Volumetric Quantitative Analysis of Receptor Function

In neurological disorders such as Parkinson's disease, examination of the living brain by high-resolution positron emission tomography (PET), combined with the appropriate pharmacokinetic and physiological analyses, can provide valuable quantitative information of altered brain function. The primary objective of the thesis is accomplished by developing a volumetric process for the quantitative analysis of receptor function in a primate model of Parkinson's disease. This method of analysis can be used to more accurately quantify the neuropathological and neurochemical alterations observed in Parkinson's disease and can thus be used to gain a greater understanding of the disease. Fusion of PET images and magnetic resonance (MR) images provides the necessary step for combining the functional information obtained from PET with the anatomical information obtained from MR. The benefit of having fused images is that it enables the visualization of functional PET data by superimposing it on a high-resolution anatomical MR image and therefore improves localization of radioligand-receptor binding and metabolic activity in the brain. When the living brain is examined through PET and magnetic resonance imaging (MRI) with this new technique and the data is combined with the appropriate pharmacokinetic compartmental analyses, the technique provides a critical tool for use in the analysis of brain function and its underlying neural network. A secondary goal of extending the analysis using more optimal statistical reconstruction techniques for PET data, such as Bayesian MAP, is also accomplished. The image data reconstructed using statistical reconstruction routines more optimally model the physics of the imaging process. Parametric images of experimental data are generated using the DIRECT algorithm. Results are compared to results from standard compartmental model analyses. Good agreement with standard compartmental model fits is obtained, demonstrating the power of the method. Thesis Supervisor: Anna-Liisa Brownell Title: Associate Professor of Radiology, Harvard Medical School