Surrogate-Guided Motion-Compensated Reconstruction for Medical Imaging

• Applicant: Dr. Christoph Kolbitsch
• Subject Area: Medical Physics, Biomedical Technology
• Term: from 2016 to 2018
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 317231546

Magnetic resonance imaging (MRI) is a highly versatile medical imaging technique which provides 3D high-resolution images with excellent soft tissue contrast. MRI is applied in a wide range of different fields, such as cancer, neurology and cardiology. Despite its long-time use in clinical practice, physiological organ motion due to breathing is still a mayor challenge especially in 3D cardiac MRI. It can negatively impair image quality and even lead to scan abortions.The aim of this project is to correct for respiratory motion during cardiac MR scans with a motion compensated reconstruction and external motion sensors. These respiratory belts yield high temporal resolution signals which optimally complement the high spatial resolution of MR information. However, respiratory belts are so far only used qualitatively to synchronise data acquisition with organ motion. Here we want to calibrate these qualitative signals with quantitative MR measurements of organ movement in the individual patient and utilise this information to correct for physiological motion. We will apply multiple respiratory belts to ensure that the complex respiratory motion of the heart is accurately captured. The obtained motion information will be used in a motion-compensated image reconstruction approach to transform all data to one motion state, i.e. to calculate the image in one phase of the respiratory cycle. This will yield one motion-corrected high quality image utilising all the available data in a very efficient approach. Uncertainties in motion estimation will be assessed and minimised using additional constraints during image reconstruction.The feasibility of the developed techniques will be demonstrated in volunteers and patients. Prof. Dr. Schulz-Menger, head of the cardiac MRI group at the Experimental and Clinical Research Center (ECRC) at Charité Berlin has assured her full support and will provide clinical guidance. The project will be carried out by a student in the first year of their PhD project. Funding for year 2 and 3 will either be obtained from external funding or be provided by the PTB. A motion compensation approach based on external surrogate information will improve image quality in a wide range of different cardiac MRI applications ensuring short and reliable scans. Furthermore, it can be extended to respiratory motion of the abdomen using simultaneous Positron-Emission-Tomography (PET)-MR.

• DFG Programme: Research Grants
• Co-Investigator: Professor Dr. Tobias Schäffter