Weakly Supervised Classification of Vital Sign Alerts as Real or Artifact

TL;DR

This paper presents a weakly supervised machine learning approach to classify vital sign alerts as real or artifact, reducing the need for extensive manual labeling and addressing alarm fatigue in healthcare.

Contribution

It introduces a weak supervision method using multiple heuristics to automatically label data, achieving competitive performance with less expert involvement.

Findings

Weak supervision models perform comparably to supervised models.

Reduced need for manual data labeling in healthcare ML.

Potential to mitigate alarm fatigue by improving alert classification.

Abstract

A significant proportion of clinical physiologic monitoring alarms are false. This often leads to alarm fatigue in clinical personnel, inevitably compromising patient safety. To combat this issue, researchers have attempted to build Machine Learning (ML) models capable of accurately adjudicating Vital Sign (VS) alerts raised at the bedside of hemodynamically monitored patients as real or artifact. Previous studies have utilized supervised ML techniques that require substantial amounts of hand-labeled data. However, manually harvesting such data can be costly, time-consuming, and mundane, and is a key factor limiting the widespread adoption of ML in healthcare (HC). Instead, we explore the use of multiple, individually imperfect heuristics to automatically assign probabilistic labels to unlabeled training data using weak supervision. Our weakly supervised models perform competitively with traditional supervised techniques and require less involvement from domain experts, demonstrating their use as efficient and practical alternatives to supervised learning in HC applications of ML.