OSR 2023 | Emergent Session: Enabling federated analysis on large datasets with COINSTAC Vaults

Cover Photo

Jul

25

12:00pm

OSR 2023 | Emergent Session: Enabling federated analysis on large datasets with COINSTAC Vaults

By ossig2024

COINSTAC promotes collaborative research by removing large barriers to traditional data-centric approaches. It allows groups of users to run common analyses on their own machines over their own datasets with ease. The results of these analyses are synchronized to the cloud and undergo aggregate analysis processes using all contributor data. Federated (decentralized) pipelines enable distributed, iterative, and feature-rich analyses, opening up new possibilities for collaborative computation. It also offers data anonymity through differentially private algorithms, so members do not need to fear protected health information (PHI) traceback.
The goal of this discussion is to introduce COINSTAC and COINSTAC Vaults (CVs). COINSTAC's federated analysis capabilities integrate seamlessly with CVs to reduce barriers by hosting standardized, persistent, and highly available datasets. A CV streamlines collaboration by providing a user interface for self-service analysis and collaboration that eliminates manual coordination with data owners. Importantly, CVs can also be used in conjunction with open data by simply creating a CV that hosts the available data. This can then be a part of future analyses, thus filling an essential gap in the data-sharing ecosystem. We illustrate the impact of CVs through several functional and structural neuroimaging studies utilizing federated analysis. These analyses showcase their potential to improve the reproducibility of research and increase sample sizes in neuroimaging studies.
In this session, we will also demonstrate commonly used data analysis pipelines in COINSTAC framework. Other features will be showcased including Singularity container platform support. We would like to hear feedback about our software such as how to improve the experience for researchers. We welcome anyone who wants to contribute to this open source and open data project with their datasets, algorithms, and code. We would also like to work with other organizations to pursue grants together, including small business grants. Collaborating with other organizations is the best way for us to answer interesting neuroscience-related questions that would not have been possible without COINSTAC and COINSTAC Vaults.

hosted by

ossig2024

share

Open in Android app

for a better experience