Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the development of privacy metrics for genomic data analysis in this 14-minute conference talk from PEPR '25. Learn about collaborative research between the National Institute of Standards and Technology (NIST), the US Census Bureau, and other organizations addressing the critical challenge of sharing genomic datasets across institutions while maintaining patient privacy. Discover how privacy-preserving federated learning techniques can enable cross-institutional genomic research for understanding and treating diseases like rare cancers, while navigating complex legal and ethical concerns. Examine the project's technical architecture, goals, and initial performance results obtained using plant genomic data as a proxy for human genomic datasets. Gain insights into the metrics and use cases being developed for privacy-preserving machine learning applications in genomic research, presented by Curtis Mitchell from the US Census Bureau's xD division alongside collaborators from NIST, Knexus, and MITRE.
