Image Analysis and Classification Protocol for Characterization of Normal and Abnormal Aging via Structural Mri

resolution, FOV of 24 cm, 62.5 kl-Iz acquisition bandwidth, 4ram slice thickness, 17 sagittal slices). Two different memory tasks were employed: A) Identification of in-/outdoor scenes, B) Comparison of left/right picture sizes. A stimulus-block paradigm with 30 seconds on (task A) and 30 seconds off (task B) were repeated for 4 cycles and followed by a 6 rain resting paradigm in a 10-minute-total scan. A high resolution anatomical scan was conducted for image registration. Data Analysis: fMRI voxel time courses were cross-correlated with the stimulus function and a student t-test was performed to localize significantly activated voxels (p < 0.001). These functionally and spatially localized voxels in the entorhinal cortex region were used to guide the calculation of functional synchrony based on the voxel time courses during the 6-rain resting paradigm (1). Results and Discussion: The designed stimulus paradigms significantly activated the regions of the entorhinal cortex, angular cortex, and dorsolateral prefrontal cortex. An average of 46 voxels in the entorhinal cortex regions of these subjects were activated by the tasks A and B. The voxel time courses at resting state of these localized voxels were cross-correlated. It was found that the functional synchrony in the entorhinal cortex region in the healthy volunteers was similar with those in the hippocampal region [1]. It is suggested that these methods and results could be applied to mild cognitive impairment subjects to determine their possible AD progression. Reference 1. S-J. Li et al., Radiology, 225(1):253-259, 2002.2. J.R. Petrella, et al., Radiology, 226(2):315-336, 2003. Acknowledgments This work is supported by NIH grants AG20279, EB02014, RR00058