Large language models: a primer and gastroenterology applications.

Area of Science:

• Artificial Intelligence in Medicine
• Clinical Informatics
• Gastroenterology Innovation

Background:

• Large Language Models (LLMs) like ChatGPT represent a significant advancement in artificial intelligence (AI).
• These models generate high-quality text without extensive task-specific training or complex engineering.
• LLMs show potential to revolutionize clinical workflows, patient outcomes, medical education, and research.

Purpose of the Study:

• To provide gastroenterologists with a practical understanding of LLMs.
• To highlight LLM technical foundations, strengths, limitations, and safe usage.
• To explore potential applications and implementation barriers of LLMs in gastroenterology.

Main Methods:

• Review of current state-of-the-art LLMs and their capabilities.
• Discussion of LLM technical underpinnings, focusing on practical aspects for clinicians.
• Exploration of use cases in clinical gastroenterology, education, and research.
• Analysis of implementation challenges and ongoing solutions.

Main Results:

• LLMs offer powerful text-generation capabilities applicable to various medical tasks.
• Effective and safe interaction with LLMs requires understanding their strengths and limitations.
• Potential applications span clinical decision support, medical documentation, patient education, and research acceleration.

Conclusions:

• LLMs present transformative potential for gastroenterology practice, education, and research.
• Understanding LLM technology is crucial for clinicians to actively participate in its development and implementation.
• Addressing implementation barriers is key to realizing the full benefits of LLMs in healthcare.