Characterization of Drug Use Patterns Using Process Mining and Temporal Abstraction Digital Phenotyping

Abstract

Understanding and identifying executed patterns, activitiesand processes for patients of different characteristics provides medicalexperts a deep understanding of which tasks are critical in the providedcare, and may help identify ways to improve them. However, extractingthese events and data for patients with complex clinical phenotypes isnot a trivial task. This paper provides an approach to identifying specificpatient cohorts based on complex digital phenotypes as a starting pointto apply process mining tools and techniques and identify patterns orprocess models. Using temporal abstraction-based digital phenotypingand pattern matching, we identified a cohort of patients with sepsis fromthe MIMIC II database, and then apply process mining techniques todiscover medication use patterns. In the case study we present, the useof temporal abstraction digital phenotyping helped us discover a relevantpatient cohort, aiding in the extraction of the data required to generatedrug use patterns for medications of different types such as vasopressors,vasodilators and systemic antibacterial antibiotics. For sepsis patients,combining the use of temporal abstraction digital phenotyping and process mining tools and techniques, was proven to help extract accuratecohorts of patients for health care process mining.