On harmonisation of brain MRI data across scanners and sites
Magnetic resonance imaging (MRI) of the brain has revolutionised neuroscience by opening unique opportunities for studying unknown aspects of brain organisation, function and pathology-induced dysfunction. Despite the huge potential, MRI measures can be limited in their consistency, reproducibility...
| Main Author: | |
|---|---|
| Format: | Thesis (University of Nottingham only) |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://eprints.nottingham.ac.uk/74417/ |
| _version_ | 1848800867213377536 |
|---|---|
| author | Ntata, Asante |
| author_facet | Ntata, Asante |
| author_sort | Ntata, Asante |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | Magnetic resonance imaging (MRI) of the brain has revolutionised neuroscience by opening unique opportunities for studying unknown aspects of brain organisation, function and pathology-induced dysfunction. Despite the huge potential, MRI measures can be limited in their consistency, reproducibility and accuracy which subsequently restricts their quantifiability. Nuisance non-biological factors, such as hardware, software, calibration differences between scanners and post-processing options can contribute or drive trends in neuroimaging features to an extent that interferes with biological variability and obstructs scientific explorations and clinical applications. Such lack of consistency, or harmonisation across neuroimaging datasets poses a great challenge for our capabilities in quantitative MRI. This thesis contributes to better understanding and addressing it. We specifically build a new resource for comprehensively mapping the extent of the problem and objectively evaluating neuroimaging harmonisation approaches. We use a travelling heads paradigm consisting of multimodal MRI data of 10 travelling subjects, each scanned at 5 different sites on 6 different 3.0T scanners from all the 3 major vendors and using 5 imaging modalities. We use this dataset to explore the between-scanner variability of hundreds of imaging-extracted features and compare these to within-scanner (within-subject) variability and biological (between-subject) variability. We identify subsets of features that are/are not reliable across scanners and use our resource as a testbed to enable new investigations which until now have been relatively unexplored. Specifically, we identify optimal pipeline processing steps that minimise between-scanner variability in extracted features (implicit harmonisation). We also test the performance of post-processing harmonisation tools (explicit harmonisation) and specifically check their efficiency in reducing between-scanner variability against baseline gold standards provided by our data. Our explorations allow us to come up with good practice suggestions on processing steps and sets of features where results are more consistent and reproducible and also set references for future studies in this field. |
| first_indexed | 2025-11-14T20:58:23Z |
| format | Thesis (University of Nottingham only) |
| id | nottingham-74417 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T20:58:23Z |
| publishDate | 2023 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-744172023-12-13T04:40:09Z https://eprints.nottingham.ac.uk/74417/ On harmonisation of brain MRI data across scanners and sites Ntata, Asante Magnetic resonance imaging (MRI) of the brain has revolutionised neuroscience by opening unique opportunities for studying unknown aspects of brain organisation, function and pathology-induced dysfunction. Despite the huge potential, MRI measures can be limited in their consistency, reproducibility and accuracy which subsequently restricts their quantifiability. Nuisance non-biological factors, such as hardware, software, calibration differences between scanners and post-processing options can contribute or drive trends in neuroimaging features to an extent that interferes with biological variability and obstructs scientific explorations and clinical applications. Such lack of consistency, or harmonisation across neuroimaging datasets poses a great challenge for our capabilities in quantitative MRI. This thesis contributes to better understanding and addressing it. We specifically build a new resource for comprehensively mapping the extent of the problem and objectively evaluating neuroimaging harmonisation approaches. We use a travelling heads paradigm consisting of multimodal MRI data of 10 travelling subjects, each scanned at 5 different sites on 6 different 3.0T scanners from all the 3 major vendors and using 5 imaging modalities. We use this dataset to explore the between-scanner variability of hundreds of imaging-extracted features and compare these to within-scanner (within-subject) variability and biological (between-subject) variability. We identify subsets of features that are/are not reliable across scanners and use our resource as a testbed to enable new investigations which until now have been relatively unexplored. Specifically, we identify optimal pipeline processing steps that minimise between-scanner variability in extracted features (implicit harmonisation). We also test the performance of post-processing harmonisation tools (explicit harmonisation) and specifically check their efficiency in reducing between-scanner variability against baseline gold standards provided by our data. Our explorations allow us to come up with good practice suggestions on processing steps and sets of features where results are more consistent and reproducible and also set references for future studies in this field. 2023-12-13 Thesis (University of Nottingham only) NonPeerReviewed application/pdf en cc_by https://eprints.nottingham.ac.uk/74417/1/Thesis_for_PhD_Degree.pdf Ntata, Asante (2023) On harmonisation of brain MRI data across scanners and sites. PhD thesis, University of Nottingham. Magnetic resonance imaging; Neuroimaging; Neuroimaging datasets; Neuroimaging harmonisation |
| spellingShingle | Magnetic resonance imaging; Neuroimaging; Neuroimaging datasets; Neuroimaging harmonisation Ntata, Asante On harmonisation of brain MRI data across scanners and sites |
| title | On harmonisation of brain MRI data across scanners and sites |
| title_full | On harmonisation of brain MRI data across scanners and sites |
| title_fullStr | On harmonisation of brain MRI data across scanners and sites |
| title_full_unstemmed | On harmonisation of brain MRI data across scanners and sites |
| title_short | On harmonisation of brain MRI data across scanners and sites |
| title_sort | on harmonisation of brain mri data across scanners and sites |
| topic | Magnetic resonance imaging; Neuroimaging; Neuroimaging datasets; Neuroimaging harmonisation |
| url | https://eprints.nottingham.ac.uk/74417/ |