What are LLMs for injury surveillance?
A preprint study analyzes e-bike incidents using two data sources. The study develops an LLM-assisted framework.

Illustration: generated for Physis Systems
Short answer: The study analyzes e-bike-related injury incidents using two data sources. The data sources are CPSRMS and NEISS. The study develops an LLM-assisted framework.
The primary source for this article is arXiv cs.AI's Large Language Modelsfor Injury Surveillance: Evidence from E-Bike Incidents. It is a preprint, which means it has not yet been peer reviewed.
What is the LLM-assisted framework?
The LLM-assisted framework converts unstructured incident narratives into structured variables.
How is injury severity analyzed?
Injury severity is harmonized into three ordered categories. Separate ordered logit models are estimated for CPSRMS and NEISS.
What are the key findings?
Human-related factors are associated with higher injury severity in CPSRMS. Equipment- and environment-related factors are associated with lower severity categories in CPSRMS. Riders under age 18 are associated with lower treatment severity in NEISS.
What do we know, and what is still open?
What is confirmed: The study was submitted on June 5, 2025, and last revised on November 13, 2026.
What is still open: The preprint study has not been peer reviewed.
Why does Physis Systems cover this?
Physis Systems follows research at the frontier, from space to fundamental science. For more stories like this one, visit the Physis Systems blog.
About Physis Systems
Physis Systems (Physis Defense Systems, Inc.) builds Inventi Matrix, an invention engineering platform available in the United States and Canada at inventimatrix.com, and Lumina AI Companion, a personal AI companion you can meet on the Lumina page.
Frequently asked questions
What is the purpose of the study?
The study analyzes e-bike-related injury incidents using two data sources.
What are the data sources used?
The data sources are CPSRMS and NEISS.
What is the LLM-assisted framework?
The LLM-assisted framework converts unstructured incident narratives into structured variables.


