Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
“Off-road” MRI - magnetic resonance imaging under harsh eddy current conditions.
A master project for a student of physics, electrical engineering or computer science involving experimental, theoretical and programming work in the domain of ultra-fast magnetic resonance imaging technology.
Keywords: MRI, image reconstruction, magnetic field monitoring, linear system characterization.
Magnetic Resonance Imaging (MRI) requires a very high stability and homogeneity of the magnetic field during signal acquisition. However, with ultra-fast MRI methods, which require rapid switching of strong gradient fields, these conditions cannot be guaranteed. This is the case with our newly developed gradient system producing up to 200 millitesla per meter (about seven time more than clinical scanners), a unique piece of equipment that allows dynamic MRI with hundreds of frames per seconds. With such extreme gradients and scanning speed, successful image reconstruction requires precise monitoring of spurious fields caused by eddy currents, as we have recently demonstrated. We continue working on this strategy to make it more versatile and faster.
Your contribution will be to replace the currently used procedure of field monitoring for each gradient pulse sequence and every imaging geometry, which is too long for clinical practice, by a prediction of the dynamic field based on a linear time-invariant model. This will require a series of experiments to measure impulse responses of high-order field components, and extending our reconstruction software by the calculation of dynamic fields for any sequence and geometry based on these results. Further investigation options include the dimensionality reduction to minimize the number of necessary impulse responses and ways to deal with nonlinear effects in parts of the gradient system.
Magnetic Resonance Imaging (MRI) requires a very high stability and homogeneity of the magnetic field during signal acquisition. However, with ultra-fast MRI methods, which require rapid switching of strong gradient fields, these conditions cannot be guaranteed. This is the case with our newly developed gradient system producing up to 200 millitesla per meter (about seven time more than clinical scanners), a unique piece of equipment that allows dynamic MRI with hundreds of frames per seconds. With such extreme gradients and scanning speed, successful image reconstruction requires precise monitoring of spurious fields caused by eddy currents, as we have recently demonstrated. We continue working on this strategy to make it more versatile and faster.
Your contribution will be to replace the currently used procedure of field monitoring for each gradient pulse sequence and every imaging geometry, which is too long for clinical practice, by a prediction of the dynamic field based on a linear time-invariant model. This will require a series of experiments to measure impulse responses of high-order field components, and extending our reconstruction software by the calculation of dynamic fields for any sequence and geometry based on these results. Further investigation options include the dimensionality reduction to minimize the number of necessary impulse responses and ways to deal with nonlinear effects in parts of the gradient system.
The ideal candidate should have interest in medical imaging technology, a good background in signal processing and affinity to programming. Knowledge of the basics of MRI will be an important plus. We offer an opportunity to develop this knowledge further by experimenting with a uniquely equipped MRI system, and to gain skills in image reconstruction methods through an ambitious research project.
The ideal candidate should have interest in medical imaging technology, a good background in signal processing and affinity to programming. Knowledge of the basics of MRI will be an important plus. We offer an opportunity to develop this knowledge further by experimenting with a uniquely equipped MRI system, and to gain skills in image reconstruction methods through an ambitious research project.
Supervisor: Dr. Franciszek Hennel, hennel@biomed.ee.ethz.ch
Professor: Prof. Dr. Klaas Pruessmann
Supervisor: Dr. Franciszek Hennel, hennel@biomed.ee.ethz.ch Professor: Prof. Dr. Klaas Pruessmann