Using partial least squares regression as a predictive tool in describing equine third metacarpal bone shape.

Area of Science:

• Biomechanical Engineering
• Comparative Anatomy
• Veterinary Orthopedics

Background:

• Subject-specific computational models require population variance considerations for accurate application.
• Understanding bone shape variability is crucial for biomechanical analyses in equine populations.

Purpose of the Study:

• To demonstrate the efficacy of partial least squares regression (PLSR) in improving shape prediction of the equine third metacarpal epiphysis.
• To show that PLSR can effectively utilize easily obtainable measurements for enhanced predictive accuracy.

Main Methods:

• Partial least squares regression (PLSR) was employed to model the relationship between anatomical measurements and bone shape.
• The study focused on the equine third metacarpal (cannon bone) epiphysis.

Main Results:

• PLSR provided an improved shape prediction of the equine third metacarpal epiphysis compared to traditional methods.
• The model successfully utilized two easily obtained measurements, highlighting the method's practicality.

Conclusions:

• Partial least squares regression offers a valuable tool for enhancing the accuracy of computational bone models in equine populations.
• This approach facilitates more reliable application of subject-specific models to broader populations by accounting for bone shape variance.