Researchers have developed a novel four-agent framework called MARCA (Multi-Agent Retrieval-Augmented Causality Assessment) to assist in the causality assessment of individual case safety reports in pharmacovigilance. This system utilizes large language models like Gemini 2.5 Pro and DeepSeek-V3.2, combined with retrieval-augmented generation, to analyze scientific literature and drug labels. MARCA demonstrated high agreement with human experts on various assessment questions, with a final causality classification agreement of 79.4%. Exploratory results also suggested improved accuracy when using retrieval-augmented scores. AI
IMPACT This framework could enhance the efficiency and accuracy of drug safety monitoring by automating aspects of causality assessment.
RANK_REASON The cluster describes a research paper published on arXiv detailing a new AI framework for a specific domain. [lever_c_demoted from research: ic=1 ai=1.0]
- arXiv
- DeepSeek V3.2
- FDA Adverse Event Reporting System
- Gemini 2.5 Pro
- MARCA
- Maurizio Sessa
- pharmacovigilance
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
