Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Investigation of bone regeneration during fracture healing in humans using time-lapsed advanced medical imaging
The ability for bone to repair itself after fracture is highly variable across the population. This project aims to understand why this is the case by building tools to quantify fracture healing in humans using state-of-the-art medical imaging, and to associate those outcomes to clinical biomarkers.
Keywords: Bone, Fracture Healing, Biomechanics, Medical Imaging, Biomarkers, Machine Learning, Clinical Research
Bone undergoes rapid change and remodelling after a traumatic fracture to rebuild the bone structure, but patient outcomes vary. Some patients recover quickly and form a stronger bone while others require longer recovery time and may not fully recover their bone’s original strength at all. High resolution CT has recently allowed us to visualise fractures in 3D as they repair with such high resolutions that local bone remodelling can be observed, including callus formation and mineralization. This presents an opportunity to study why some people heal better than others by exploring the interplay between bone regeneration, clinical metrics (e.g. motion, grip strength and pain), and biochemical markers of bone turnover in the blood.
However, methods to quantify bone characteristics as it heals are limited, especially for quantifying changes of the inflamed soft tissue surrounding the bone and the callus that drives bone regeneration. In this emerging field of research there is extensive opportunities to develop new methods based on conventional image processing or machine learning techniques.
The project will focus on advancing our understanding of fracture healing in humans and can be designed to meet the personal interests and skills of the student. This could include applying machine learning methods and advanced image processing techniques to segment and quantify bone, callus, and soft tissue during fracture healing. Alternatively, the interplay between bone regeneration parameters, biochemical markers, and clinical metrics can be explored to evaluate what metrics are the most predictive of patient outcomes.
Bone undergoes rapid change and remodelling after a traumatic fracture to rebuild the bone structure, but patient outcomes vary. Some patients recover quickly and form a stronger bone while others require longer recovery time and may not fully recover their bone’s original strength at all. High resolution CT has recently allowed us to visualise fractures in 3D as they repair with such high resolutions that local bone remodelling can be observed, including callus formation and mineralization. This presents an opportunity to study why some people heal better than others by exploring the interplay between bone regeneration, clinical metrics (e.g. motion, grip strength and pain), and biochemical markers of bone turnover in the blood.
However, methods to quantify bone characteristics as it heals are limited, especially for quantifying changes of the inflamed soft tissue surrounding the bone and the callus that drives bone regeneration. In this emerging field of research there is extensive opportunities to develop new methods based on conventional image processing or machine learning techniques.
The project will focus on advancing our understanding of fracture healing in humans and can be designed to meet the personal interests and skills of the student. This could include applying machine learning methods and advanced image processing techniques to segment and quantify bone, callus, and soft tissue during fracture healing. Alternatively, the interplay between bone regeneration parameters, biochemical markers, and clinical metrics can be explored to evaluate what metrics are the most predictive of patient outcomes.
This project is an opportunity to develop a deeper knowledge of bone as a mechanical and biological system. You will learn how to work with state of the art high-resolution medical imaging data and apply computational methods to evaluate bone (including morphometrics and potentially machine learning techniques).
This project is an opportunity to develop a deeper knowledge of bone as a mechanical and biological system. You will learn how to work with state of the art high-resolution medical imaging data and apply computational methods to evaluate bone (including morphometrics and potentially machine learning techniques).
The project is suited for a Bachelor or Master student with basic to expert knowledge in programming in Python (depending on your area of interest within the project). You will get access to our remote Jupyter Lab resources and many pre-developed tools that you can use alongside access to a dataset of real patient data. This work can accommodate home office, but physical workspace is available.
Please contact Dr. Danielle Whittier via email danielle.whittier@hest.ethz.ch if you are interested in learning more and discussing project ideas that would suit your interests.
The project is suited for a Bachelor or Master student with basic to expert knowledge in programming in Python (depending on your area of interest within the project). You will get access to our remote Jupyter Lab resources and many pre-developed tools that you can use alongside access to a dataset of real patient data. This work can accommodate home office, but physical workspace is available.
Please contact Dr. Danielle Whittier via email danielle.whittier@hest.ethz.ch if you are interested in learning more and discussing project ideas that would suit your interests.