Description
- Abstract:
- This thesis discusses two topics, both related to the goal of determining tumor blood flow parameters from time-sequenced contrast-enhanced medical imaging data in an effort to quickly measure the efficacy of cancer treatments. This work builds upon existing work in the radiology field by using a commonly accepted two-compartment model. In this model, the blood flow parameters are represented as four constant coefficients: perfusion, permeability surface area product, and relative volumes of the two compartments. For the purposes of the testing the ideas presented in this thesis, computational baseline data is used; although not actual patient data, this baseline was developed to emulate real patient data. The first major topic is a parametric study. Since the application of this work is in the form of an inverse problem, i.e., the coefficients are unknown, it is important to understand the effects of each of the four coefficients on the overall signal intensity of the medical images. Therefore, a set of numerical experiments was designed to explore these effects. The results of these experiments are presented along with a discussion of how the numerical results relate to the physiological system being modeled. The second major topic has two parts. In the first part, questions are raised and explored involving the validity of a model simplification in the existing literature. The second part proposes new methods for computing two of the coefficients; these alternative methods do not require the use of the simplification in question. This simplification arises from a difficulty in measuring the time-dependent outflow of contrast agent. The assumption leading to this simplification is explored, and the existence of the error is verified through calculations. The beginning stages of alternative methods that do not require the use of the erroneous simplification are also presented. Two of the four coefficients are recovered with high accuracy through numerical experiments. A method for determining the missing boundary condition is also presented. To demonstrate the robustness of these methods in the presence of low image frequency, only subsets of baseline analytical data are passed as input to the computations.
- Notes:
- Thesis (Ph.D.) -- Brown University (2008)
Access Conditions
- Rights
- In Copyright
- Restrictions on Use
- Collection is open for research.
Citation
Libertini, Jessica Meade,
"Determining Tumor Blood Flow Parameters Using Dynamic Imaging Data"
(2008).
Applied Mathematics Theses and Dissertations.
Brown Digital Repository. Brown University Library.
https://doi.org/10.7301/Z02R3PZ7
Relations
Collection:
-
Applied Mathematics Theses and Dissertations
Theses and Dissertations for the Applied Mathematics department....