745.4 MB of data from consumer-grade wearable sensors aims to shortcut expensive lab-based musculoskeletal modeling. Gradient boosting models predict internal forces in the lower body, a major driver of overuse injuries in runners. The dataset was created by John Davis and last updated on May 9, 2026.
Use Cases
- Predicting lower-body internal forces based on wearable sensor data
- Developing injury risk models for runners based on movement data
- Training gradient boosting models to shortcut traditional biomechanical modeling
- Comparing predictions from wearable sensors against lab-based force estimates
Strengths
- 745.4 MB of data provides a substantial volume for model training
- The MIT license allows for broad use and modification
- Focus on consumer-grade sensors suggests potential for real-world, out-of-lab application
Limitations
- Column-level documentation is absent; field semantics must be inferred after download
- Row count is unknown, which may limit suitability assessment
- Data may reflect temporal or source bias inherent to figshare
Provenance
- Source
- John Davis
- Collection Method
- Likely collected from consumer-grade wearable sensors worn by runners.
- Freshness
- Last updated 2026-05-09 07:38:45