Stroke is a devastating disease that is caused by a lack of perfusion to the brain, and is the third leading cause of death in the US. Arterial spin labeling (ASL) is an MRI based method for noninvasive imaging of perfusion, using no radiation and no injections, and holds tremendous potential for aiding in the diagnosis and management of stroke. The overall objective of this project is to develop two new ASL methods for the evaluation of cerebrovascular diseases, with stroke as the primary focus. The first is Velocity Selective ASL (VSASL), which addresses the inability of conventional ASL techniques to measure cerebral perfusion in the presence of collateral and/or slow flow, a condition that is often present in stroke patients. In conventional ASL, radiofrequency pulses magnetically tag arterial blood in major arteries leading to the brain. Under conditions of slow or collateral flow, the tag, which decays away with a half life of approximately one second, can dissipate before it can be delivered to the target tissue, and can give a false reading of no perfusion. In VSASL, arterial blood is tagged based purely on flow velocity, rather than location, allowing for tagging in all locations simultaneously. VSASL is currently the only ASL technique that can in principle give accurate measures of perfusion under these conditions.
Our specific aims related to bringing VSASL into clinical practice are: 1) Development of a robust 3D image acquisition method for VSASL; 2) Development of a direction independent VSASL tagging method; and 3) Optimization of VSASL for the detection of slow flow. The result of this work will be a robust ASL method for quantitative perfusion imaging in stroke. The second method is Vascular Source Imaging (VSI), which allows for the identification of the vascular source that supplies blood to the tissue of interest. VSI is a new technique with a broad range of potential clinical applications including the evaluation of carotid stenosis, risk assessment for stroke, and the identification of tumor blood supplies. VSI is early in it's development, and our aims in this area are to address two basic technical issues: 1) Optimize VSI tagging parameters based on pseudo-continuous labeling; and 2) Develop efficient VSI encoding schemes to maximize the SNR of the VSI measurement and allow for simultaneous encoding of multiple arteries. The result of this work will be robust methods for measuring not only the amount of perfusion, but also the vascular source from which that perfusion was derived. ? ? ?
Showing the most recent 10 out of 17 publications
