From bytes to bites: application of large language models to enhance nutritional recommendations

Area of Science:

• Healthcare technology
• Artificial intelligence in nutrition
• Chronic kidney disease management

Background:

• Large language models (LLMs) are being integrated into daily life, with potential applications in healthcare.
• Personalized nutritional guidance is crucial for patients with chronic kidney disease (CKD).
• Collaboration among healthcare professionals, patients, caregivers, and LLM experts is vital for effective LLM integration.

Purpose of the Study:

• To review the current literature on using LLMs for nutrition recommendations in CKD.
• To identify limitations and potential improvements for LLM-based nutrition support in CKD.
• To explore the future role of generative AI in CKD nutrition.

Main Methods:

• Comprehensive literature review focusing on LLMs for CKD nutrition.
• Inclusion of findings from adjacent fields like diabetes and obesity management.
• Expert curation of reviewed literature.

Main Results:

• Current LLM applications for CKD nutrition support are limited and require enhancement.
• LLMs can generate recipes but often underestimate critical nutritional components like electrolytes and calories.
• Advancements in AI are expected to improve nutritional analysis and food identification for kidney health.

Conclusions:

• LLM-based nutritional support for CKD patients is in its early stages but evolving rapidly.
• Future AI advancements may enable accurate nutritional analysis, visual cooking aids, and restaurant menu recommendations.
• Community engagement is essential for a balanced and optimistic approach to AI in CKD nutritional care.