Privacy-preserving data-sharing middleware for federated
This technology allows multiple organizations to collaborate and share data insights without ever exposing their raw, sensitive data to each other. Each participant first compresses their data into a stripped-down 'distilled' form that has had identifying or sensitive features mathematically removed. A central system then figures out which participants have data most relevant to a given task — say, a fraud detection model or a medical prediction — and shares only the relevant distilled snippets with the requesting party. The result is that an organization can improve its AI model or analytics service by learning from others' data without anyone handing over private records.
What you could build
A privacy-preserving data marketplace middleware platform for federated industries — such as healthcare networks, financial services consortia, or smart-city operators — where members want shared AI model improvements without surrendering patient records, transaction logs, or sensor data. Buyers would be enterprise data platform vendors and regulated-industry data consortia.
Who in Virginia should care
Northern Virginia's dense concentration of federal contractors, health systems, and data center operators would find this relevant for privacy-compliant data collaboration across agency or enterprise boundaries.
Readiness: Concept
Concept — described but not yet demonstrated. Lab validated — supported by experimental results in the patent. Prototype likely — the text describes a built, working embodiment.
Readiness is inferred from the patent text, not from a lab visit.
The record
- Inventors
- Ismini Lourentzou, Ran Jin
- Filed
- Patent pending — filed September 25, 2024
- Status
- Application
- Publication number
- US20250217506A1
Ready to talk?
Virginia Tech Intellectual Properties handles licensing for this technology.
Prosim summaries are generated from public patent text and are not legal advice.