Predicting physiological hip joint loads with inverse bone remodeling using clinically available QCT images.

Area of Science:

• Biomechanics
• Computational modeling
• Medical imaging

Background:

• Assessing in vivo joint loading is difficult due to invasive methods or complex computations.
• Inverse bone remodeling (IBR) recovers loading conditions from bone microstructure images.
• Homogenized finite element (hFE) models offer an efficient IBR approach.

Purpose of the Study:

• Compare hip joint load predictions using hFE-based IBR with clinically feasible CT scans versus micro-FE-based IBR (gold standard).
• Validate a new, efficient IBR workflow for predicting physiological hip joint loads.

Main Methods:

• Proximal femora (n=20) scanned using clinical QCT (0.3 mm) and high-resolution Xtreme CT II (0.03 mm).
• Finite element (FE) models of decreasing complexity were generated.
• hFE models from QCT data and micro-FE (µFE) models from high-resolution data were used for IBR.
• IBR predicted optimal scaling factors for twelve unit load cases.

Main Results:

• hFE-based IBR with QCT data showed trends consistent with high-resolution models.
• Peak load magnitudes between µFE and hFE-based IBR were well correlated (R²=76.8%).
• Quantitative agreement (CCC=82.8%) required additional peak load calibration.

Conclusions:

• First direct comparison of µFE-based IBR and hFE-based IBR using QCT data.
• A clinically feasible workflow, including peak calibration, enables rapid prediction of physiological peak hip joint loads.
• This method simplifies the assessment of joint loading conditions.