Federal Project Grant Award Summary

The Regents of the University of Minnesota received a $600,000 project grant from the National Science Foundation's Division of Information and Intelligent Systems (CISE Program, CFDA 47.070) awarded February 1, 2026, with a completion date of January 31, 2029. The award supports the development of a federated learning framework for artificial intelligence-driven diagnosis of rare genetic disorders, specifically ciliopathies and related rare diseases. The research deliverables include a privacy-preserving machine learning architecture that enables collaborative model training across multiple healthcare institutions without requiring direct exchange of patient-level data, addressing critical barriers to rare disease diagnosis posed by geographically dispersed patient populations, limited clinical expertise, and strict privacy regulations.

The technical products developed under this award encompass deep clinical phenotyping capabilities through advanced natural language processing and large language models to extract and normalize observational data from electronic health records in multiple languages, integration of biomedical ontologies and rare disease knowledge bases, and patient similarity modeling systems to support diagnostic decision-making across diverse healthcare environments. The research aims to reduce diagnostic delays and improve clinical accuracy for rare genetic disorders by enabling large-scale collaborative learning from distributed clinical data while maintaining local data security and regulatory compliance across participating institutions.