Super Resolution algorithms and software tools for MR imaging techniques
Project Title: Super Resolution algorithms and software tools for MR imaging techniques
Collaborating with The University of Edinburgh (UK)
Funded by the Royal Society Newton Mobility Grant (2016 - this collaboration is ongoing due to synergies in research interests)
Co-Investigators: Maria Valdes-Hernandez, Esin Öztürk Işık
Researchers: Sevim Cengiz, Asım Şamlı
Affiliated Departments: Row Fogo Centre for Research into Ageing and the Brain, Edinburgh Imaging, Centre for Clinical Brain Sciences in The University of Edinburgh
Project Summary: A vast amount of radiological data is currently considered suboptimal, and is archived without further use due to their poor spatial resolution and sometimes low quality. Subtle indications of early disease stages are also missed for the same reason. The spatial resolution choice is about finding a good balance between image quality, voxel size and acquisition time. This project aims to deliver computational methods and software tools to increase the resolution of different modalities of magnetic resonance images (MRI) to ensure better brain research, diagnosis and timely treatment strategies.
Keywords: Magnetic resonance imaging, brain imaging, super resolution, software tools, deep learning
Ozturk-Isik E, Marshall I, Filipiak P, Benjamin AJ, Ones VG, Ramón RO, Valdés Hernández MD. Workshop on reconstruction schemes for magnetic resonance data: summary of findings and recommendations. R Soc Open Sci. 2017;4(2):160731.
Sevim Cengiz, Maria del C. Valdes-Hernandez, Ozturk-Isik E. Super Resolution Convolutional Neural Networks for Increasing Spatial Resolution of 1H Magnetic Resonance Spectroscopic Imaging. Medical Image Understanding and Analysis 2017, Communications in Computer and Information Science, Vol. 723, pp.641-650, Springer, 2017. ISSN 1865-0929. Invited and Research Talks
Invited conference talks and scientific workshops:
Sevim Cengiz, Maria del C. Valdes-Hernandez, Ozturk-Isik E. Super Resolution Convolutional Neural Networks for Increasing Spatial Resolution of 1H Magnetic Resonance Spectroscopic Imaging. Medical Image Understanding and Analysis 2017 conference, 13th July 2017.
Cengiz, S. Advances in magnetic resonance spectroscopic image processing. ‘Recent Advances in Magnetic Resonance Imaging’ workshop, Bogazici University, Istanbul, June 20th, 2017.
Ozturk-Isik E. Compressed Sensing for Fast 31P-MRSI of Brain Tumors and 1H-MRSI of Mild Cognitive Impairment in Parkinson’s Disease. ‘Reconstruction schemes for MR data' workshop, University of Edinburgh, UK, August 17th, 2016