Auckland Bioengineering Institute


Musculoskeletal System current projects

The Musculoskeletal Atlas Project (MAP)


Computational models have tremendous potential to complement clinical trials, reducing the cost of bringing a new product to market through in silico testing across a virtual population. However, the predictive power of computational models is dependent on the ability to accurately capture the complex geometry of the musculoskeletal system and the ability to describe appropriate loads and boundary conditions (i.e. muscle and joint forces).

Project goal: to develop a population-based anatomical and functional atlas of the human musculoskeletal system to generate computational models for virtual clinical trials. The key deliverable from this proposal is an open-source software platform to rapidly generate accurate, surface meshes of the bones, muscles, and soft tissue of the lower limb, using statistical shape models extracted from a large imaging dataset (https://simtk.org/home/map).

We have leveraged existing software tools developed for the IUPS Physiome Project to facilitate image segmentation, meshing, and data query. To facilitate exchange of data and models, accelerate validation and provide robust peer review, the MAP database has been implemented within the Physiome Model Repository. Using these software tools, we are actively engaging the biomechanics community to build the MAP database and provide a unique resource for population-based orthopaedic modeling and in silico testing and validation.

Figure 1: The Musculoskeletal Atlas Project (MAP) framework.
Figure 1: The Musculoskeletal Atlas Project (MAP) framework.

In Figure 1, an open-source, python-based desktop application (MAP Client) is used as a front-end to input imaging and functional data into a database. Using statistical shape models from imaging datasets, we rapidly fit 3D anatomical models of the musculoskeletal system to match experimental data. Those models can be stored in the MAP Database and used to perform statistical-based population studies (MAP Query). Final models can be exported to a number of formats consistent with other open-source and commercial computational codes.
 

Figure 2: An example workflow which uses the MAP Client to generate a subject-specific musculoskeletal model, compatible with OpenSim.
Figure 2: An example workflow which uses the MAP Client to generate a subject-specific musculoskeletal model, compatible with OpenSim.

Project leader

Portrait of

Thor Besier
Associate Professor
Email: t.besier@auckland.ac.nz

Funding partner

US Food and Drug Administration (BAA-12-00118)

Researchers and collaborators

Top

Changing the way we move: a novel framework for motion retraining


Movement disorders such as stroke, cerebral palsy and osteoarthritis dramatically impact quality of life. In walking these disorders can alter muscle and joint forces leading to rapid joint degeneration. An effective and efficient movement-retraining tool would enable patients to resume activities of daily living and reduce the burgeoning cost of health care.

Project goal: The goal of this project is to investigate the use of dielectric elastomers (artificial muscle) to provide accurate and efficient haptic (touch) feedback to patients with movement disorders.
 

Workflow for providing real-time haptic feedback using novel sensors and actuators.
Workflow for providing real-time haptic feedback using novel sensors and actuators.

Inertial sensors (IMU’s) are placed on lower limb segments to provide estimates of joint kinematics. Together with electromyography (EMG), these data are used within a surrogate contact model that includes muscle force estimates using an EMG-driven musculoskeletal model. Estimates of peak contact force or pressure are then used in a data-driven gait prediction model to provide a stimulus to the subject in real-time to alter their gait. The stimulus is applied as skin stretch or vibration using dielectric elastomer (artificial muscle).
 

Project leader

Portrait of

Thor Besier
Associate Professor
Email: t.besier@auckland.ac.nz

Researchers and collaborators

  • Daniel Chen
  • Mousa Kazemi
  • Iain Anderson

Funding partner

Marsden Fund (UOA-1211), administered by the Royal Society of New Zealand

Top

Virtual testing of orthopaedic devices


Successful designs of total joint replacement need to be robust to patient and surgery-related variability. Significant morphometric and material property variations are known to exist between individuals. This, coupled with the significant variation in the placement and alignment of components and soft tissue balancing can mean that an implant is subjected to a diverse range of mechanical environments.

Accounting for this variability needs to be an integral part of the design and development process and for product validation.

Project goal: to assess how a joint replacement is likely to behave when subjected to these diverse conditions, prior to clinical trial.

Advances in computational modeling, from design of computer experiments, through probabilistic analyses and to population based modelling have significant potential to improve the robustness of orthopaedic implant design.

Workflow of musculoskeletal models to estimate loads and boundary conditions for orthopaedic applications.
Workflow of musculoskeletal models to estimate loads and boundary conditions for orthopaedic applications.

In the figure above, medical imaging data provide accurate geometry of muscles and bones, which are then coupled with rigid body dynamics analyses to estimate kinematics and muscle and joint contact forces. These loads and boundary conditions are used in a finite element model, along with estimates of material properties, to estimate tissue-level stresses and strains.
 

