An ensemble meta-prediction framework to integrate multiple external models into a current study
share
Disease risk prediction models are used throughout clinical biomedicine. With the discovery of new biomarkers these models could be improved and provide better predictions. However, the data that includes the new biomarkers will typically have a limited sample size. We aim to build improved prediction models based on individual-level data from an "internal" study while incorporating summary-level information from "external" models. We propose a meta-analysis framework along with two weighted estimators as the ensemble of empirical Bayes estimators, which combines the estimates from the different external models. The proposed framework is flexible and robust in the ways that (i) it is capable of incorporating external models that use a slightly different set of covariates; (ii) it is able to identify the most relevant external information and diminish the influence of information that is less compatible with the internal data; and (iii) it nicely balances the bias-variance trade-off while preserving the most efficiency gain. The proposed estimators are more efficient than the naive analysis of the internal data and other naive combinations of external estimators.
READ FULL TEXT