Title: Near real-time parallel-transmit pulse design for ultra-high field Magnetic Resonance Imaging
Emre Kopanoglu, Ph.D.
School of Psychology
Dec 22, 13:40
Aysel Sabuncu Brain Research Center
Ultra-high field (UHF) MRI offers improved signal- and contrast-to-noise ratio (SNR/CNR), which can be leveraged towards higher-resolution images. However, acquiring higher-resolution images increases scan times, making MRI scans prone to patient motion. Motion may become a problem especially with uncooperative patients such as in paediatric imaging and patients with Parkinson’s or dementia. A common solution in such cases is to use sedatives. Unfortunately, sedation makes MRI an invasive imaging modality, and may commonly have side effects. Alternatively, real-time motion correction techniques can adapt the scan parameters accordingly and ‘freeze’ patient motion from a data processing perspective these techniques are not directly applicable at UHF MRI.
At higher field strengths, artificial contrast variations are imposed on the acquired images due to the shorter wavelength. To correct such effects, parallel-transmit (pTx) arrays are commonly used. However, using pTx arrays may lead to increased localized heating inside the patient. Therefore, complex pTx pulses are designed to mitigate contrast variations while ensuring adherence to stringent safety limits. Due to the increased complexity of designing such pulses, design process may take between a few seconds and several minutes, making real-time motion correction infeasible.
This talk will introduce a method that can design pTx pulses in less than a second while adhering to safety limits, and demonstrate initial computational results.
Emre Kopanoglu obtained his BSc and PhD degrees from the Department of Electrical and Electronics Engineering of Bilkent University in 2006 and 2012, respectively. During his PhD studies, he focused on improving patient safety in Magnetic Resonance Imaging (MRI) through the use of novel nonlinear gradient hardware. He joined Yale University in 2012 as a Post-Doctoral Associate, where he further investigated using nonlinear gradient fields for patient motion tracking, improved excitation fidelity, and accelerated imaging in MRI, leading to a publication featured on the cover of the journal of Magnetic Resonance in Medicine. He joined Aselsan Research Centre in 2015 as a Senior Research Scientist, focusing on joint reconstruction of undersampled multi-contrast MRI data through compressive sensing in order to reduce MRI scan times without sacrificing image quality. In 2017, he joined Cardiff University as a Lecturer in the School of Psychology and the Cardiff University Brain Research Imaging Centre (CUBRIC), where he investigates RF pulse design at Ultra-high field (UHF) MRI. His work involves developing fast algorithms to correct the artificial contrast variations inherent to UHF-MRI in near real-time, while keeping patient heating under control and below safety limits.