Timeseries biomarkers clustering for Alzheimer’s Disease progression

Abstract

Neurodegenerative diseases are complex and highly time-dependent diseases. Among them, the most common is Alzheimer’s Disease (AD), in which the patient goes through a series of symptomatic stages before receiving the diagnosis of dementia caused by AD. Due to its temporal characteristics, it is necessary to study the biomarkers associated with the AD from a time series point of view. In this work, we have applied to the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort the Dynamic Time Warping (DTW) technique combined with hierarchical clustering. We extensively applied this technique to several datasets: unidimensional (only one biomarker) and multidimensional (two or more biomarkers) datasets. The results obtained with both dataset types corresponded very clearly with the expected clinical outcomes. The work presented here raises the enormous potential of time series clustering to discover new knowledge in time-dependent diseases such as AD.