For Alzheimer's disease (AD), the most common neurodegenerative disease, there is no simple, cost-effective biomarker for disease identification. Using novel mass spectrometry (MS)-based techniques, and analysis of the albumin-enriched low molecular weight proteome, minute amounts of human serum were analyzed for the measurement of thousands of peptides and proteins in parallel. The mass spectrograms were then evaluated with a novel computer algorithm to identify spectral peaks that discriminate between samples from patients with and without AD. There are four peaks that distinguish AD from control subjects and AD subjects from those with Parkinson's disease (PD). Additionally, after analyzing data from a recently published study of AD and control subjects, we found three discriminating peaks in common with the four from our patient serum samples. The identification of these peptides/proteins, and their direct measurement in patient serum, may allow the development of a simple, cost-effective test for AD.