Scribd
Upload
58 views18 pages

Pediatric Exoskeleton Design Optimization

This document describes the design of a pediatric exoskeleton for children with neuromuscular disorders like cerebral palsy. The researchers created a computational model to optimize exoskeleton parameters like spring stiffness and pulley radii based on a child's weight and leg length. Their model was verified against experimental data. Parameter optimization found that varying the hip and ankle pulley radii provided the best fit to normative joint moments, while force was highly dependent on spring stiffness but independent of knee pulley radius. This suggests a device with adjustable hip and ankle pulleys would be most effective at fitting individual patients. Future work includes improving the model with experimental pediatric joint data.

Uploaded by

api-272723910
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
58 views18 pages

Pediatric Exoskeleton Design Optimization

This document describes the design of a pediatric exoskeleton for children with neuromuscular disorders like cerebral palsy. The researchers created a computational model to optimize exoskeleton parameters like spring stiffness and pulley radii based on a child's weight and leg length. Their model was verified against experimental data. Parameter optimization found that varying the hip and ankle pulley radii provided the best fit to normative joint moments, while force was highly dependent on spring stiffness but independent of knee pulley radius. This suggests a device with adjustable hip and ankle pulleys would be most effective at fitting individual patients. Future work includes improving the model with experimental pediatric joint data.

Uploaded by

api-272723910
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 18

Design of a Pediatric Exoskeleton for

Individuals with Neuromuscular


Disorders through Computational
Parameter Optimization

Jessica Zistatsis1, Per Lukas Hillerstrom2, Lauren Martini2, Mackenzie Andrews2


1
Mechanical Engineering, 2Bioengineering

BIOEN 485: Computational Bioengineering (Spring 2017), Dr. Wendy Thomas


University of Washington
Background
Cerebral Palsy (CP)

● Neuromuscular disorder
● Cause motor delays → gait
impairments
● Impacts ~3/1000 children per year1
● Children with CP require walking
training2

1. Data & Statistics for Cerebral Palsy [WWW Document], 2016. [WWW Document]. Centers for Disease Control and Prevention. URL https://www.cdc.gov/ncbddd/cp/data.html (accessed 2.17.16).
2. Garvey, M.A., Giannetti, M.L., Alter, K.E., Lum, P.S., 2007. Cerebral palsy: New approaches to therapy. Current Neurology and Neuroscience Reports 7, 147–155. doi:10.1007/s11910-007-0010-x
Exotendon System
Device Parameters

● Spring Stiffness - k (N*m)


Hip Pulley
● Ankle Pulley Radius - rAnkle (mm)
● Knee Pulley Radius - rKnee (mm)
Exotendon
● Hip Pulley Radius - rHip (mm)
● Preload (N) - initial stretch (force)
Knee Pulley
applied to the spring
Also:
● Slack Length (mm) - determined by
Ankle Pulley
preload, so only preload is optimized
during parameter optimization
Goals

● Create computational model for exoskeleton


○ Inputs: Child weight (kg), child leg length (m)

○ Outputs: Optimized parameters, exotendon force during gait cycle with


optimized parameters

● Use the model to inform device design and clinical implementation


○ Determine optimal device parameters for various patient weights and leg
lengths

○ Use parameter sensitivities to determine device component properties

■ i.e. Should the pulleys have variable radii? If so, which pulleys?
Hypothesis

● The model will be best optimized by varying spring stiffness and leaving
pulley radii constant

● This would allow for a simple in-clinic modification to fit a range of patient
sizes and needs
Van den Bogert3 Model

3. Van den Bogert, A.J., 2003. Exotendons for assistance of human locomotion. BioMedical Engineering OnLine 2. doi:10.1186/1475-925X-2-17
Verifications: van den Bogert

● Experimental force and moment data for an adult exoskeleton

Mathematical Excel model: Our MATLAB model:


Comparison with van den Bogert

van den Bogert: Our model:


Parameter Optimization - Fitting to Normative Moments

● Impaired data generated with scaling


factors obtained from literature

Impaired data = unimpaired moments *


scaling factor for each joint moment4

● Objective function: minimize


difference between normative
moments and assisted patient
moments

Residual = Normative Moments -


(Impaired Moments + Exotendon Moments)

4. Lam, W., Leong, J., Li, Y., Hu, Y., Lu, W., 2005. Biomechanical and electromyographic
evaluation of ankle foot orthosis and dynamic ankle foot orthosis in spastic cerebral palsy. Gait &
Posture 22, 189–197. doi:10.1016/j.gaitpost.2004.09.011
Verifications: unimpaired patients
● Unimpaired patients should require no assistance → exotendon
moments should be zero for adults and children
Parameter Optimization
Varying child weight and leg length
Parameter and Force Sensitivities

Spring Stiffness:
Parameter and Force Sensitivities

Ankle Pulley Knee Pulley Hip Pulley

Note the difference in y-axis scale → knee << ankle, hip


Clinical Application
● Interpreting fractional changes between child size and parameters
Conclusions

● Force is highly dependent on spring stiffness


● Varying hip and ankle pulley radii provide for a more optimized gait
● Force optimization is nearly independent of knee pulley radius

This infers...

● A device with variable hip and ankle radii would be most effective for fitting
a patient
Future Work

● Create dataset of impaired pediatric joint angles through experimental


measurement
○ Integrate with model - replace scaling currently used, remove normative joint angle
assumption

● Improve optimization function to accommodate increases in muscle force


(currently assume that impairment → decrease in force)
● Add weighting in objective function for the different parameters
Thank you! Any questions?

You might also like