Fall risk assessment and system design based on an intelligent rehabilitation shoe
OU Jian-lin
YAN Yi-feng
CHEN Fang-ting
OUYANG Hui
LU Jian-liang
LIAO Cheng-qiang
GUAN Ye-yi
LIANG Zhi-hao
CHEN Zhuo-ming
Abstract：Objective To develop an intelligent shoe fall risk assessment system by integrating flexible sensor-based plantar pressure measurement,computer-aided feature extraction,and machine learning modeling.Methods Based on human foot biomechanics,an intelligent rehabilitation shoe system was designed to assess fall risk using plantar pressure data.Participants were categorized into high and low fall risk groups using a Berg Balance Scale cutoff score of 40.Features were extracted according to the weak foot hypothesis,and feature selection was performed using filter,wrap-per,and embedded methods.Multiple machine learning classifiers-including Logistic Regression(LogiR),k-Nearest Neighbors(KNN),Support Vector Machine(SVM),Decision Tree(DT),Random Forest(RF),Gradient Boosting Decision Tree(GBDT),and AdaBoost-were applied to classify fall risk.Finally,clinical validation was conducted using plantar pressure data collected from 48 independently ambulating elderly subjects.Results After feature selection,clas-sifier evaluation,and hyperparameter tuning,the optimal model achieved an accuracy of 87.5%,with a sensitivity of 100%,specificity of 75%,positive predictive value of 80%,and negative predictive value of 100%.Conclusion The intelligent rehabilitation shoe fall risk assessment system offers a convenient and efficient approach for integrating rehabilita-tion evaluation with computer technology,demonstrating significant potential for clinical application in fall risk prediction.
Keywords：smart rehabilitation shoefall riskdiagnosissystem design
Publication Date：2025-03-15
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：7( 350-356 )