Proof of Concept
FoliumNovum has progressed beyond the conceptual stage to an initial software proof of concept. Two complementary prototype approaches have been developed: a Random Forest machine-learning model using structured clinical inputs and a large language model (LLM)-based prototype designed to evaluate clinical presentations relevant to emergency-care decision making.
Structured machine-learning prototype designed to evaluate clinical variables and generate reproducible predictions relevant to emergency-care decision making.
LLM-based prototype designed to interpret clinical presentations and provide contextual decision support complementary to the structured model.
Development & Validation
Development of FoliumNovum will use a de-identified statewide emergency-department dataset spanning approximately eight years of encounters to develop, optimize, and internally validate the models. Independent clinical data will subsequently be used to evaluate generalizability and external validity.
The long-term objective is an interpretable clinical decision-support platform that augments—not replaces—clinical judgment.
Research Status
FoliumNovum is an investigational research platform currently under development. It is not intended for clinical use and has not been cleared or approved by the U.S. Food and Drug Administration.