Magnetic resonance imaging (MRI) is used widely and frequently in the clinical setting to image and diagnose patients. In addition to the anatomical scans that can be acquired using MRI, different kinds of physiological parameters, such as blood flow, can be obtained by utilizing pulse sequence, scan protocol and post-processing. In many types of clinical applications, the quantification of their respective physiological parameters can provide many benefits for management of patients and treatment decisions. By quantifying parameters using MRI, it gives an advantage over other established imaging modalities, such as computed tomography (CT) and positron emission tomography (PET), that expose patients to radiation. In this thesis, we discuss the quantification of physiological parameters in the workup of glioblastoma multiform (GBM), a deadly malignant tumor in the brain, and acute ischemic stroke. One of the difficulties of designing treatment plans for patients is determining which will do well and which will do poorly. For treating GBMs, it is known that the genetic makeup of the tumor influences the response to treatment. Because currently, invasive methods are used to obtain genetic information, we study the relationship between MR physiological parameters, T1 and T2, and genomics to determine whether genetic information can be inferred from MRI. In the clinical setting, time is an issue, and thus cannot use T1 and T2 calculating scans that take a long time. Therefore, a method of quantifying T1 and T2 from anatomical T1-weighted and T2-weighted scans, which are used commonly in the clinical setting, is presented in this study. Ischemic stroke occurs when blood supply to the brain is interfered by a blockage in a blood vessel. The ability to assess a patient’s cerebral blood flow (CBF) is important in determining the prognosis. In addition to ischemic stroke, CBF plays a large role in other cerebrovascular diseases as well. In this thesis, we study the dynamic susceptibility contrast (DSC) MRI method of calculating absolute CBF and compare against reference standard microsphere deposition method. Another critical factor in the setting ischemic stroke is the growth of infarct in the affected tissue. In major vessel occlusion, compromised blood flow may be sustained by the recruitment of pial collateral blood vessels, and thus alter the outcome of the infarct growth. To study this relationship, we present a method of automatically calculating the infarct volume using diffusing weighted imaging (DWI), and develop a predictive model of infarct volume per time as a function of pial collateral recruitment.

Creator

• Jeong, Yong Ik

DOI

• https://doi.org/10.21985/n2-4zya-8q85

Subject

• Biomedical engineering

Language

• en

Alternate Identifier

• etdadmin_upload_628020
• http://dissertations.umi.com/northwestern:14483

Date created

• 2018-01-01

Resource type

• Dissertation

Rights statement

• In Copyright