About the Creator
A privacy-preserving framework for simulating public health interventions and optimizing resource allocation in real-time.
See It In Action
Public health resources are frequently allocated based on lagging, aggregate data. This "top-down" approach fails to capture hyper-local realities, creating neighborhoods that lack access to care. Strict privacy laws (HIPAA) often block access to the detailed data needed to solve this.
SynthEquity bridges this gap using Generative AI. We generate a Privacy-Preserving Synthetic Population that mirrors a real city but contains no real individuals. This allows policymakers to model interventions freely using our real-time dashboard. See a live demo here.
SynthEquity utilizes a Generative AI model (CTGAN) to produce a privacy-preserving synthetic population from public data. A multivariate regression model is then trained on this synthetic population to power real-time inference for the browser-based dashboard, calculating health disparity scores to identify medical deserts and simulate intervention impacts.
Uses Conditional Tabular GANs to ingest Census data and output a synthetic cohort that retains statistical fidelity.
A multivariate regression engine that quantifies health vulnerability based on transit, income, and proximity.
Client-side React/TypeScript engine that recalculates community risk scores in under 100ms.