Recent workshops
February 21, 2025
NeuroFLAME Virtual Demo
Get a sneak peak at the newest version of COINSTAC, now called NeuroFLAME. With NVIDIA FLARE's federated learning framework, we've re-engineered our platform from the ground-up, taking all our years of experience and learning and creating a new tool that we think will open new doors and revolutionize global collaborative research, especially the field of Neuroscience.
Why NeuroFLAME?
Neuroimaging Federated Learning Analysis for Multi-Site Environments
The Problem
You want to do research, and you want to include data from around the world. Unfortunately, orchestrating such an event is anything but trivial.
Coordinating data-driven research can be difficult. Who's going to collect all of the files? Who is going to actually "run" group analysis on all of the data?
Ensuring privacy can be difficult. Can I trust other people or institutions with my research participants' data? Am I even allowed to share it?
Valuable research data may often not be shared due to privacy or IRB constraints.
Large datasets can be expensive to transfer. When file sets are in the GB and TB range, network transfers are not immediately trivial or even practical.
"Smart bullies" have demonstrated ability to extract personal information from various aggregated, anonymized datasets. How can we share data without revealing confidential information?
Bottom line: collaborative group research requires a great deal of coordination. Human and business factors can hamper research from happening at a pace that we are able to handle! Constraints may even forbid group research to occur at all.
The Solution
NeuroFLAME removes the barriers to collaborative analysis by:
Decentralizing analyses and computation
Each user performs analyses via vetted decentralized pipelines on their own computers. Derivatives of the analysis from each users' output is sent to a central compute node.
A central compute node performs a complimentary component of the group analysis, generally a federated average etc. This node may trigger further computations on users' machines, generally in effort to improve a model, which the research is trying to predict!
Not synchronizing full datasets.
Instead, synchronizing only resultant analysis metrics, central compute nodes aggregates these metrics, and attempts to draw conclusions from the contributor swarm.
Because machine learning algorithms can be designed to model outcomes via artifacts of your analysis pipelines, we keep your data safely and conveniently on your own machine, untouched.
Applying differential privacy strategies to truly anonymize private data, whilst still permitting collaboration.
Partners
NeuroFLAME has been made possible through
the past and present efforts of these institutions
The following research groups
are currently using NeuroFLAME in their research
Of course, none of this would be possible
without grants from the
National Institutes of Health
Places Using NeuroFLAME
Get Started
To download and run NeuroFLAME on your local machine click the button below
To create and run a local development build and contribute to the project start here:
Contribute by creating computations to with our handy documentation guide.