Maximum-Entropy-Rate Selection of Features for Classifying Changes in Knee and Ankle Dynamics During Running

TL;DR

This study introduces a maximum-entropy-rate feature selection method for classifying changes in knee and ankle dynamics during running, demonstrating its effectiveness in estimating distance and energy expenditure from wearable sensor data.

Contribution

The paper presents a novel maximum-entropy-rate approach for selecting relevant features in classifying joint stiffness changes during running.

Findings

Distance traveled can be accurately estimated from knee and ankle motion data.

Energy expenditure correlates with observed joint dynamics.

The method improves classification accuracy over traditional feature selection techniques.

Abstract

This paper investigates deteriorations in knee and ankle dynamics during running. Changes in lower limb accelerations are analyzed by a wearable musculo-skeletal monitoring system. The system employs a machine learning technique to classify joint stiffness. A maximum-entropyrate method is developed to select the most relevant features. Experimental results demonstrate that distance travelled and energy expended can be estimated from observed changes in knee and ankle motions during 5 km runs.