Motion compensation for magnetic resonance imaging with compressed-sensing reconstruction

• Applicants: Professor Dr. Alex Peter Frydrychowicz; Professor Dr.-Ing. Alfred Mertins
• Subject Area: Medical Physics, Biomedical Technology
• Term: from 2016 to 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 290129674

Patient motion during magnetic resonance imaging (MRI) data acquisition leads to unwanted image artifacts that may be so severe that the entire measurement becomes unusable for diagnosis. Based on previous works of the applicants and the principle of compressed sensing, we intend to develop and test novel reconstruction methods that include efficient motion estimation and compensation. The medical focus will be laid on abdominal and head/neck MRI. The goal is to allow patients to swallow and catch breath occasionally or to breathe freely without severely affecting the image quality. At the center of the proposed approach, multidimensional complex wavelet transforms will be applied for both compressed sensing based image reconstruction and the estimation of the motion vector fields that occur in the above mentioned body regions. One particularly suited complex wavelet transform was developed in previous works by the applicant. The proposed novel approach will be fundamentally investigated and optimized with regard to its computational efficiency, so that it may find its way into clinical praxis in the near future.

• DFG Programme: Research Grants