Machine learning model for predicting shear forces at the body-seat interface in wheelchair users: A novel approach

Area of Science:

• Biomechanics
• Machine Learning
• Rehabilitation Engineering

Background:

• Shear forces at the body-seat interface are a significant mechanical factor contributing to pressure injuries.
• Accurate prediction of these forces is essential for developing effective prevention strategies.

Purpose of the Study:

• To introduce and evaluate a novel machine learning approach for predicting shear forces at the body-seat interface.
• To assess the model's performance across individuals without disabilities and wheelchair users.

Main Methods:

• A supervised learning model, specifically Random Forest Regression, was employed.
• Data from pressure mapping systems and an experimental seat were utilized.
• The model was trained on data from individuals without disabilities and validated on both this group and wheelchair users.

Main Results:

• The Random Forest Regression model achieved promising accuracy, with an average error below 20% for individuals without disabilities and some wheelchair users.
• Model performance decreased for wheelchair users with significantly lower shear forces.
• Key input features included a calculated variable, backrest force, feet normal force, seat pan force, backrest area, and backrest center of pressure location.

Conclusions:

• The developed machine learning model shows potential for predicting shear forces relevant to pressure injury prevention.
• Further dataset expansion with diverse anthropometric characteristics and seated postures is needed to enhance model generalizability.
• Continued research is vital for improving predictive accuracy across all wheelchair user populations.