Clustering Pseudo Time Series: Exploring Trajectories in the Ageing Process

Abstract

Investigating the natural ageing process typically involves the use of extensive longitudinal datasets that can capture changes associated with the progression of ageing. However, they are often resource-intensive and time-consuming to conduct. Cross-sectional data, on the other hand, provides a snapshot of a population at many different ages and can capture many disease processes but do not incorporate the time dimension. Pseudo time series can be reconstructed from cross sectional data, with the aim to explore dynamic processes (such as the ageing process). In this paper we focus on employing pseudo time series analysis on cross-sectional population data that we constrain using age information to create realistic trajectories of people with different degrees of cardiovascular disease. We then use clustering methods to construct and label trajectory-based phenotypes, aiming to enhance our understanding of ageing and disease progression.

Keywords: ageing; clustering analysis; disease progression; pseudo time series.