An information theory based formalism for medical image analysis proposed in Young et al. [Young, K., Chen, Y., Kornak, J., Matson G. B., Schuff, N., 2005. Summarizing Complexity in High Dimensions, Phys. Rev. Lett. 94 098701-1] is described and used to estimate image complexity measures as a means of generating interpretable summary information. An analysis of anatomical brain MRI data exhibiting cortical thinning, currently considered to be a sensitive early biomarker for neurodegenerative diseases, is used to illustrate the method. Though requiring no previous assumptions about the detailed shape of the cortex or other brain structures, the method performed comparably (sensitivity=0.91) to direct cortical thickness estimation techniques (sensitivity=0.93) at separating populations in a data set designed specifically to test the cortical thickness estimation algorithms. The results illustrate that the complexity estimation method, though general, is capable of providing interpretable diagnostic information.