Defining functional groups based on running kinematics using Self-Organizing Maps and Support Vector Machines.

Area of Science:

• Biomechanics
• Sports Science
• Human Movement Analysis

Background:

• Matching footwear to individuals can improve comfort, performance, and reduce injury risk.
• Defining functional groups by movement patterns is key to personalized footwear.
• Current grouping methods often overlook distinct movement patterns.

Purpose of the Study:

• To propose and apply a methodological approach for defining functional groups based on movement patterns.
• To identify differences in age, gender, and comfort preferences among these functional groups.
• To highlight the importance of pattern recognition for personalized product development.

Main Methods:

• Utilized Self-Organizing Maps (SOM) and Support Vector Machines (SVM) for pattern recognition.
• Analyzed kinematic data and subjective comfort preferences from 88 participants (aged 16-76).
• Defined eight distinct functional groups based on unique movement patterns.

Main Results:

• Eight functional groups with distinctive movement patterns were successfully identified.
• Significant differences in age and gender were observed across several groups.
• Comfort preferences varied, indicating group-specific footwear requirements.
• Some groups with unique footwear needs did not differ by age or gender, underscoring limitations of traditional grouping.

Conclusions:

• The proposed pattern recognition approach effectively defines functional groups by movement patterns.
• This method can uncover specific user needs missed by age or gender-based classifications.
• Tailoring footwear based on identified functional groups can enhance comfort and potentially reduce injuries.