2 decay compensation produced carotid artery images without motion artifacts. However, the image sharpness is a little inferior to that of the TSE images. Further work needs to be done to improve the image sharpness for better visualization of the carotid artery wall. |
