Katherine M. Steele

Human motion data expansion from arbitrary sparsesensors with shallow recurrent decoders

This paper develops a deep learning method to reconstruct dense human motion data from sparse sensors, enabling accurate tracking of unmonitored body segments and improving digital biomarker estimates.

Quantifying changes in individual-specific template-based representations of center-of-mass dynamics during walking with ankle exoskeletons using Hybrid-SINDy

This study uses Hybrid-SINDy to analyze changes in center-of-mass dynamics during walking with ankle exoskeletons in unimpaired adults and a post-stroke individual, finding no changes in unimpaired dynamics but increased rotary stiffness in the paretic leg with exoskeletons.

A machine learning approach to quantify individual gait responses to ankle exoskeletons

This study uses neural networks to quantify individual gait responses to ankle exoskeletons by modeling discrepancies between nominal and exoskeleton gait data.

Muscle synergies are flexibly recruited during gait pattern exploration using motor control-based biofeedback

This study investigates how synergies are recruited during gait pattern exploration using motor control-based biofeedback, finding that synergy complexity decreases for most patterns, with distal mechanics predicting these changes.

Predicting walking response to ankle exoskeletons using data-driven models

This study evaluates three data-driven phase-varying models to predict kinematic and myoelectric responses to ankle exoskeletons during walking, finding LPV models provide better accuracy than PV models for kinematic predictions.