Figure 2.

PCA scores plots representing each single class model for the training set (90 total samples). CN, cognitively normal subjects (A, green squares); SMD, subjective memory decline (B, blue squares); MCI, mild cognitive impairment (C, purple squares); and AD, Alzheimer’s disease (red squares). All class-models indicate that the samples are homogeneous as no outliers were detected. The labels t[1] and t[2] along the axes represent the scores (the first 2 partial least-squares components) of the model, which are sufficient to build a satisfactory classification model.