GCP – Powering open source healthcare research on Google Cloud

Google Cloud is pleased to announce our partnership with Odysseus Data Services to launch ATLAS on GCP. ATLAS is an open source application that is built and maintained by The Observational Health Data Sciences and Informatics (OHDSI) program. This program is a multi-stakeholder, interdisciplinary collaboration to create open-source solutions that bring out the value of observational health data through large-scale analytics. The tool can support everything from data visualization and cohort building to deep learning models. All of these tools are deployed leveraging the Observational Medical Outcomes Partnership (OMOP) common data model.

OHDSI and ATLAS 

ATLAS is an open source application that is built and maintained by The Observational Health Data Sciences and Informatics (OHDSI) program. This program is a multi-stakeholder, interdisciplinary collaboration to create open-source solutions that bring out the value of observational health data through large-scale analytics. The tool can support everything from data visualization and cohort building to deep learning models. All of these tools are deployed leveraging the Observational Medical Outcomes Partnership (OMOP) common data model.  ATLAS v2.7.8 with BigQuery support is available on Google Cloud Marketplace.  

ATLAS is intended to provide a unified interface to standardized patient-level data converted to the OMOP Common Data Model (CDM) and to conduct observational research. It allows researchers to define patient cohorts, select analytical designs, set parameters and execute analytical methods against data.

Benefits of ATLAS on Google Cloud

Odysseus in partnership with Google Cloud provides users with the infrastructure for these open source tools. This means users no longer need to be concerned with the storage of their data, the management of databases or the scalability of their workloads since autoscaling, and managed products take care of this. Using Google Cloud with OHDSI tools allows users to shift their focus from implementations to insights in a cost effective manner. 

Per a recent analysis by Stanford Medicine, ATLAS performs significantly better running on BigQuery than on traditional databases:

“We present the performance of Google BigQuery…we compare Oracle vs BigQuery and show that BigQuery is 10-100x faster…​In nearly all cases (for Postgres), the performance of ATLAS is comparable or faster when using BigQuery.”¹

Summary

ATLAS is free, publicly available, web based, open source software created by the Observational Health Data Sciences and Informatics (OHDSI) community. Odysseus, one of the ATLAS developers and an active OHDSI community participant, collaborated with Google Cloud to enable GCP BigQuery support. This has enabled ATLAS to be deployed on GCP platforms in multiple healthcare organizations. These academic and healthcare institutions are now active participants in the scientific medical research conducted by the OHDSI community.

With these tools, we can improve our ability to share health insights and learnings across disciplines, departments, sectors and geographies, to improve outcomes. 

Google Cloud’s Commitment to Healthcare Data and Research

Google Cloud has long supported the healthcare research community from health researchers  to biomedical scientists to policy makers.  Google Cloud has enabled open source tools, built unique product offerings, and created strong partnerships with leading public and private sector organizations to advance healthcare research.  Researchers, enterprises, and government entities choose Google Cloud to advance their research goals due to our strong security and reliability, our data analytics, and our cutting edge ML/AI capabilities.

In addition to our new partnership with OHDSI, Google Cloud provides the basis for other open source data platforms such as Terra, developed by Verily and the Broad Institute of MIT and Harvard, and we have longstanding partnerships with research focused organizations such as Tempus, Schrodinger, and Clear Labs.  In addition, we have worked closely with the National Institutes of Health, and have enabled Accenture’s INTIENT platform on Google Cloud.

^{1. Stanford Medicine – Datta, Somalee, et al. Mar 17 2020. https://arxiv.org/abs/2003.10534}