Predicting Center of Pressure Velocity Based on Regional Plantar Force in Elderly Men Using Artificial Neural Networks

Abstract: This study aimed to develop an artificial neural network (ANN) model to predict the forward velocity of the center of pressure (COP) in reference to regional plantar force in older adults and verify its validity. Plantar pressure variables in eight artificially divided anatomical areas were recorded barefoot with a Footscan plantar pressure plate system in sixteen community-dwelling males over sixty-five. The ANN was employed to build a predictive model for the velocity of COP based on measured regional plantar force information. The validation test showed that the determination coefficient (R2) for the forward velocity of COP were 0.65 for hindfoot, 0.66 for midfoot, and 0.38 for the forefoot. These results add additional insights into the basis for building an ANN model for gait analysis and pathologic evaluation in older adults. Relevant information may be necessary for clinical applications, such as further elucidating the causes of common age-related foot diseases.