Identifying TBI Physiological States by Clustering Multivariate Clinical Time-Series Data

TL;DR

This paper introduces SLAC-Time, a self-supervised clustering method that preserves data integrity in multivariate time series, enabling the identification of distinct TBI physiological states without imputation or aggregation.

Contribution

The study applies SLAC-Time to cluster TBI patient data, revealing three physiological states and their features, while incorporating clinical validation and insights into state transitions.

Findings

Identified three distinct TBI physiological states.

Validated states with clinical expert input.

Linked clinical events to state transitions.

Abstract

Determining clinically relevant physiological states from multivariate time series data with missing values is essential for providing appropriate treatment for acute conditions such as Traumatic Brain Injury (TBI), respiratory failure, and heart failure. Utilizing non-temporal clustering or data imputation and aggregation techniques may lead to loss of valuable information and biased analyses. In our study, we apply the SLAC-Time algorithm, an innovative self-supervision-based approach that maintains data integrity by avoiding imputation or aggregation, offering a more useful representation of acute patient states. By using SLAC-Time to cluster data in a large research dataset, we identified three distinct TBI physiological states and their specific feature profiles. We employed various clustering evaluation metrics and incorporated input from a clinical domain expert to validate and interpret the identified physiological states. Further, we discovered how specific clinical events and interventions can influence patient states and state transitions.