What is Flywheel?
Maybe you’ve heard that 80% of a data scientist’s time is spent finding, curating and organizing data. We’re changing that. Flywheel is a medical imaging AI development platform that automates time-consuming processes so scientists and researchers can focus on making discoveries.
What Can Flywheel Do?
Data Exchange
Curated Datasets
Accelerate the development of models with access to analysis-ready data from public and private sources.
Publish Your Datasets
Share data to Flywheel Exchange for others to preview & license for their projects.
Access Data from Gradient Health
License real-world, de-identified patient data to access representative datasets & test your algorithms to develop trustworthy AI.
Data Discovery
Create Cohorts With Data From Your PACS
Filter imaging in your PACS, search clinical findings from radiology reports, and extract imaging cohorts for curation & analysis.
Create Cohorts with Nuance
Access Nuance mPower to harness large language models & identify actionable study datasets to develop AI-supported insights.
Data Curation
Curation Pipelines
Automate quality and pre-processing workflows to efficiently and consistently create analysis-ready datasets.
Reader Studies
Orchestrate blinded reader studies with task assignment workflows, searchable annotations, and guided data collection.
Model Development
Develop with Jupyter Notebooks
Test and verify algorithms in an integrated environment with data from your Flywheel project & organize model results.
Develop on Azure ML
Develop algorithms at scale with curated data. Publish models to collaborators in Flywheel to track quality and compute consumption and test across projects.
Develop with MONAI
Work with your preferred open source algorithms and test them against your data to uncover bias and improve model performance.