Project leader

Portrait of

Thor Besier
Associate Professor
Email: t.besier@auckland.ac.nz


Funding partners

  • DePuy Orthopaedics
  • Ossis Orthopaedics
  • Australian Research Council (ARC) (LP130100122)
  • Ministry of Business, Innovation and Employment (MBIE) (UOAX1407)

Researchers and collaborators

  • Mark Taylor (PI, Flinders University)
  • Ju Zhang
  • Justin Fernandez
  • Peter Hunter
  • Chris Bradley
  • Vickie Shim
  • Andi Liu
Top

ICON Bionic Joint: Human joint sensing and actuation in preventive ergonomics and rehabilitation therapy & monitoring in an ageing society


This NZ-German collaboration will enable the first generation of highly integrated sensing and actuation technology for human exoskeletons. Latest flexible polymer-based soft sensing technology will be combined with computational models and high torque actuation and control technology to develop a novel, lightweight upper arm exoskeleton. The exoskeleton will be designed for reducing workplace musculoskeletal injuries and accelerating the recovery of upper arm function in stroke survivors, both of which have tremendous impact in our ageing society.

ICON Bionic Joint
Subject specific musculoskeletal models will be coupled with wearable sensors to provide an intuitive real-time controller for a robotic exoskeleton.

 

Project leader

Portrait of

Thor Besier
Associate Professor
Email: t.besier@auckland.ac.nz


Funding partners

  • Ministry of Business, Innovation and Employment (MBIE)
  • The University of Auckland

Researchers and collaborators

  • Peter Hunter
  • Massoud Alipour
  • Thorben Pauli
  • Ted Yeung
  • Desney Greybe
  • Urs Schneider, Fraunhofer and Stuttgart University, Germany
  • Oliver Röhrle, Fraunhofer and Stuttgart University, Germany
Top

Assistive technologies for stroke rehabilitation


A stroke is a leading cause of adult disability worldwide. In New Zealand, approximately 7,800 people experience a stroke each year, and more than two-thirds of the country’s 64,000 stroke survivors require assistance with activities of everyday living. Impaired movement is common after stroke, and recovery of upper limb function is vital to regaining independence.

The use of wearable sensors to measure and monitor stroke rehabilitation is an exciting area and one that is receiving much attention due to our ageing population.

Project goal: to develop assistive technologies for stroke rehabilitation to measure and monitor the upper limb and control robotic exoskeletons.

Our framework will combine wearable sensors (IMUs and stretch sensors) with musculoskeletal models to derive kinematic and kinetic data for the upper limb.

By linking wearable sensor technology, such as IMU’s and stretch sensors with a musculoskeletal model, this project promises to provide a low-cost quantitative assessment of stroke patients in their home environment.
By linking wearable sensor technology, such as IMU’s and stretch sensors with a musculoskeletal model, this project promises to provide a low-cost quantitative assessment of stroke patients in their home environment.

 

Project leader

Portrait of

Thor Besier
Associate Professor
Email: t.besier@auckland.ac.nz


Funding partner

Medical Technologies Centre of Research Excellence (MedTech CoRE)

 

Researchers and collaborators

  • Marcus King, Callaghan Innovation
  • Denise Taylor, School of Clinical Sciences, Auckland University of Technology
  • Nicola Kayes, School of Clinical Sciences, Auckland University of Technology
  • Winston Byblow, Department of Sport and Exercise Science, University of Auckland
  • Cathy Stinear, School of Medicine, University of Auckland
  • Angus McMorland, Department of Sport and Exercise Science, University of Auckland
  • Edgar Rodriguez, School of Design, Victoria University of Wellington
  • Simon Fraser, School of Design, Victoria University of Wellington
Top

Musculoskeletal modelling of children with Cerebral Palsy


Cerebral palsy (CP) is a common neurological disorder affecting one in every 325 children. CP is caused by abnormal development or damage to the brain and causes deficits in motor control, balance and posture, as well as structural impairments to muscles and bones. Many techniques exist to help children overcome limitations in mobility, including rehabilitation, orthoses, medication, and surgery.

Project goal: to improve the outcome of these interventions by using neuromusculoskeletal modelling. By developing better anatomical and functional models of skeletal muscle specific to CP, we will create a framework to assist decision-making and improve the process of prescription and design of interventions for children with CP.

illustration-cerebral-palsy-project
Heterogeneous patterns of muscle weakness in children with Cerebral Palsy

 

Project leader


Funding partners

  • Whitaker Foundation
  • Medical Technologies Centre of Research Excellence (MedTech CoRE)
  • Wishbone Trust

Researchers and collaborators

  • Julie Choisne
  • Thor Besier
  • Justin Fernandez
  • David Lloyd, Griffith University, Australia
  • Chris Carty, Griffith University, Australia
  • Morgan Sangeux, Murdoch Childrens Research Institute, Melbourne, Australia
Top

Modelling outcomes of Femoral Acetabular Impingement (FAI) surgery


Modelling outcomes of FAI
Predicted hip joint cartilage stress during the stance phase of gait.

Femoroacetabular impingement (FAI), estimated to affect 10-25% of the general population, is a condition where the bones of the hip are abnormally shaped leading to mechanical impingement and hip pain in young active adults.

Although untested, the repetitive mechanical impingement in movement and locomotion is believed to lead to labral and chondral stresses that cause irreversible structural pathology. It is also believed that this process may be responsible for the development of over 90% of the cases of hip osteoarthritis. 

Project goal: to use musculoskeletal modelling to understand the hip joint contact mechanics in patients who have undergone hip arthroscopic surgery for FAI.

