Date of Award
Summer 2013
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Biomedical Engineering
First Advisor
DeYoe, Edgar A.
Second Advisor
Ropella, Kristina M.
Abstract
The overall goal of this study is to evaluate the existing, and develop new visual field mapping paradigms, which consist of visual stimulation scheme, post-processing and displaying tools using fMRI for both research and clinical applications.
We first directly compared phase mapping and random multifocal mapping paradigms with respect to clinically relevant factors. Multifocal mapping was superior in immunity to noise and was able to accurately decompose the response of single voxels to multiple stimulus locations. In contrast, phase mapping activated more extrastriate visual areas and was more efficient per run in achieving a statistically efficient response in a minimum time but required separate runs to map eccentricity and angle. Multifocal mapping became less efficient as the number of simultaneous stimulus locations increased from 13 to 25 to 49 and when duty cycle increased from 25% to 50%. In sum, each paradigm offers advantages that may be optimal for different applications.
Given the respective advantages and weaknesses of phase-encoded design and random multifocal design, we further developed a novel paradigm by combining the phase-encoded stimuli and one or two isolated random segments. The addition of the random stimuli was shown to have insignificant effect on the retinotopic mapping by the phase-encoded stimuli. Three applications were demonstrated for this combined paradigm: Simultaneous mapping the retinotopy and selected ROIs, automated calibration of the temporal phases, and delineation of the hemodynamic response function for selected voxels.
At present the representation of the visual field by the visual cortex is displayed as a diagram of a subject's visual field with circular symbols placed at locations to which voxels have shown a response. The diagram provides an intuitive way of interpreting the fMRI cortical maps in terms of visual function. However, it provides little information about the relative probability of obtaining a brain response from different locations within the field of view. Therefore, we derived a quantitative form of such a diagram, on which a probability distribution could be drawn. The quantitative diagrams from five subjects showed highly variable patterns of coverage, which made it questionable whether any meaningful probabilistic distribution can be obtained.