Related Experiment Video
Updated: May 29, 2026
06:36
Lower Limb Biomechanical Analysis of Healthy Participants
Published on: April 15, 2020
Functional knee axis based on isokinetic dynamometry data: Comparison of two methods, MRI validation, and effect on
A Van Campen1, F De Groote, L Bosmans
1Department of Mechanical Engineering, Katholieke Universiteit Leuven, Division PMA, Celestijnenlaan 300B, B-3001 Heverlee, Belgium. anke.vancampen@mech.kuleuven.be
Journal of Biomechanics
|September 20, 2011
Summary
Functional knee axes of rotation (fAoR) better describe knee motion than geometry-based axes. The Gamage and Lasenby (GL) algorithm is more accurate than the SARA algorithm for calculating fAoRs from dynamometry data.
Area of Science:
- Biomechanics
- Orthopedics
- Medical Imaging
Background:
- Accurate determination of knee axes of rotation (AoR) is crucial for understanding knee joint mechanics and designing prosthetics.
- Existing methods include geometry-based axes (transepicondylar axis, geometric center axis) and motion-based functional axes (fAoR).
- The precision of fAoR calculation depends on the chosen algorithm and the quality of motion data.
Purpose of the Study:
- To compare the accuracy of geometry-based knee axes versus motion-based fAoRs.
- To evaluate two algorithms (Gamage and Lasenby's sphere fitting - GL, and Ehrig et al.'s axis transformation - SARA) for calculating fAoRs.
- To assess the performance of different knee axes in describing knee joint kinematics and dynamics.
Main Methods:
- Calculated fAoRs using GL and SARA algorithms based on 3D motion data from isokinetic dynamometry.
- Validated calculated AoRs against an equivalent axis derived from static MRI images.
- Quantified differences in orientation and location between various knee axes.
- Estimated knee joint angle and acceleration trajectories using Kalman smoothing.
Main Results:
- Functional knee axes of rotation (fAoR) demonstrated smaller differences compared to the MRI-based equivalent axis than geometry-based axes.
- Maximum differences in orientation and location were 5.4°/11.5mm (GL) and 8.6°/12.8mm (SARA) for fAoR vs. MRI.
- Geometry-based axes showed larger maximum differences: 10.6° in orientation and 20.8mm in location.
- The GL algorithm performed better than SARA in calculating fAoRs from experimental dynamometry data, with maximum differences of 5.7° and 15.4mm between the two algorithms.
- RMS differences for joint angles and accelerations were within acceptable limits (3° and 20°/s², respectively) when using the MRI-based axis as reference.
Conclusions:
- Functional knee axes of rotation provide a more accurate representation of knee joint motion compared to geometry-based axes.
- The Gamage and Lasenby (GL) algorithm is superior to the SARA algorithm for calculating fAoRs from isokinetic dynamometry data.
- These findings have implications for improving the biomechanical analysis of knee function and the design of orthopaedic interventions.
