Abstract
MotivationTemporal biomarker discovery in longitudinal data is based on detecting reoccurring trajectories, the so-called shapelets. The search for shapelets requires considering all subsequences in the data. While the accompanying issue of multiple testing has been mitigated in previous work, the redundancy and overlap of the detected shapelets results in an a priori unbounded number of highly similar and structurally meaningless shapelets. As a consequence, current temporal biomarker discovery methods are impractical and underpowered.ResultsWe find that the pre- or post-processing of shapelets does not sufficiently increase the power and practical utility. Consequently, we present a novel method for temporal biomarker discovery: Statistically Significant Submodular Subset Shapelet Mining (S5M) that retrieves short subsequences that are (i) occurring in the data, (ii) are statistically significantly associated with the phenotype and (iii) are of manageable quantity while maximizing structural diversity. Structural diversity is achieved by pruning non-representative shapelets via submodular optimization. This increases the statistical power and utility of S5M compared to state-of-the-art approaches on simulated and real-world datasets. For patients admitted to the intensive care unit (ICU) showing signs of severe organ failure, we find temporal patterns in the sequential organ failure assessment score that are associated with in-ICU mortality.Availability and implementation S5M is an option in the python package of S3M: github.com/BorgwardtLab/S3M.
Highlights
In hospitals, critically ill patients are transferred to intensive care units (ICUs) and are subjected to increased intensity of monitoring and care
For patients admitted to the intensive care unit (ICU) showing signs of severe organ failure, we find temporal patterns in the sequential organ failure assessment score that are associated with in-ICU mortality
We discover novel interpretable biomarkers in the sequential organ failure assessment (SOFA) score of patients associated with in-ICU mortality on the eICU database (Pollard et al, 2018)
Summary
Critically ill patients are transferred to intensive care units (ICUs) and are subjected to increased intensity of monitoring and care. The sequential organ failure assessment score (SOFA) describes the severity of a patient’s organ dysfunction where a high score is associated with high in-ICU mortality (Singer et al, 2016; Vincent et al, 1996). The SOFA score is assessed in 24 h intervals or more (Ferreira et al, 2001; Tee et al, 2018). The high resolution of critical care databases, makes more frequent evaluations of the SOFA scores possible. We mine critical care databases for statistically significant temporal patterns to provide additional information and assistance to clinicians in recognizing and interpreting clinical data
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call