Surrogate Assisted Semi-supervised Inference for High Dimensional Risk Prediction

TL;DR

This paper introduces a surrogate-assisted semi-supervised learning method for high-dimensional risk prediction using EHR data, effectively leveraging unlabeled data and surrogates to improve inference and prediction accuracy.

Contribution

The paper develops a novel semi-supervised approach that combines surrogates and high-dimensional predictors for robust risk modeling with valid inference.

Findings

Outperforms existing supervised methods in simulations

Provides valid risk inference even with model mis-specification

Successfully applied to genetic risk prediction of type-2 diabetes

Abstract

Risk modeling with EHR data is challenging due to a lack of direct observations on the disease outcome, and the high dimensionality of the candidate predictors. In this paper, we develop a surrogate assisted semi-supervised-learning (SAS) approach to risk modeling with high dimensional predictors, leveraging a large unlabeled data on candidate predictors and surrogates of outcome, as well as a small labeled data with annotated outcomes. The SAS procedure borrows information from surrogates along with candidate predictors to impute the unobserved outcomes via a sparse working imputation model with moment conditions to achieve robustness against mis-specification in the imputation model and a one-step bias correction to enable interval estimation for the predicted risk. We demonstrate that the SAS procedure provides valid inference for the predicted risk derived from a high dimensional working model, even when the underlying risk prediction model is dense and the risk model is mis-specified. We present an extensive simulation study to demonstrate the superiority of our SSL approach compared to existing supervised methods. We apply the method to derive genetic risk prediction of type-2 diabetes mellitus using a EHR biobank cohort.