sureLDA: A Multi-Disease Automated Phenotyping Method for the Electronic Health Record

Abstract

ObjectiveA major bottleneck hindering utilization of electronic health record (EHR) data for translational research is the lack of precise phenotype labels. Chart review as well as rule-based and supervised phenotyping approaches require laborious expert input, hampering applicability to studies that require many phenotypes to be defined and labeled de novo. Though ICD codes are often used as surrogates for true labels in this setting, these sometimes suffer from poor specificity. We propose a fully automated topic modeling algorithm to simultaneously annotate multiple phenotypes. MethodssureLDA is a label-free multidimensional phenotyping method. It first uses the PheNorm algorithm to initialize probabilities based on two surrogate features for each target phenotype, and then leverages these probabilities to constrain the Latent Dirichlet Allocation (LDA) topic model to generate phenotype-specific topics. Finally, it combines phenotype-feature counts with surrogates via clustering ensemble to yield final phenotype probabilities. ResultssureLDA achieves reliably high accuracy and precision across a range of simulated and real-world phenotypes. Its performance is robust to phenotype prevalence and relative informativeness of surogate versus non-surrogate features. It also exhibits powerful feature selection properties. DiscussionsureLDA combines attractive properties of PheNorm and LDA to achieve high accuracy and precision robust to diverse phenotype characteristics. It offers particular improvement for phenotypes insufficiently captured by a few surrogate features. Moreover, sureLDAs feature selection ability enables it to handle high feature dimensions and produce interpretable computational phenotypes. ConclusionsureLDA is well suited toward large-scale EHR phenotyping for highly multi-phenotype applications such as PheWAS.