Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Cardiac Magnetic Resonance Image Reconstruction Using Machine Learning
Dynamic Magnetic Resonance (MR) imaging offers exquisite views of cardiac anatomy and function. The objective of this project is to develop and implement methods that allow learning a data model from large sets of training data to be used in nonlinear data recovery from highly undersampled MR data.
Time-resolved cardiac Magnetic Resonance (MR) imaging is the reference standard for assessing cardiac anatomy and function. Advanced image reconstruction approaches permit image recovery from highly undersampled data and hence enable improved spatiotemporal resolution during clinically feasible scan durations. A challenge of image recovery from undersampled data is related to residual image artefacts. To this end, model-based non-linear image reconstruction techniques have been devised.
Time-resolved cardiac Magnetic Resonance (MR) imaging is the reference standard for assessing cardiac anatomy and function. Advanced image reconstruction approaches permit image recovery from highly undersampled data and hence enable improved spatiotemporal resolution during clinically feasible scan durations. A challenge of image recovery from undersampled data is related to residual image artefacts. To this end, model-based non-linear image reconstruction techniques have been devised.
The objective of the present project is to develop and implement methods that allow learning a data model from large sets of training data to be used in nonlinear data recovery. Analysis of generalization ability and reproducibility of the developed approach should be emphasized. Simulated and in-vivo MR data of the heart are available to test and validate the method. The project entails:
• Simulation of undersampled MR data acquisition and reconstruction
• Adaptation and application of dictionary/machine learning on large training sets
• Implementation and testing of model-based non-linear data recovery
• Application to highly undersampled clinical imaging data of the heart
Programming experience in Matlab or Python is required.
The objective of the present project is to develop and implement methods that allow learning a data model from large sets of training data to be used in nonlinear data recovery. Analysis of generalization ability and reproducibility of the developed approach should be emphasized. Simulated and in-vivo MR data of the heart are available to test and validate the method. The project entails:
• Simulation of undersampled MR data acquisition and reconstruction • Adaptation and application of dictionary/machine learning on large training sets • Implementation and testing of model-based non-linear data recovery • Application to highly undersampled clinical imaging data of the heart
Programming experience in Matlab or Python is required.
Supervisors: Dr. Valery Vishnevskiy (vishnevskiy@biomed.ee.ethz.ch); Dr. Johannes Schmidt (schmidtjf@gmail.com); Prof. Dr. Sebastian Kozerke (kozerke@biomed.ee.ethz.ch)
Supervisors: Dr. Valery Vishnevskiy (vishnevskiy@biomed.ee.ethz.ch); Dr. Johannes Schmidt (schmidtjf@gmail.com); Prof. Dr. Sebastian Kozerke (kozerke@biomed.ee.ethz.ch)