Article type
Abstract
Objectives
This session aims to illuminate the transformative potential of artificial intelligence (AI), particularly large language models (LLMs), in enhancing guideline development and use. Attendees will explore AI's role in evidence synthesis and guideline development and implementation, including its application in the assessment of GRADE domains and completion of the Evidence-to-Decision (EtD) framework.
Description and interaction plans
As AI continues to reshape health care practice and decision-making, this session delves into its pivotal role in guideline development and implementation through the following segments:
1. AI extension for GIN-McMaster Checklist: introducing a new extension that ensures that the integration of AI tools and methodologies into the guideline development process is evidence-based and ethically sound and improves the quality and efficiency of guidelines.
2. Question generation with LLM: demonstrating LLMs’ capabilities in understanding queries posed by internet users and identifying gaps in existing questions to support the process of generating guideline questions.
3. GRADE GPT: showcasing how GPT supports the transparent and consistent application of GRADE criteria in the systematic evaluation of the certainty of evidence and strength of recommendation in guideline development.
4. AI-integrated EtD framework: introducing an LLM-enhanced approach to transition from guideline recommendations to implementation, integrating a “virtual expert panel” that synthesizes information into comprehensive, context-specific EtD frameworks.
5. RecChat: evidence-based AI assistant as a novel channel for guideline dissemination and implementation.
To engage attendees, each segment will include Q&A and/or an interactive live demonstration of AI products, concepts, and prototypes. We will further facilitate a forward-thinking dialogue to encourage the exchange of ideas and experiences among participants.
Target audience
Guideline developers, methodologists, and health care professionals interested in leveraging AI to enhance the quality and efficiency of clinical practice guidelines. Participants will gain a deeper understanding of AI's potential and insights into the ethical, practical, and methodological challenges of integrating AI into guideline development.
Level of knowledge
All levels.
This session aims to illuminate the transformative potential of artificial intelligence (AI), particularly large language models (LLMs), in enhancing guideline development and use. Attendees will explore AI's role in evidence synthesis and guideline development and implementation, including its application in the assessment of GRADE domains and completion of the Evidence-to-Decision (EtD) framework.
Description and interaction plans
As AI continues to reshape health care practice and decision-making, this session delves into its pivotal role in guideline development and implementation through the following segments:
1. AI extension for GIN-McMaster Checklist: introducing a new extension that ensures that the integration of AI tools and methodologies into the guideline development process is evidence-based and ethically sound and improves the quality and efficiency of guidelines.
2. Question generation with LLM: demonstrating LLMs’ capabilities in understanding queries posed by internet users and identifying gaps in existing questions to support the process of generating guideline questions.
3. GRADE GPT: showcasing how GPT supports the transparent and consistent application of GRADE criteria in the systematic evaluation of the certainty of evidence and strength of recommendation in guideline development.
4. AI-integrated EtD framework: introducing an LLM-enhanced approach to transition from guideline recommendations to implementation, integrating a “virtual expert panel” that synthesizes information into comprehensive, context-specific EtD frameworks.
5. RecChat: evidence-based AI assistant as a novel channel for guideline dissemination and implementation.
To engage attendees, each segment will include Q&A and/or an interactive live demonstration of AI products, concepts, and prototypes. We will further facilitate a forward-thinking dialogue to encourage the exchange of ideas and experiences among participants.
Target audience
Guideline developers, methodologists, and health care professionals interested in leveraging AI to enhance the quality and efficiency of clinical practice guidelines. Participants will gain a deeper understanding of AI's potential and insights into the ethical, practical, and methodological challenges of integrating AI into guideline development.
Level of knowledge
All levels.