From PET and SPECT Development, Towards Imaging Alzheimer's
Date
Featured Speaker:
Sepideh Shokouhi, PhD
Instructor, Vanderbilt University Institute of Imaging Science
Department of Radiology and Radiological Sciences
Vanderbilt University, Nashville, TN
Abstract:
We are developing new functional, structural and pathological imaging biomarkers of Alzheimer's disease (AD) in both human subjects from the ADNI (Alzheimer's Disease Neuroimaging Initiative) database and transgenic mouse models using 18-FDG-PET and 11C-PIB-PET.
One of current ADNI studies is focused on evaluating longitudinal changes in brain glucose metabolism based on a within-subject correlation analysis of FDG-PET images to capture progressive heterogeneous metabolic changes and their association to the subject's changes in cognitive status. In another ADNI study we are modeling longitudinal changes of PIB-PET images with n-point correlation functions. Statistical descriptors such as n-point correlation functions (Sn) have been used in astronomy and materials science to provide detailed characterization of patterns in spatial data. Sn measures the stochastic relationship between the image voxel values at different distances, typically referred to as lumpiness or flocculence. Our objective is to evaluate the feasibility of translating these methods into PIB-PET imaging to characterize longitudinal amyloid beta plaque accumulation in prodromal stages of AD, which could be difficult to detect due to low image resolution and contrast-to-noise ratio. We are also interested in developing stochastic object models of Amyloid beta plaque distribution and progression in transgenic mouse brain and incorporating them into in-vivo imaging using optimization techniques, such as simulated annealing.
Please note that Sepideh Shokouhi, PhD is an applicant for the Biomarker Imaging faculty position.
Type
Time Duration
Featured Speaker:
Sepideh Shokouhi, PhD
Instructor, Vanderbilt University Institute of Imaging Science
Department of Radiology and Radiological Sciences
Vanderbilt University, Nashville, TN
Abstract:
We are developing new functional, structural and pathological imaging biomarkers of Alzheimer's disease (AD) in both human subjects from the ADNI (Alzheimer's Disease Neuroimaging Initiative) database and transgenic mouse models using 18-FDG-PET and 11C-PIB-PET.
One of current ADNI studies is focused on evaluating longitudinal changes in brain glucose metabolism based on a within-subject correlation analysis of FDG-PET images to capture progressive heterogeneous metabolic changes and their association to the subject's changes in cognitive status. In another ADNI study we are modeling longitudinal changes of PIB-PET images with n-point correlation functions. Statistical descriptors such as n-point correlation functions (Sn) have been used in astronomy and materials science to provide detailed characterization of patterns in spatial data. Sn measures the stochastic relationship between the image voxel values at different distances, typically referred to as lumpiness or flocculence. Our objective is to evaluate the feasibility of translating these methods into PIB-PET imaging to characterize longitudinal amyloid beta plaque accumulation in prodromal stages of AD, which could be difficult to detect due to low image resolution and contrast-to-noise ratio. We are also interested in developing stochastic object models of Amyloid beta plaque distribution and progression in transgenic mouse brain and incorporating them into in-vivo imaging using optimization techniques, such as simulated annealing.
Please note that Sepideh Shokouhi, PhD is an applicant for the Biomarker Imaging faculty position.