Measured and predicted mechanical internal work in human locomotion

Area of Science:

• Biomechanics
• Human Locomotion Analysis
• Exercise Physiology

Background:

• Estimating mechanical internal work (W(int)) is crucial for understanding locomotion biomechanics, especially when experimental data are unavailable.
• A prior model predicted W(int) using velocity, stride frequency, duty factor, and a critical term (q), but validation was limited to young males and horses.
• The need for broader validation of predictive W(int) methods across diverse human populations and locomotion conditions is evident.

Purpose of the Study:

• To extend the validation of a predictive model for mechanical internal work (W(int)) in human locomotion.
• To compare model predictions (PW(int)) with experimentally measured W(int) (MW(int)) across varying gender, age, gait, velocity, and gradient conditions.
• To determine reference values for the critical term (q) under different locomotion scenarios.

Main Methods:

• Seventy healthy subjects (males and females, aged 6-65 years) performed level walking and running on a treadmill at various velocities.
• A subset of subjects (25-35 years) also underwent uphill and downhill walking and running trials.
• Model predictions (PW(int)) were compared against experimentally measured W(int) (MW(int)) to assess model accuracy and derive reference values for term q.

Main Results:

• The critical term (q) was found to be similar between males and females and independent of velocity and gradient.
• Reference values for q were determined: 0.08 for level locomotion and 0.10 for gradient conditions.
• Despite a ~20% underestimation compared to the previous model due to data filtering, the predictive model demonstrated a close match in trends with experimental data.

Conclusions:

• The predictive model for mechanical internal work (W(int)) is applicable across diverse human locomotion conditions, including variations in gender, age, gait, velocity, and gradient.
• The derived reference values for the critical term (q) provide a basis for estimating W(int) in populations and scenarios not previously studied.
• Further refinement of data processing methods may improve the accuracy of W(int) predictions.