Gaussian Process Regression and Classification using International Classification of Disease Codes as Covariates

TL;DR

This paper introduces a novel Gaussian process framework utilizing a string kernel for measuring similarity between ICD code sets, enabling improved automated prediction of biomedical responses from electronic health records.

Contribution

It develops a new similarity measure for ICD code sets and integrates it into Gaussian process models for nonparametric regression and classification in healthcare data.

Findings

Outperforms competitors in classifying primary cancer sites.

Provides better sensitivity and specificity in predictions.

Estimates associations between chronic conditions and cancer types.

Abstract

International Classification of Disease (ICD) codes are widely used for encoding diagnoses in electronic health records (EHR). Automated methods have been developed over the years for predicting biomedical responses using EHR that borrow information among diagnostically similar patients. Relatively less attention has been paid to developing patient similarity measures that model the structure of ICD codes and the presence of multiple chronic conditions, where a chronic condition is defined as a set of ICD codes. Motivated by this problem, we first develop a type of string kernel function for measuring similarity between a pair of subsets of ICD codes, which uses the definition of chronic conditions. Second, we extend this similarity measure to define a family of covariance functions on subsets of ICD codes. Using this family, we develop Gaussian process (GP) priors for Bayesian nonparametric regression and classification using diagnoses and other demographic information as covariates. Markov chain Monte Carlo (MCMC) algorithms are used for posterior inference and predictions. The proposed methods are free of any tuning parameters and are well-suited for automated prediction of continuous and categorical biomedical responses that depend on chronic conditions. We evaluate the practical performance of our method on EHR data collected from 1660 patients at the University of Iowa Hospitals and Clinics (UIHC) with six different primary cancer sites. Our method has better sensitivity and specificity than its competitors in classifying different primary cancer sites and estimates the marginal associations between chronic conditions and primary cancer sites.