deletions | additions       diff --git a/Cerebral_metabolic_rate_of_glucose__.tex b/Cerebral_metabolic_rate_of_glucose__.tex  index c8689de..d457249 100644  --- a/Cerebral_metabolic_rate_of_glucose__.tex  +++ b/Cerebral_metabolic_rate_of_glucose__.tex 
...
The glucose metabolism measures with FDG-PET scan have been used with good accuracy to distinguish AD from both NCs and other dementias \cite{Mosconi_2007,Ballard_2011}.  FDG-PET has also been used to predict MCI to AD converters with better accuracy than structural MRI \cite{Yuan_2008}.  Longitudinal structural MRIs have been widely used studied  as an imaging biomarker for AD. At present, brain atrophy is measured from the high-resolution acquisitions with MRI scanners of 1.5T or 3T magnets.  The best established and validated atrophy assessment methods are based on T1-weighted MRIs \cite{Frisoni2010}.  Progressive death of neurons or neurodegeneration leads to structural changes in the brain which can be observed in strucutral MRI.  Structural changes seen in MRI correlates well with the cognitive impairment \cite{Jack_2013}.  By this time, A$\beta$ abnormality is already saturated.  Thus for monitoring the impact of disease modifying drugs, the ability to track and predict structural changes in MRI can play an important role.  Figure \ref{fig:structuralMRI} shows an example of the changes seen in the brain structure of an AD patient from the patient's two brain scans acquired two years apart.  In this work the figure,  we will lay can see hippocampal volume loss, medial temporal lobe atrophy and ventricular expansion.  Several algorithms performing brain morphometry from longitudinal MRIs have been proposed in  the foundations literature to be used as a biomarker  of AD progression.  Since more and more atrophy estimation algorithms are being used to track volume changes in longitudinal MRIs, it becomes imperative to validate and evaluate these atrophy estimation algorithms.  In this work we build a foundation towards developing a framework for a comprehensive  modeling and simulation of longitudinal structural  MRIs for of  AD patients. One of the immediate benefits of this framework is in evaluating and validating large number of atrophy/volume measurement algorithms existing in the literature.  Large database of ground truth synthetic structural MRIs that can be generated from the proposed framework could also be used to train machine learning algorithms that intend to compute local volume changes.  In the long run, we hope that researchers will build upon this foundational framework to develop more complex models for predicting and simulating patient-specific structural changes in the brain with AD.