 This paper proposes a novel multivariate multi-scale methodology for analyzing complex physiological signals. It combines permutation entropy with a multi-scale framework in order to simultaneously analyze multiple channels of data. This approach is faster than traditional methods and is more robust to noise and artifacts. It was tested on a data set of brain activity from Alzheimer's disease patients and mild cognitive impairment subjects compared to healthy elderly individuals. The results suggest that the proposed methodology could be used to differentiate between these two groups. This article was authored by Isabella Palamara, Giuseppe Moribito, Alessia Bramanti, and others.