Peter E Yoo1,2, Jon O Cleary3, Scott C Kolbe3,4, Roger J Ordidge3, Terence J O'Brien5, Nicholas L Opie6,7,8, Sam E John6,7,8,4, Thomas J Oxley7,8,4,5, Bradford A Moffat3. 1. Department of Medicine and Radiology, Melbourne Medical School, The University of Melbourne, Kenneth Myer Building, 30 Royal Parade, Parkville, VIC, 3052, Australia. peter.eli.yoo@gmail.com. 2. Vascular Bionics Laboratory, Department of Medicine, Melbourne Brain Centre, The University of Melbourne, Melbourne, VIC, Australia. peter.eli.yoo@gmail.com. 3. Department of Medicine and Radiology, Melbourne Medical School, The University of Melbourne, Kenneth Myer Building, 30 Royal Parade, Parkville, VIC, 3052, Australia. 4. The Florey Institute of Neuroscience and Mental Health, Parkville, VIC, Australia. 5. Departments of Medicine and Neurology, Melbourne Brain Centre at The Royal Melbourne Hospital, The University of Melbourne, Melbourne, VIC, Australia. 6. Department of Biomedical and Electronic Engineering, The University of Melbourne, Melbourne, VIC, Australia. 7. Vascular Bionics Laboratory, Department of Medicine, Melbourne Brain Centre, The University of Melbourne, Melbourne, VIC, Australia. 8. Center for Neural Engineering, The University of Melbourne, Melbourne, VIC, Australia.
Abstract
OBJECTIVE: Ultra-high-field functional MRI (UHF-fMRI) allows for higher spatiotemporal resolution imaging. However, higher-resolution imaging entails coverage limitations. Processing partial-coverage images using standard pipelines leads to sub-optimal results. We aimed to develop a simple, semi-automated pipeline for processing partial-coverage UHF-fMRI data using widely used image processing algorithms. MATERIALS AND METHODS: We developed automated pipelines for optimized skull stripping and co-registration of partial-coverage UHF functional images, using built-in functions of the Centre for Functional Magnetic Resonance Imaging of the Brain's (FMRIB's) Software library (FSL) and advanced normalization tools. We incorporated the pipelines into the FSL's functional analysis pipeline and provide a semi-automated optimized partial-coverage functional analysis pipeline (OPFAP). RESULTS: Compared to the standard pipeline, the OPFAP yielded images with 15 and 30% greater volume of non-zero voxels after skull stripping the functional and anatomical images, respectively (all p = 0.0004), which reflected the conservation of cortical voxels lost when the standard pipeline was used. The OPFAP yielded the greatest Dice and Jaccard coefficients (87 and 80%, respectively; all p < 0.0001) between the co-registered participant gyri maps and the template gyri maps, demonstrating the goodness of the co-registration results. Furthermore, the greatest volume of group-level activation in the most number of functionally relevant regions was observed when the OPFAP was used. Importantly, group-level activations were not observed when using the standard pipeline. CONCLUSION: These results suggest that the OPFAP should be used for processing partial-coverage UHF-fMRI data for detecting high-resolution macroscopic blood oxygenation level-dependent activations.
OBJECTIVE: Ultra-high-field functional MRI (UHF-fMRI) allows for higher spatiotemporal resolution imaging. However, higher-resolution imaging entails coverage limitations. Processing partial-coverage images using standard pipelines leads to sub-optimal results. We aimed to develop a simple, semi-automated pipeline for processing partial-coverage UHF-fMRI data using widely used image processing algorithms. MATERIALS AND METHODS: We developed automated pipelines for optimized skull stripping and co-registration of partial-coverage UHF functional images, using built-in functions of the Centre for Functional Magnetic Resonance Imaging of the Brain's (FMRIB's) Software library (FSL) and advanced normalization tools. We incorporated the pipelines into the FSL's functional analysis pipeline and provide a semi-automated optimized partial-coverage functional analysis pipeline (OPFAP). RESULTS: Compared to the standard pipeline, the OPFAP yielded images with 15 and 30% greater volume of non-zero voxels after skull stripping the functional and anatomical images, respectively (all p = 0.0004), which reflected the conservation of cortical voxels lost when the standard pipeline was used. The OPFAP yielded the greatest Dice and Jaccard coefficients (87 and 80%, respectively; all p < 0.0001) between the co-registered participant gyri maps and the template gyri maps, demonstrating the goodness of the co-registration results. Furthermore, the greatest volume of group-level activation in the most number of functionally relevant regions was observed when the OPFAP was used. Importantly, group-level activations were not observed when using the standard pipeline. CONCLUSION: These results suggest that the OPFAP should be used for processing partial-coverage UHF-fMRI data for detecting high-resolution macroscopic blood oxygenation level-dependent activations.
Authors: E Yacoub; A Shmuel; J Pfeuffer; P F Van De Moortele; G Adriany; P Andersen; J T Vaughan; H Merkle; K Ugurbil; X Hu Journal: Magn Reson Med Date: 2001-04 Impact factor: 4.668
Authors: Stephen M Smith; Mark Jenkinson; Mark W Woolrich; Christian F Beckmann; Timothy E J Behrens; Heidi Johansen-Berg; Peter R Bannister; Marilena De Luca; Ivana Drobnjak; David E Flitney; Rami K Niazy; James Saunders; John Vickers; Yongyue Zhang; Nicola De Stefano; J Michael Brady; Paul M Matthews Journal: Neuroimage Date: 2004 Impact factor: 6.556
Authors: Peter E Yoo; Sam E John; Shawna Farquharson; Jon O Cleary; Yan T Wong; Amanda Ng; Claire B Mulcahy; David B Grayden; Roger J Ordidge; Nicholas L Opie; Terence J O'Brien; Thomas J Oxley; Bradford A Moffat Journal: Neuroimage Date: 2017-03-08 Impact factor: 6.556
Authors: Peter E Yoo; Maureen A Hagan; Sam E John; Nicholas L Opie; Roger J Ordidge; Terence J O'Brien; Thomas J Oxley; Bradford A Moffat; Yan T Wong Journal: Hum Brain Mapp Date: 2018-03-08 Impact factor: 5.038
Authors: Mark Jenkinson; Christian F Beckmann; Timothy E J Behrens; Mark W Woolrich; Stephen M Smith Journal: Neuroimage Date: 2011-09-16 Impact factor: 6.556
Authors: Petroc Sumner; Parashkev Nachev; Peter Morris; Andrew M Peters; Stephen R Jackson; Christopher Kennard; Masud Husain Journal: Neuron Date: 2007-06-07 Impact factor: 17.173
Authors: Peter E Yoo; Thomas J Oxley; Sam E John; Nicholas L Opie; Roger J Ordidge; Terence J O'Brien; Maureen A Hagan; Yan T Wong; Bradford A Moffat Journal: Sci Rep Date: 2018-10-22 Impact factor: 4.379