Keywords: machine learning for health, temporal clustering, patient outcome prediction, medical time series
TL;DR: We trained different time series model to embed medical time series data from mechanical ventilation episodes, and then we clustered these to uncover hidden patient subtypes in the data.
Abstract: The advancement of Electronic Health Records (EHRs) and machine learning have enabled a data-driven and personalised approach to healthcare. One step in this direction is to uncover patient sub-types with similar disease trajectories in a heterogeneous population. This is especially important in the context of mechanical ventilation in intensive care, where mortality is high and there is no consensus on treatment. In this work, we present an approach to clustering mechanical ventilation episodes, using a multi-task combination of supervised, self-supervised and unsupervised learning techniques. Our dynamic clustering assignment is guided to reflect the phenotype, trajectory and outcomes of the patient. Experimentation on a real-world dataset is encouraging, and we hope that this could translate into actionable insights in guiding future clinical research.