Software for Distributed Computation on Medical Databases: A Demonstration Project

Abstract

Bringing together the information latent in distributed medical databases promises to personalize medical care by enabling reliable, stable modeling of outcomes with rich feature sets (including patient characteristics and treatments received). However, there are barriers to aggregation of medical data, due to lack of standardization of ontologies, privacy concerns, proprietary attitudes toward data, and a reluctance to give up control over end use. Aggregation of data is not always necessary for model fitting. In models based on maximizing a likelihood, the computations can be distributed, with aggregation limited to the intermediate results of calculations on local data, rather than raw data. Distributed fitting is also possible for singular value decomposition. There has been work on the technical aspects of shared computation for particular applications, but little has been published on the software needed to support the "social networking" aspect of shared computing, to reduce the barriers to collaboration. We describe a set of software tools that allow the rapid assembly of a collaborative computational project, based on the flexible and extensible R statistical software and other open source packages, that can work across a heterogeneous collection of database environments, with full transparency to allow local officials concerned with privacy protections to validate the safety of the method. We describe the principles, architecture, and successful test results for the site-stratified Cox model and rank-k singular value decomposition.

Caring without sharing: Meta-analysis 2.0 for massive genome-wide association studies

“…For GWAS, our pipeline leverages an algorithm for performing PCA by communicating the LD-matrices (Algorithm 1), and subsequently a method for performing generalized regression (most commonly, linear and logistic regression) by iteratively solving a regularized regression at each silo. This general, iterative approach, known as Alternating Directions Method of Multipliers (11; 12), is guaranteed to converge (13), has suc-2 cessfully been applied to many problems (14; 15; 16), and for GWAS we show that moderately high accuracies can be achieved with a few iterations (also see (14)). We show that our pipeline is accurate, scalable, practical, and a significant improvement over the meta 1.0 approach.…”

Imputation of Mixed Data With Multilevel Singular Value Decomposition

“…Both obstacles can be overcome by turning to distributed computations, which consists in leaving the data on sites and distributing the calculations, so that hospitals only share some intermediate results instead of the raw data (Narasimhan et al, 2017). Among other methods, SVD, which only involves inner products and sums, can be very straightforwardly implemented in a distributed manner.…”

“…Indeed, all the computations in Algorithm 4 can be done in parallel with a master-slave architecture (Narasimhan et al, 2017), where a central server collects summary statistics computed locally on sites, as illustrated Figure 5. Here, the local right singular vectors v j , j ∈ {1, .…”

DataSHIELD – New Directions and Dimensions

“…Brown et al 2010a;Brown et al 2010b) and open source software (e.g. Carter et al 2016;Narasimhan et al 2017). The Canadian Network for Observational Drug Effect Studies (CNODES, Suissa et al 2012) and Mini-Sentinel (a safety surveillance system developed by the U.S. Food and Drugs Administration, Platt and Carnahan, 2012) are both platforms to facilitate the running of analysis requests from approved users locally, along with disclosure checks, prior to securely combining the results centrally as a meta-analysis.…”