Semi-supervised Clustering Through Representation Learning of Large-scale EHR Data

TL;DR

This paper introduces SCORE, a semi-supervised learning framework for EHR data that improves patient phenotyping and disease prediction by leveraging large-scale, high-dimensional, and partially labeled health records.

Contribution

SCORE combines a novel Poisson-Adapted Latent factor Mixture model with a hybrid EM and GVA algorithm, providing theoretical guarantees and improved accuracy over existing methods.

Findings

SCORE outperforms existing methods in simulations.

Increases accuracy with limited labeled data.

Produces more informative patient embeddings for MS prediction.

Abstract

Electronic Health Records (EHR) offer rich real-world data for personalized medicine, providing insights into disease progression, treatment responses, and patient outcomes. However, their sparsity, heterogeneity, and high dimensionality make them difficult to model, while the lack of standardized ground truth further complicates predictive modeling. To address these challenges, we propose SCORE, a semi-supervised representation learning framework that captures multi-domain disease profiles through patient embeddings. SCORE employs a Poisson-Adapted Latent factor Mixture (PALM) Model with pre-trained code embeddings to characterize codified features and extract meaningful patient phenotypes and embeddings. To handle the computational challenges of large-scale data, it introduces a hybrid Expectation-Maximization (EM) and Gaussian Variational Approximation (GVA) algorithm, leveraging limited labeled data to refine estimates on a vast pool of unlabeled samples. We theoretically establish the convergence of this hybrid approach, quantify GVA errors, and derive SCORE's error rate under diverging embedding dimensions. Our analysis shows that incorporating unlabeled data enhances accuracy and reduces sensitivity to label scarcity. Extensive simulations confirm SCORE's superior finite-sample performance over existing methods. Finally, we apply SCORE to predict disability status for patients with multiple sclerosis (MS) using partially labeled EHR data, demonstrating that it produces more informative and predictive patient embeddings for multiple MS-related conditions compared to existing approaches.