Within the next 50 years Alzheimer's disease is expected to affect 100 million people worldwide. The progressive decline in the mental health of the patient is caused by severe brain atrophy generated by the breakdown and aggregation of proteins, resulting in β-amyloid plaques and neurofibrillary tangles. The greatest challenge to Alzheimer's disease lies in the pursuit of an early and definitive diagnosis, in order that suitable treatment can be administered. At the present time, definitive diagnosis is restricted to post-mortem examination. Alzheimer's disease also remains without a long-term cure. This research demonstrates the potential role of Raman spectroscopy, combined with principle components analysis (PCA), as a diagnostic method. Analyses of ethically approved ex vivo post-mortem brain tissues (originating from frontal and occipital lobes) from control (3 normal elderly subjects and 3 Huntingdon's disease subjects) and Alzheimer's disease (12 subjects) brain sections, and a further set of 12 blinded samples are presented. Spectra originating from these tissues are highly reproducible, and initial results indicate a vital difference in protein content and conformation, relating to the abnormally high levels of aggregated proteins in the diseased tissues. Further examination of these spectra using PCA allows for the separation of control from diseased tissues. The validation of the PCA models using blinded samples also displays promise for the identification of Alzheimer's disease, in conjunction with secondary information regarding other brain diseases and dementias. These results provide a route for Raman spectroscopy as a possible non-invasive, non-destructive tool for the early diagnosis of Alzheimer's disease.