Use of noisy labels as weak learners to identify incompletely ascertainable outcomes: A Feasibility study with opioid-induced respiratory depression

• DOI: 10.1016/j.heliyon.2024.e26434
• Publication Date: 2024-02-16T17:54:38Z
• Authors: Alvin D. Jeffery; Daniel Fabbri; Ruth M. Reeves; Michael E. Matheny
• Keywords: Depression

Assigning outcome labels to large observational data sets in a timely and accurate manner, particularly when outcomes are rare or not directly ascertainable, remains significant challenge within biomedical informatics. We examined whether noisy generated from subject matter experts' heuristics using heterogenous types programming paradigm could provide large, set. chose the clinical condition of opioid-induced respiratory depression for our use case because it is rare, has no administrative codes easily identify condition, typically requires at least some unstructured text ascertain its presence.