A principal component analysis approach to correcting the knee flexion axis during gait

Area of Science:

• Biomechanics
• Gait Analysis
• Orthopedic Surgery

Background:

• Accurate knee flexion axis identification is crucial for tibial and femoral derotation osteotomies.
• Marker misplacement during gait analysis frequently leads to errors in knee flexion axis identification.

Purpose of the Study:

• To develop an efficient algorithm for post-hoc correction of knee flexion axis errors.
• To evaluate the efficacy of the developed algorithm against existing methods.

Main Methods:

• Gait data were collected from twelve healthy subjects with both standard and intentionally misplaced lateral knee markers.
• A principal component analysis (PCA)-based algorithm was developed for error correction.
• Efficacy was assessed by quantifying reductions in knee angle errors (crosstalk and rotation offset).

Main Results:

• The PCA-based algorithm significantly reduced crosstalk error (r(2)) (p<0.001).
• It improved the certainty of knee kinematics, with knee rotation converging to 11.9±8.0° external rotation.
• Precision was enhanced, reducing the within-subject standard deviation of hip rotation offset error from 13.5±1.5° to 0.7±0.2° (p<0.001).

Conclusions:

• The PCA-based algorithm effectively corrects knee flexion axis errors caused by marker misplacement.
• This method offers improved precision and certainty in gait analysis for knee kinematics.
• Its performance is comparable to established methods and superior to the null space algorithm.