Alzheimer's disease, bio-markers, and the role of classification techniques in early diagnosis from neuro-images — An analysis

Various classification methods have been proposed for automatically detecting AD patients from normal patients based on features extracted from MRI and PET neuro-images. The results have achieved high accuracies up to 87% sensitivity and 95% specificity [5]–[21]. But their sensitivity in the detection of patients with Mild Cognitive Impairment (MCI) from normal patients is very lower in the range 60%–70% [5]–[21]. This paper presents a review on the various invasive and non-invasive biomarkers related to AD patients and the classification techniques used in the early diagnosis of Alzheimer's disease. It highlights the factors that affect early AD detection. It concludes that further research on bio-markers is necessary for early AD detection. Optimization of classification techniques can help improve the classification accuracy to some extent. Combining the features obtained from different neuro-images including functional and structural MRI and PET images and other biomarkers can help to improve the prediction rate of MCI from normal patients.