AI RESEARCH
DeepEN: A Deep Reinforcement Learning Framework for Personalized Enteral Nutrition in Critical Care
arXiv CS.AI
•
ArXi:2510.08350v3 Announce Type: replace-cross Objective: Enteral nutrition (EN) delivery in the ICU remains suboptimal due to limited personalization and uncertainty regarding appropriate calorie, protein, and fluid targets under dynamic metabolic demands. We Methods: DeepEN was trained on over 11,000 ICU patients from MIMIC-IV to generate 4-hourly, patient-specific caloric, protein, and fluid targets. The state representation incorporated graphics, comorbidities, vital signs, laboratory values, and recent interventions.