Pharmasight
Sign inSee a worked packBook the bench
Design & Stress-Test · Gretel

generated 16M+ synthetic patient records for labor-and-delivery data to train ML forecasting models while preserving privacy

Unnamed major healthcare institution (2,000+ care sites)Design & Stress-TestGretel

The customer is anonymised in Gretel’s own case study — not withheld by us.

Gretel is a synthetic-data generation platform (acquired by NVIDIA) used to create privacy-preserving stand-ins for sensitive genomic and patient data.

This sits in the Design & Stress-Test stage of the AI use-case lifecycle: Named examples from the tools pharma teams use to scope, prototype and benchmark an AI use case before committing to build it — data-catalog, sandbox and evaluation platforms.

Source

Published by Gretel on its own site — read the original case study on info.gretel.ai ↗

More like this

Related use cases

  1. 01used Gretel's synthetic-data platform to create safe synthetic genomic datasets, enabling genomics research and safe/private data sharing between researchers, healthcare providers and industryIlluminaGretelDesign & Stress-Test
  2. 02turned synthetic data into a strategic-growth capability (finance; non-pharma) — referenced on Gretel's finance solutions page, dedicated case-study URL not directly resolved in this pass.Global payments platform (unnamed)GretelDesign & Stress-Test
  3. 03synthetic time-series data for cross-institution use (finance; non-pharma) — referenced on Gretel's finance solutions page, dedicated case-study URL not directly resolved in this pass.Unnamed global financial institution(s)GretelDesign & Stress-Test
  4. 04pharmaceutical company using Agent Bricks for clinical and pharmacovigilance workflowsNovo NordiskDatabricksDesign & Stress-Test
  5. 05runs its platform on Databricks/AWS to accelerate drug discovery; the Regeneron Genetics Center built one of the largest genetics databases pairing sequenced exomes with EHRs for 400,000+ peopleRegeneronDatabricksDesign & Stress-Test
  6. 06built a knowledge graph and recommendation engine on Databricks to help scientists mine millions of data points across thousands of sources and generate novel drug-target hypotheses; separately used Agent Bricks to parse 400,000+ clinical trial documents into structured data in under 60 minutes with no codeAstraZenecaDatabricksDesign & Stress-Test