Abstract Details

Title Network Diffusion Model Predicts Future Dementia Patterns

Topic Aging and Dementia

Presentation(s) P06 - (-)

Poster/Presentation
Number 021

Objective

Background BACKGROUND: A novel network diffusion model capturing prion-like mechanistic model of neurodegenerative progression in the brain was recently proposed (Raj 2012). The model predicts distinct spatial patterns of disease called "eigen-modes" which recapitulate the atrophy in Alzheimers and other dementias. Here we apply this model to predict future atrophy of individuals using baseline MRI.

Design/Methods DESIGN/METHODS: Baseline and end-of-study atrophy and its rate of change were obtained using ADNI longitudinal MRI Freesurfer volumetrics from 687 normal, AD and MCI subjects. Future atrophy patterns was predicted for each subject by applying the time-evolution equation y(t) = exp(-beta L y(o)) of the network diffusion model to connectivity networks obtained from 14 healthy brains via 86-region Freesurfer atlas-based parcellation and tractography. Measured and predicted atrophy patterns at the end of study were correlated for all individuals and regions. Another experiment correlated baseline atrophy and its rate of change, using proposed and 2 other plausible non-network models - exponential and Sigmoid model based on Jack et al, Lancet 2010.

Results RESULTS: Pearson correlation of measured and predicted atrophy at the end of study was: R=0.93, p<0.00001. This compares favorably to the raw correlation between baseline and end atrophy, which gave R of 0.90 (MCI) and 0.86 (AD). Pearson correlation between baseline and rate of change of atrophy was: R = 0.65 (network diffusion), R=0.19 (exponential) and R=0.31 (Sigmoid).

Conclusions CONCLUSIONS: Striking agreement with longitudinal regional atrophy supports the network diffusion model's ability to predict future atrophy patterns of individuals. Head-to-head comparison with other plausible models of disease spread showed an order of magnitude improvement using the network diffusion model. Successful predictability of future dementia patterns can greatly impact patient care, prognosis and therapy monitoring.

Authors/Disclosures
Ashish Raj PRESENTER	No disclosure on file
	No disclosure on file
Amy Kuceyeski, PhD (Weill Cornell Medical College, Radiology)	An immediate family member of Dr. Kuceyeski has received personal compensation for serving as an employee of Heyer Physical Therapy. The institution of Dr. Kuceyeski has received research support from National Institutes of Health. Dr. Kuceyeski has received personal compensation in the range of $500-$4,999 for serving as a Grant Panel Member with Department of Defense.
Lauren B. Krupp, MD, FAAN (NYU Langone Medical Center)	Dr. krupp has received personal compensation in the range of $500-$4,999 for serving as a Consultant for Bristol Myers Squibb. Dr. krupp has received personal compensation in the range of $500-$4,999 for serving as a Consultant for Celgene. Dr. krupp has received personal compensation in the range of $500-$4,999 for serving as a Consultant for Medscape. Dr. krupp has received personal compensation in the range of $500-$4,999 for serving as a Consultant for EBIX. Dr. krupp has received personal compensation in the range of $500-$4,999 for serving on a Scientific Advisory or Data Safety Monitoring board for Biogen. Dr. krupp has received personal compensation in the range of $500-$4,999 for serving on a Scientific Advisory or Data Safety Monitoring board for Hoffman LaRoche. Dr. krupp has received personal compensation in the range of $5,000-$9,999 for serving as an Expert Witness for MMMK. Dr. krupp has received personal compensation in the range of $500-$4,999 for serving as an Expert Witness for Patrick, Dolan, and Kaufman. Dr. krupp has received intellectual property interests from a discovery or technology relating to health care.

��ɫ��

��ɫ��