Sexual Dimorphism in Plantar Pressure Distribution Patterns.

Area of Science:

• Biomechanics and Human Movement Analysis
• Orthopedics and Podiatry
• Computational Biology and Machine Learning

Background:

• Effective management of foot disorders necessitates understanding potential sexual dimorphism in foot structure and function.
• Existing knowledge regarding foot dimorphism remains inconclusive, prompting further investigation.
• Plantar pressure distribution is a key biomechanical indicator of foot function.

Purpose of the Study:

• To investigate sexual dimorphism in foot structure and function by analyzing multizone plantar pressure distribution.
• To test the hypothesis that sexual foot dimorphism influences plantar pressure patterns.
• To employ machine learning for accurate foot phenotype classification based on plantar pressure and anthropometric data.

Main Methods:

• Piezoelectric pedobarography was used to record multizone plantar pressure distribution in 298 healthy subjects.
• A k-nearest neighbors (k-NN) classifier, a supervised machine learning algorithm, was utilized for foot phenotype classification.
• Plantar pressure, anthropometric, and structural features were input into the k-NN model.

Main Results:

• Significant sex-specific foot dimorphism was identified in healthy individuals.
• The k-NN classifier accurately predicted sex based on foot features, demonstrating a low misclassification rate.
• Plantar pressure distribution on the great toe emerged as the most discriminative feature for sex classification.

Conclusions:

• The study confirms the presence of sexual foot dimorphism, with distinct plantar pressure patterns between sexes.
• Pedobarography combined with k-NN machine learning offers a highly accurate method for discriminating foot sexual phenotypes.
• The findings suggest that foot phenotypes exist beyond the gender binary, with potential implications for clinical management of foot disorders.