Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Using cardiac Diffusion Tensor Imaging data for patient-specific modelling of the heart
The aim of this project is to implement an automatic identification and delineation pipeline of the right ventricle and to parameterize and map microstructural information from MRI data onto a Finite Element (FE) model.
Keywords: FE Modelling, Patient-specific Model, Biomedical Imaging, Diffusion Tensor Imaging
To run patient-specific simulations a biomechanical model needs to be personalized based on a set of input data. One aspect in such a multiscale model is the cardiac fiber architecture. This can be imaged non-invasively using cardiac Diffusion tensor imaging (cDTI), a Magnetic Resonance (MR) imaging technique. To populate a Finite Element Model (FEM model) with clinical and experimental data it is crucial to translate microstructural information from an image into the model. Therefore, image segmentation, the construction of a physiological meaningful local coordinate system and a mapping incorporating prior knowledge is required. Currently, the segmentation of the left ventricle of the heart (LV) is done using an automatic segmentation and data assimilation pipeline. It is the aim to extend the model to a more realistic biventricular one in the future.
To run patient-specific simulations a biomechanical model needs to be personalized based on a set of input data. One aspect in such a multiscale model is the cardiac fiber architecture. This can be imaged non-invasively using cardiac Diffusion tensor imaging (cDTI), a Magnetic Resonance (MR) imaging technique. To populate a Finite Element Model (FEM model) with clinical and experimental data it is crucial to translate microstructural information from an image into the model. Therefore, image segmentation, the construction of a physiological meaningful local coordinate system and a mapping incorporating prior knowledge is required. Currently, the segmentation of the left ventricle of the heart (LV) is done using an automatic segmentation and data assimilation pipeline. It is the aim to extend the model to a more realistic biventricular one in the future.
The objective of the student project is to implement an automatic tool, to identify and delineate the right ventricle. Based on this segmented geometry of the RV a physiological local coordinate system should be be defined and current techniques of parameterizing microstructure shall be employed in view of a personalized biventricular FE model of the heart.
The objective of the student project is to implement an automatic tool, to identify and delineate the right ventricle. Based on this segmented geometry of the RV a physiological local coordinate system should be be defined and current techniques of parameterizing microstructure shall be employed in view of a personalized biventricular FE model of the heart.
Supervisors
Johanna Stimm, stimm@biomed.ee.ethz.ch, ETZ F 61.2;
Christian Stoeck, stoeck@biomed.ee.ethz.ch, ETZ F 93
Professor
Sebastian Kozerke, kozerke@biomed.ee.ethz.ch, ETZ F94
Supervisors Johanna Stimm, stimm@biomed.ee.ethz.ch, ETZ F 61.2; Christian Stoeck, stoeck@biomed.ee.ethz.ch, ETZ F 93
Professor Sebastian Kozerke, kozerke@biomed.ee.ethz.ch, ETZ F94