Researchers have developed a system called ICD-Deepresearch that can predict which medical diagnosis codes will be documented in a patient's future visit based on their longitudinal record. The system combines predictive foundation models with medical search and ICD dictionaries to generate flexible diagnostic hypotheses. It achieves higher precision and recall compared to existing systems, and physicians rate its retrieved documents as more useful. The system uses a combination of machine learning and natural language processing techniques, including GPT-5, to make predictions. --- Why it matters: This matters because it could lead to better patient care by providing healthcare professionals with accurate and relevant information for diagnosis and treatment planning. It also demonstrates the potential of combining different AI models and techniques to improve clinical decision-making. Source: https://arxiv.org/abs/2608.17075