Project leader

Portrait of

Thor Besier
Associate Professor
Email: t.besier@auckland.ac.nz


Funding partner

Australian National Health and Medical Research Council (NHMRC) (APP1069278)

Researchers and collaborators

  • David Hunter (PI), University of Sydney, Australia
  • David Lloyd, Griffith University, Australia
  • Justin Fernandez
  • Ju Zhang
  • Vickie Shim
Top

Improving surgical management of complex hip deformities


Improving surgical management of hip deformity

Surgical management of complex hip deformities in adolescents & young adults focuses on technically demanding femoral and pelvic reconstructive osteotomies to re-align the femoral head inside the hip socket. Such osteotomies are believed to improve force distribution around the hip joint, and thus, delay the need for total joint replacements into middle adulthood. However, predicting the functional effect of surgical osteotomies on joint forces is difficult and clinical outcomes remain variable.

Project goal: to develop a simulation-based educational software application based on biomechanical principles to model surgical management of complex hip deformities.

Workflow to simulate hip joint contact pressures in patients with hip deformities.

Project leader

Portrait of

Thor Besier
Associate Professor
Email: t.besier@auckland.ac.nz

Researchers and collaborators

  • Susan Stott, Orthopaedics Surgery, University of Auckland
  • Andrew Graydon, Orthopaedics Surgery, University of Auckland
  • Alex Carleton
  • Vickie Shim
Top

Osteoarthritis of the CMC (thumb) joint


The saddle-shaped articulation at the trapezium’s terminal post defines the large and complex range of motion of the first carpometacarpal (CMC) joint. Such range of motion permits the thumb to be finely controlled and coordinated to perform precise motion in concert with the fingers, such as pinching and grasping.

The CMC joint is also a common site of osteoarthritis in the hand, causing pain and disability, ultimately, requiring surgery.

Project goal: to determine in vivo articular cartilage stress at the CMC joint using bone geometry and joint orientation obtained from CT images.

Osteoarthritis of the CMC joint
Workflow to automatically generate finite element meshes of the trapezium and first metacarpal using statistical shape modelling.

Project leader

Portrait of

Thor Besier
Associate Professor
Email: t.besier@auckland.ac.nz


Funding partners

  • Stanford University Department of Orthopaedics
  • Brown University Department of Orthopaedics
  • Auckland Bioengineering Institute

Researchers and collaborators

  • Marco Schneider
  • Kumar Mithraratne
  • Poul Nielsen
  • Joseph “Trey” Crisco, Dept Orthopaedics, Brown University, USA
  • Amy Ladd, Dept Orthopaedics, Stanford University, USA
Top

Joint mechanics of the equine fetlock joint


Joint mechanics of equine fetlock joint
Proposed mechanical aetiology of fetlock joint fracture.

Fractures of the cannon and other bones in the fetlock joint are common, can have tragic consequences, and are related to distances and speeds at which Thoroughbreds race and train. 

A new approach is to develop a technology array that can predict fetlock joint fracture risk in an individual horse, based upon shape, bone density and forces acting on the joint.  Such models have enormous potential as clinical tools to prevent catastrophic injury.  

Project goal: to develop the first component of the array, a statistical Shape Model, to predict the size and shape of the joint surfaces, and bone mineral densities at all points beneath the joint surfaces, using data from routine digital imaging.

This project focuses on using statistical shape modelling to characterise the bone morphology and material properties. Finite element models will then be used to determine mechanical stress-strain and predict failure.
 

Project leaders

Portrait of

Justin Fernandez
Senior Research Fellow
Email: j.fernandez@auckland.ac.nz

Portrait of

Thor Besier
Associate Professor
Email: t.besier@auckland.ac.nz

Researchers and collaborators

  • Helen Liley
  • Helen Davies, Melbourne University, Australia
Top

Mapping and predicting femur trabecular structure


Preventing and treating hip fractures is an ever growing challenge in an ageing population. Capturing the complex micro-structure of trabecular bone in the femur can lead to more accurate predictions of hip strength and fracture risk. However, the micro-scale structure is beyond the capabilities of clinical imaging systems.

Project Goal: Using statistical methods, predict trabecular micro-structure from clinical measurements to improve hip strength predictions. We will first map the variations in trabecular structure throughout the femur and across a population using a set of high-resolution CT image. Statistical models will then predict the trabecular structure map of an individual from routine clinical images and patient information.

Three scans of the femur labelled a,b and c. a is Clinical CT, b is Micro-CT study, c is Tensor Field of Anisotropy. belled a,b and c. a is Clinical CT, b is Micro-CT study, c is Tensor Field of Anisotropy.
Current clinical imaging of the femur (a) cannot capture the micro-structure of trabecular bone (b). By mapping the trabecular structure (c) from a population of high-resolution CT images, we can derive statistical models that predict trabecular structure from clinical resolution CT images.

Project leader

Portrait of

Ju Zhang
Research Fellow
Email: ju.zhang@auckland.ac.nz

Funding Partners

  • University of Auckland Faculty Research Development Fund

Researchers and Collaborators

  • Michelle Deacon
Top