BiomechAgent: AI-Assisted Biomechanical Analysis Through Code-Generating Agents
1 st R. James Cotton, 2 nd Thomas Leonard · 2026
Read BiomechAgent: AI-Assisted Biomechanical Analysis Through Code-Generating Agents by 1 st R. James Cotton, 2 nd Thomas Leonard (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-06975.txt.
Description
R. James Cotton and Thomas Leonard present BiomechAgent, an AI agent that lets clinicians analyze movement data without writing code. Markerless motion capture has made measured movement analysis widely available, but turning the resulting data into answers still needs programming skill. BiomechAgent takes plain-language requests to query data stores, make charts, and read results.
The authors built a benchmark spanning data retrieval, chart making, activity sorting, time splitting, and clinical thinking. BiomechAgent did well on retrieval and charts and showed growing clinical thinking skill. Domain-specific guidance beat generic prompts, and tested tools for gait event detection sharply improved hard space-time analysis. A local open-weight model did much worse than a frontier cloud model except on retrieval, a trade-off the paper documents for setups that cannot use cloud models.
Catalog record
| CREATOR | 1 st R. James Cotton, 2 nd Thomas Leonard |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-06975