Back

Calibration of instrumented treadmills using an instrumented pole; a modified version to use relatively smaller forces

Hosseini-Yazdi, S.-S.

2024-07-11 bioengineering
10.1101/2024.07.08.602562 bioRxiv
Show abstract

The instrumented treadmills quality of the generated Ground Reaction Forces (GRF) may degrade over time, as the original calibration matrix may not accurately represent the exerted forces. A cost-effective alternative to manufacturer recalibration is to use an instrumented pole for calibration. Collins et al. presented a simple method to collect multiple data points by exerting forces in various directions. The sensor on the instrumented pole provides instantaneous force magnitudes, while motion capture records the poles instantaneous directions. They recommended a relatively large force magnitude (1000N), requiring at least two individuals. Using an optimization method, the new calibration may be estimate by relating the exerted forces (pole) to the treadmill signals. Here, we attempted to simplify the process further, allowing a single individual to perform force exertion with additional force exertion direction. Thus, the calibrating forces were reduced to one-third of the prior recommendation. This maintained the structural integrity of the pole and helped avoid inducing bending moments that could affect calibration results. The cross-validation score for test data (prediction score) was at least 0.92. Additionally, comparing GRFs in posterior/anterior and vertical directions with a benchmark treadmill for even walking revealed an average cross-correlation of 0.97.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.