Using binary silver labels in electronic health records-based computable phenotyping algorithms

2026-07-20

AI Summary

This work addresses the scarcity of gold-standard phenotype labels in electronic health records and the limited applicability of existing weakly supervised methods—such as PheNorm—to binary silver-standard labels, which are commonly available but underutilized. The authors propose Binary PheNorm, a novel extension that directly incorporates binary silver-standard labels into the PheNorm framework. By employing denoising regression to generate continuous phenotype scores, the method eliminates the need for log-transformation and Gaussian mixture modeling, thereby simplifying the pipeline and enhancing the utilization of binary signals. Binary PheNorm also supports Lasso regularization for high-dimensional features and enables joint modeling of both binary and count-based silver labels. Evaluated on anaphylaxis and acute pancreatitis phenotyping tasks, the approach substantially improves AUC from 0.793 to 0.891–0.892 and from 0.736 to 0.805–0.819, respectively, demonstrating its effectiveness.

Abstract

Gold-standard phenotype labels are often unavailable at scale in electronic health record (EHR) studies because they require manual chart review. Weakly supervised phenotyping methods instead use silver-standard labels, such as diagnosis-code counts, natural language processing (NLP) mentions, medication indicators, or laboratory thresholds. PheNorm is widely used for this purpose, but its original formulation was designed for count-valued silver labels and relies on log transformation, utilization normalization, and Gaussian mixture modeling. These steps are not directly suited to binary silver labels, which are common and may be highly informative. We propose Binary PheNorm, an extension that uses binary silver labels directly in the corruption-and-regression denoising step and produces a continuous phenotype score without EM calibration. We also consider a lasso-regularized version for high-dimensional EHR settings and combined models using both binary and count labels. In simulations, Binary PheNorm achieved strong discrimination using binary labels alone and often improved performance when combined with count labels. In anaphylaxis, AUC increased from 0.793 for an epinephrine-mention indicator to 0.891-0.892 after Binary PheNorm. In acute pancreatitis, AUC increased from 0.736 for a lipase-threshold indicator to 0.805-0.819. These results support Binary PheNorm as a practical weakly supervised approach when informative binary silver labels are available.

Problem

• electronic health records
• computable phenotyping
• silver-standard labels
• binary labels
• weakly supervised learning

Innovation

• Binary PheNorm
• weakly supervised learning
• electronic health records
• computable phenotyping
• silver-standard labels