Researchers have developed the Nimblemind Multi-Agent System (nMAS), a novel pipeline designed to automate the complex process of feature engineering from electronic health records (EHRs) for heart failure research. This system addresses a significant bottleneck in clinical research, where feature engineering can consume up to 45% of a data scientist's time. nMAS successfully generated and verified a substantial number of structured and aggregated features, which, when added to existing models, significantly improved the accuracy of phenotyping for both heart failure with reduced ejection fraction (HFrEF) and heart failure with preserved ejection fraction (HFpEF). The system's ability to provide evidence traceability and methodological soundness was also highlighted by an independent LLM assessment. AI
IMPACT Automates a critical bottleneck in clinical research, potentially accelerating AI-driven discoveries in cardiovascular disease.
RANK_REASON Academic paper detailing a new system for feature engineering in EHRs. [lever_c_demoted from research: ic=1 ai=1.0]
- electronic health records
- heart failure
- Heart failure with preserved ejection fraction
- heart failure with reduced ejection fraction
- New Mexico Academy of Science
- Nimblemind Multi-Agent System
- U.S.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
