A new arXiv preprint introduces TRACE, a model for inpatient risk prediction that accounts for the messy, real-world way clinical data arrives. Unlike methods that assume regularly spaced measurements, TRACE is built for heterogeneous evidence that accumulates at uneven intervals during a hospital stay.

The abstract lists three core prediction targets: in-hospital mortality, prolonged length of stay, and readmission. Because the abstract is truncated, details on the model's architecture and evaluation are not available in the source, but the stated focus is on handling multimodal clinical trajectories.

The significance is practical: hospital data is rarely tidy, and risk tools that ignore timing or data type may miss important patterns. TRACE appears to address that gap directly, though full validation details await the complete paper.