lunduniversity.lu.se

Biomedical Engineering

Faculty of Engineering, LTH

Denna sida på svenska This page in English

Biomechanics

Interested?

If you are interested in the projects below, please contact Hanna Isaksson.

Computational model on bone fracture mechanisms

Objective: Develop and validate a computational 3D finite element (FE) model on bone fracture mechanisms for improved predictions of fractures.

Approach: The candidate will develop and validate a computational 3D FE model using the determined global and local strains, trabecular bone microstructures, and mineral density distributions from µCT images that have been acquired while bone deforms under compressive loading.

Application: Trabecular bone tissue from cadavers with normal and low density bone is used. Using combined micro tomographic imaging, in situ mechanical loading, and digital volume correlation (DVC), the global mechanical characteristics and local tissue strains of human trabecular bone have been assessed. After development and validation of the FE model, the roles of local tissue mineral density and trabecular bone structure to the local and global fracture resistance in trabecular bone are evaluated. The overall aim is to predict the mechanical integrity of trabecular bone tissue and bone material for a better understanding of the damage and fracture mechanisms.

Collaboration opportunities: The work is part of a collaborative project between Lund University (Lund, Sweden, Department of Biomedical Engineering), and University of Eastern Finland (Kuopio, Finland, Department of Applied Physics). The thesis supervisors will be Mikael Turunen, Ph.D., (Kuopio) and Associate professor Hanna Isaksson, Ph.D., (Lund). The work will be performed mainly at University of Lund.

Student background: Basic knowledge in finite element modeling, image analysis and solid mechanics.

Finite element models including a fracture mechanics approach to improve failure estimates at the human femur

Objective: to develop a fracture mechanics approach that predicts crack patterns and strength in human femurs.

Approach: the candidate will help in the development of a finite element modelling procedure for human femurs from clinical CT scans. The models will include (i) accurate morphological and densitometric mapping of cortical and trabecular bone tissue; (ii) consideration of fracture properties according to fracture mechanics theory. This new finite element modelling procedure will be assessed against experimental measurements obtained during tests to failure of a set of human cadaver femurs.

Collaboration opportunities: the work is part of a joint initiative of University of Lund (Lund, Sweden), Department of Biomedical Engineering, and Istituto Ortopedico Rizzoli (Bologna, Italy), Laboratorio di Tecnologia Medica. The thesis supervisors will be Lorenzo Grassi (Lund) and Enrico Schileo (Bologna). The work will be performed mainly at University of Lund, although a part of the activities may take place in Bologna.

Student background: Basic knowledge in finite element modelling, image analysis and solid mechanics

Functional imaging for detecting hip osteoarthritis - project 1

Objective: Develop an automatic segmentation method to distinguish the calcified tissues from the soft.

Approach: Create a 2D Statistical Shape Model (SSM) model that can be used to identify the region of interest.

Application: Merge this technique with an existing algorithm with a Statistical Appearance Model (SAM) that estimates the 3D shape of patient hips based on their available 2D radiographs. The information from the developed 2D Statistical Shape Model (SSM) can be used to improve the speed and performance of the existing optimization algorithm and SAM. The project adds value to existing research by removing the human error which is created when segmentation is performed manually.

Clinical relevance: Measure geometrical hip parameters in 3D that could detect patients that risk developing osteoarthritis later in life.

Student background: Basic knowledge in mechanics and biomechanics. In-depth knowledge and interest in mathematical modelling, image analysis and programming (mainly Matlab). 

Functional imaging for detecting hip osteoarthritis - project 2

Objective: Develop a new similarity measure for 2D-3D multi-modal image registration

Approach: Create and test new similarity measures that combine existing measures, such as Mutual Information and Cross-Correlation, normally adopted for analysis of multi-modal images.

Application: The method can be tested on existing hip radiographs (2D) and CT images of the hip (3D) of the same patient. It can further be tested in an optimization scheme where a 3D Statistical Appearance Model (SAM) is expected to adapt its shape to fit a 2D radiograph. Improved similarity measures for 2D-3D image registration has vast applications in many fields of medical science.

Clinical relevance: Measure geometrical hip parameters in 3D that could detect patients that risk developing osteoarthritis later in life.

Student background: Basic knowledge in mechanics and biomechanics. In-depth knowledge and interest in mathematical modelling, image analysis and programming (mainly Matlab). 

Functional imaging to improve osteoporosis diagnostics - project 1

Objective: Assess whether existing subject-specific finite element (FE) modeling procedures for human ones can be made independent from proprietary software.

Approach: Migrate an already validated FE modeling framework (Schileo et al., JBiomech 40(13), 2007) into open-source FE suites, while keeping an eye on the automation of the method.

Application: Several open-source software have been released for subject-specific modeling of human bones (ITK-snap, Seg3D2, FEBio, etc.). However, most of the state-of-the-art FE modeling approaches rely on proprietary software packages. Similar FE modeling approaches can be implemented using open-source software, and the obtained results will be thoroughly compared both to state-of-the-art FE methods, and to experimentally measured strain data.

Clinical relevance: an open-source implementation of reliable FE modeling techniques for human bones could help towards the introduction of FE modeling in osteoporosis diagnostics, aimed at fracture risk assessment.

Student background: Basic knowledge in finite element modelling, image analysis and solid mechanics 

Functional imaging to improve osteoporosis diagnostics - project 2

Objective: Develop an automatic method for the creation of finite element (FE) models of human femurs from computed tomography (CT) images.

Approach: Create a new, automatic, segmentation framework that is able to accurately reconstruct the 3D shape of a human femur from CT data.

Application: Different software platforms can be used to implement the code (Matlab, MevisLab, etc.). The code will be able to import a CT dataset, and through a series of operations obtain the 3D geometry of the femur. This geometry will then be discretized by generating a FE model. The obtained model will be compared to the state-of-the art, manual, FE models in terms of accuracy of 3D reconstruction, and FE element quality.

Clinical relevance: FE element models are a promising tool to better assess bone strength for elderly, but the time and skills required to generate a model impair their use in clinics. Having an automated method can help towards the introduction of the method in the clinical practice.

Student background: Basic knowledge in finite element modelling, image analysis and solid mechanics 

Simulate crack propagation in cortical bone with cohesive finite elements

Objective: Explore the potential of using cohesive finite elements to simulate crack propagation in cortical bone on the microscale and compare with existing models based on the eXtended Finite Element Method (XFEM).  

Approach: Create an automatic method for inserting cohesive elements into a mesh with “ordinary” finite elements. Evaluate possible damage models as well as mesh sensitivity, computational cost and convergence rate and compare with XFEM-models.

Application: The method can be tested on simplified microstructural models of cortical bone where Haversian canals, osteons and cement lines are the main components affecting crack propagation. 

Clinical relevance: We need better tools to model crack propagation in bone to understand the connection between bone quality and resistance to fracture.

Student background: Basic knowledge in solid mechanics with in-depth knowledge and interest in finite element modelling. This project will give hands-on experience with user-defined subroutines in Abaqus, which is one of the largest and most common commercial finite element softwares.