Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Image-Registration in Fractured Bone
Image registration is an essential tool in bone research, yet existing software can only deal with easy geometries like healthy bone or single fractures. Therefore, the present study will investigate possibilities to register 3D images with higher geometrical complexity.
Keywords: Micro-CT Imaging, Image Registration, Bone Fracture
With the resolution of current Micro-CT technologies, bone images nowadays have such a high resolution that structural changes of bone in time lapsed CT image series can be observed in great detail. This allows studying structural changes due to e.g. increased loading or other forms of treatment. Before this structural analysis can be performed, the different images have to be registered to account for e.g. alignment variations between the different CT scans. However, for images of fractures, the different bone fragments can also move with respect to each other between two scans and, additionally, large structural changes occur during fracture healing. Current registration algorithms are not designed to account for these additional challenges. In this study we thus want to develop a set of algorithms that can register CT images of complex fractures. These new algorithms could potentially be used to study the effects of treatments of bone fractures.
The student will develop algorithms that can register CT images of fractured bone (C++). He will then use them to analyse structural changes during bone fracture healing. The project is thus suited for students with an interest in computational science and image processing.
Task: 80% computational, 20% image processing
With the resolution of current Micro-CT technologies, bone images nowadays have such a high resolution that structural changes of bone in time lapsed CT image series can be observed in great detail. This allows studying structural changes due to e.g. increased loading or other forms of treatment. Before this structural analysis can be performed, the different images have to be registered to account for e.g. alignment variations between the different CT scans. However, for images of fractures, the different bone fragments can also move with respect to each other between two scans and, additionally, large structural changes occur during fracture healing. Current registration algorithms are not designed to account for these additional challenges. In this study we thus want to develop a set of algorithms that can register CT images of complex fractures. These new algorithms could potentially be used to study the effects of treatments of bone fractures.
The student will develop algorithms that can register CT images of fractured bone (C++). He will then use them to analyse structural changes during bone fracture healing. The project is thus suited for students with an interest in computational science and image processing.
Task: 80% computational, 20% image processing
Developing algorithms to register CT images of bone fractures to analyse structural changes during bone fracture healing.
Developing algorithms to register CT images of bone fractures to analyse structural changes during bone fracture healing.
Nicholas Ohs (nicholas.ohs@hest.ethz.ch), Institute for Biomechanics, ETH Zurich, Professorship Ralph Müller
Nicholas Ohs (nicholas.ohs@hest.ethz.ch), Institute for Biomechanics, ETH Zurich, Professorship Ralph Müller