With the increase in the number of identified rare diseases and the intricacy involved in diagnosis, as exemplified by metabolic brain diseases, the need for computerized diagnostic systems is inevitable. We propose a pilot computer-assisted medical decision support system (mCAD) which tries to identify and further categorize these diseases, utilizing the information available from magnetic resonance spectroscopy (MRS) and diffusion-weighted imaging (DWI). In this study, we have utilized wavelets, fuzzy relational classifiers and a collection of signal/image processing routines to extract and to classify disease features. The combined MRS+ DWI system achieved a sensitivity (Se) and positive predictivity (PP) of 65.00% and 72.22%, respectively, in detecting seven categories of metabolic brain diseases. The combined MRS+ DWI system exhibits a 10% and 3.47% increase in Se and PP, respectively, in comparison to the system using only DWI information. It also increases the Se and PP of the system using only the MRS information by 15% and 22.22%, respectively.