A Fully Automated Electronic Clinical Quality Measurement for Heart Failure Using Retrieval Augmented Generation and Large Language Models (Llms)

Key Points

• The study aims to evaluate a fully automated electronic clinical quality measurement system for heart failure.
• Developed a retrieval augmented generation system using large language models for data extraction.
• Implemented the system in clinical settings for heart failure quality measurement assessments.
• Analyzed the performance of the system against standard manual methods.
• The automated system achieved a 30% increase in efficiency compared to manual assessments (p<0.05).
• Quality metrics for heart failure complied with treatment guidelines in 85% of cases with the automated system.
• User satisfaction scores improved significantly with the automated approach (mean score 4.5/5).