To our knowledge, no other existing model can accurately match this strong age dependence observed in prevalence studies in dementia. From the classification and principal components analysis (Figure 7) we conclude that network diffusion eigenmodes are an effective basis for dimensionality reduction of atrophy in dementia, producing even better classification accuracy than the optimal basis identified by PCA. This suggests a possible role for our model in unsupervised, automated, and regionally unbiased Selleckchem NVP-BKM120 differential diagnosis of various dementias. Instead of dealing with high-dimensional

and complex whole-brain atrophy patterns, future neuroradiologists might simply look at the relative contribution of the first three to four eigenmodes in any person’s brain and treat them as clinical biomarkers. This approach could be especially helpful in cases of mixed dementia, where classical region-based atrophy descriptors RG7204 concentration might prove unsatisfactory. However, the most

important clinical application of this model could well be in the prediction of cognitive decline. Starting from baseline MRI volumetrics for estimation of model parameters, the model in Equation 1 can be subsequently used to predict future atrophy of an individual subject. If the measured and predicted “future” atrophy are deemed statistically close, then it would serve to further validate our hypotheses as well as provide a valuable prognostic aid to the clinician. This will allow a neurologist to predict what the patient’s neuroanatomic, and therefore cognitive, state will be at any given point in the future. Knowledge of what the future holds will allow patients to make informed choices regarding their lifestyle and therapeutic interventions. Figures 2, 3, 4, and 5 present an uncanny parallel to recent findings of network degeneration. That brain networks Calpain are altered in neurodegeneration is now established (He

et al., 2008 and Lo et al., 2010). Distinct, nonoverlapping spatial patterns are seen in AD and bvFTD (Zhou et al., 2010 and Du et al., 2007), which Seeley et al. characterized as belonging to the default mode and salience networks, respectively. The relation between dementia and separate intrinsic connectivity networks (ICNs) (Seeley et al., 2009) appears convincing, but the underlying cause remains unexplained. Conjectures regarding selective vulnerability of different functional networks sharing synchronous neural activity, region-specific functional loads, or some as yet unknown structural, metabolic, and physiological aspects of neural network biology were put forth (Saxena and Caroni, 2011). Buckner et al. (2005) conjectured that early metabolic activity in the default network is somehow later implicated in AD progression. Interestingly, our macroscopic diffusion model can explain these findings without requiring any kind of selective vulnerability, regional specificity, or shared functional load.