Ultra-fast magnetic resonance imaging for small animal models
Abstract
[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] Magnetic Resonance Imaging (MRI) has been widely used in clinical diagnoses since 1980s. The scan speed of MRI is crucial based on several reasons. First, only around 25,000 clinical MRI scanners are in service worldwide in 2013, which makes MRI resource tremendously precious. The reduction of the scan time means more patients can be benefited and more scientific progress can be made. Second, the acceleration of the scan speed of MRI allows the examination of the rapid physiological events, such as the heart movement, the brain activity, the blood flow and the perfusion of a drug tracer. Finally, MRI is often used on small animal models for the medical research and the drug investigations. Accelerating the imaging speed can save time, decrease burdens on animals, reduce research coast and improve research outcome. In this thesis, five chapters are included with the focus of ultra-fast MRI for small animal models. A background introduction and technique overview is included in Chapter I, followed by four chapters each with a self-contained study In Chapter II, a fast and quantitative MRI method is developed on a 7 Tesla MRI scanner. The method enabled to collect high quality and high frame rate video in the rat heart. Hence, accurate measurements allowed the detection of the very early cardiac functional impairment in the diabetic and hypertensive heart in rats. In Chapter III, We applied the method developed in Chapter II to investigate the impact of high salt diet on the cardiac functions in a female hypertensive rat model and a female health rat model. The results show that the health rats with high salt intake developed hypertrophy, impaired left ventricular relaxation and lowered cardiac output. Moreover, the hypertensive Ren2 rat model is more vulnerable to the high salt diet. The cardiac dysfunction of Ren2 rats fed with high salt diet progressed into a more server stage with both diastolic and systolic dysfunctions. The MRI results were confirmed with autopsy and histopathology. In Chapter IV, a novel technique, golden-angle ultra-short echo time sequence with ultra-fast acquisition and compressed sensing reconstruction is developed. The method successfully removed the flow turbulence artifacts in the mouse cardiac images at the early diastolic period. In this study, the ultra-short echo time sequence is further improved with golden-angle acquisition. Moreover, in order to accelerate this advantageous pulse sequence, a cutting-edge technology proposed in recent five years called compressed sensing is implemented. As a result, even for a mouse with its heart rate faster than 440 beat per minute, a clean cardiac video can be achieved in less than a minute while the turbulence artifact is removed, using the golden-angle ultra-short echo time sequence combined with compressed sensing technology. In MRI, it usually takes significantly longer time to acquire a better image with higher contrast noise ratio. In Chapter V, a new method, compressed sensing k-space weighting (CSKW), is proposed to acquire the high contrast noise ratio images without additional scan time. This method exploits the fact that compressed sensing technology can preserve image quality while accelerating scan time. Instead of conventionally repeating MRI scan, an acquisition scheme is designed to utilize the advantage of compressed sensing technique. Several MRI scans on animals and phantoms are tested. Using a similar scan time, this new method increases the contrast noise ratio by over 40% or over 100% depending on the scanned subject or organs in mice. Moreover, this technique does not depend on the types of pulse sequence and study. Brain images, kidney images and tumor images as well as different pulse sequences are tested and their results also demonstrated the effectiveness of the new method. To summarize, the importance of accelerating the scan speed in tracking the fast cardiac movement is demonstrated in the studies in Chapter II and III. A concise and robust method of time shortening pulse sequence is proposed and tested in Chapter II and III. In Chapter IV, a powerful new technique called compressed sensing is used to accelerate a novel sequence to tackle the problem with turbulence artifact in cardiac images caused by the fast heart rate of small animals. In Chapter V, a novel method is proposed to enhance the contrast noise ratio of MRI without additional scan time.
Degree
Ph. D.
Thesis Department
Rights
Access is limited to the campuses of the University of Missouri.