Identifying temporal patterns in the progression of neurodegenerative disease using unsupervised clustering

Abstract

AbstractObjectivesOne of the principal goals of Precision Medicine is to stratify patients by accounting for individual variability. However, extracting meaningful information from Real-World Data, such as Electronic Health Records, still remains challenging due to methodological and computational issues.Materials & methodsA Dynamic Time Warping-based unsupervised-clustering methodology is presented in this paper for the clustering of patient trajectories of multi-modal health data on the basis of shared temporal characteristics. Building on an earlier methodology, a new dimension of time-varying numerical clinical and imaging features (six in total) is incorporated, through an adapted cost-minimization algorithm for clustering on different, possibly overlapping, feature subsets. A cluster evaluation process is also implemented, by admitting two user-defined parameters (granularity threshold and feature contribution). The model disease chosen is Huntington’s disease (HD), characterized by progressive neurodegeneration.ResultsFrom a wide range of examined user-defined parameters, four case examples are highlighted to exemplify the combined effects of feature weights and granularity threshold in the stratification of HD trajectories in homogeneous clusters. For each identified cluster, polynomial fits that describe the temporal behavior of the assessed features are provided for an informative comparison, together with their averaged values.DiscussionThe proposed data-mining methodology permits the stratification of distinct time patterns of multi-modal health data in individuals that share a diagnosis or future diagnosis, employing user-customized criteria beyond the current clinical practice.ConclusionsThis work bears implications for better analysis of individual variability in disease progression, opening doors to personalized preventative, diagnostic and therapeutic strategies.