Reproduce “Learning a patient world model for health system simulation in intensive care” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yingtai Li, Ruoyan Wang, Yu Li, Shaohua Kevin Zhou
The authors build a health-system simulation that couples longitudinal patient histories to shared ICU beds, staff, and queues. A fixed neural model supplies predictions, while external mappings and operating rules configure scenarios. In 404 held-out patient pairs, it predicted a 24-hour recovery contrast of −6.40 percentage points, near −6.68 observed; compact models predicted zero. Simulations are not validated operational forecasts.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.