Center for Health AI and Synthesis of Evidence (CHASE)

We have developed of a general framework of distributed learning (also known as collaborative learning) methods with novel inferential procedures that integrate sensitive biomedical data (e.g., EHR data and biobank data) across multiple institutes. We refer to this framework as PDA:Privacy-preserving Distributed Algorithms. Our algorithms are communication-efficient, which only require the collaborating sites to send aggregated information to the coordinating site once.

Xmeta is an open-sourced, well-documented and interactive toolboxfor meta-analysis. There are three main components to this toolbox: An R package called xmeta(), video tutorials and documentation for the package and a web-based analysis platform. Xmeta provides a convenient and straightforward platform for a wide range of users to conduct the meta-analysis.

Jiayi Tong, Jing Huang, Jessica Chubak, Xuan Wang, Jason H Moore, Rebecca A Hubbard, Yong Chen, An augmented estimation procedure for EHR-based association studies accounting for differential misclassification, Journal of the American Medical Informatics Association, Volume 27, Issue 2, February 2020, Pages 244–253, https://doi.org/10.1093/jamia/ocz180

Luo, S., Chen, Y., Su, X., & Chu, H. (2014). mmeta: An R Package for Multivariate Meta-Analysis. Journal of Statistical Software, 56(11), 1–26. https://doi.org/10.18637/jss.v056.i11

This paper describes the core features of the R package mmeta, which implements the exact posterior inference of odds ratio, relative risk, and risk difference given either a single 2 × 2 table or multiple 2 × 2 tables when the risks within the same study are independent or correlated.