Abstract
Longitudinal data provide a powerful source of information for tracking disease progression over time; yet, identifying early signs of prodromal symptoms remains a significant challenge. This paper introduces LongitProgression, a Python software tool providing computer scientists and physicians with an effective tool for longitudinal cluster analysis. It combines the k-means clustering technique with time-series analysis to account for the temporal nature of medical data and uncover latent behaviour patterns. It also provides preprocessing, visualization, and statistical tools, enabling researchers to explore and interpret complex multi-dimensional datasets. The software is publicly accessible to data scientists and domain experts, with a user-friendly interface and comprehensive documentation. LongitProgression has been successfully employed in diverse scientific papers, underscoring its efficacy and versatility as a valuable tool for longitudinal studies.