Feasibility of Data-Driven, Model-Free Quantitative MRI Protocol Design: Application to Brain and Prostate Diffusion-Relaxation Imaging
Purpose: We investigate the feasibility of data-driven, model-free quantitative MRI (qMRI) protocol design on in vivo brain and prostate diffusion-relaxation imaging (DRI).Methods: We select subsets of measurements within lengthy pilot scans, without identifying tissue parameters for which to optimi...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-11-01
|
| Series: | Frontiers in Physics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphy.2021.752208/full |
| _version_ | 1818984143786082304 |
|---|---|
| author | Francesco Grussu Francesco Grussu Francesco Grussu Stefano B. Blumberg Marco Battiston Lebina S. Kakkar Hongxiang Lin Andrada Ianuş Torben Schneider Torben Schneider Saurabh Singh Roger Bourne Shonit Punwani David Atkinson Claudia A. M. Gandini Wheeler-Kingshott Claudia A. M. Gandini Wheeler-Kingshott Claudia A. M. Gandini Wheeler-Kingshott Eleftheria Panagiotaki Thomy Mertzanidou Daniel C. Alexander |
| author_facet | Francesco Grussu Francesco Grussu Francesco Grussu Stefano B. Blumberg Marco Battiston Lebina S. Kakkar Hongxiang Lin Andrada Ianuş Torben Schneider Torben Schneider Saurabh Singh Roger Bourne Shonit Punwani David Atkinson Claudia A. M. Gandini Wheeler-Kingshott Claudia A. M. Gandini Wheeler-Kingshott Claudia A. M. Gandini Wheeler-Kingshott Eleftheria Panagiotaki Thomy Mertzanidou Daniel C. Alexander |
| author_sort | Francesco Grussu |
| collection | DOAJ |
| description | Purpose: We investigate the feasibility of data-driven, model-free quantitative MRI (qMRI) protocol design on in vivo brain and prostate diffusion-relaxation imaging (DRI).Methods: We select subsets of measurements within lengthy pilot scans, without identifying tissue parameters for which to optimise for. We use the “select and retrieve via direct upsampling” (SARDU-Net) algorithm, made of a selector, identifying measurement subsets, and a predictor, estimating fully-sampled signals from the subsets. We implement both using artificial neural networks, which are trained jointly end-to-end. We deploy the algorithm on brain (32 diffusion-/T1-weightings) and prostate (16 diffusion-/T2-weightings) DRI scans acquired on three healthy volunteers on two separate 3T Philips systems each. We used SARDU-Net to identify sub-protocols of fixed size, assessing reproducibility and testing sub-protocols for their potential to inform multi-contrast analyses via the T1-weighted spherical mean diffusion tensor (T1-SMDT, brain) and hybrid multi-dimensional MRI (HM-MRI, prostate) models, for which sub-protocol selection was not optimised explicitly.Results: In both brain and prostate, SARDU-Net identifies sub-protocols that maximise information content in a reproducible manner across training instantiations using a small number of pilot scans. The sub-protocols support T1-SMDT and HM-MRI multi-contrast modelling for which they were not optimised explicitly, providing signal quality-of-fit in the top 5% against extensive sub-protocol comparisons.Conclusions: Identifying economical but informative qMRI protocols from subsets of rich pilot scans is feasible and potentially useful in acquisition-time-sensitive applications in which there is not a qMRI model of choice. SARDU-Net is demonstrated to be a robust algorithm for data-driven, model-free protocol design. |
| first_indexed | 2024-12-20T18:14:19Z |
| format | Article |
| id | doaj.art-dde3530e76cb45218a05c5fe9fae8da8 |
| institution | Directory Open Access Journal |
| issn | 2296-424X |
| language | English |
| last_indexed | 2024-12-20T18:14:19Z |
| publishDate | 2021-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physics |
| spelling | doaj.art-dde3530e76cb45218a05c5fe9fae8da82022-12-21T19:30:24ZengFrontiers Media S.A.Frontiers in Physics2296-424X2021-11-01910.3389/fphy.2021.752208752208Feasibility of Data-Driven, Model-Free Quantitative MRI Protocol Design: Application to Brain and Prostate Diffusion-Relaxation ImagingFrancesco Grussu0Francesco Grussu1Francesco Grussu2Stefano B. Blumberg3Marco Battiston4Lebina S. Kakkar5Hongxiang Lin6Andrada Ianuş7Torben Schneider8Torben Schneider9Saurabh Singh10Roger Bourne11Shonit Punwani12David Atkinson13Claudia A. M. Gandini Wheeler-Kingshott14Claudia A. M. Gandini Wheeler-Kingshott15Claudia A. M. Gandini Wheeler-Kingshott16Eleftheria Panagiotaki17Thomy Mertzanidou18Daniel C. Alexander19Queen Square MS Centre, Queen Square Institute of Neurology, Faculty of Brain Sciences, University College London, London, United KingdomCentre for Medical Image Computing, Department of Computer Science, University College London, London, United KingdomRadiomics Group, Vall d’Hebron Institute of Oncology, Vall d’Hebron Barcelona Hospital Campus, Barcelona, SpainCentre for Medical Image Computing, Department of Computer Science, University College London, London, United KingdomQueen Square MS Centre, Queen Square Institute of Neurology, Faculty of Brain Sciences, University College London, London, United KingdomCentre for Medical Imaging, University College London, London, United KingdomCentre for Medical Image Computing, Department of Computer Science, University College London, London, United KingdomChampalimaud Research, Champalimaud Centre for the Unknown, Lisbon, PortugalPhilips United Kingdom, Guildford, United KingdomDeep Spin, Berlin, GermanyCentre for Medical Imaging, University College London, London, United KingdomDiscipline of Medical Imaging Science, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW, AustraliaCentre for Medical Imaging, University College London, London, United KingdomCentre for Medical Imaging, University College London, London, United KingdomQueen Square MS Centre, Queen Square Institute of Neurology, Faculty of Brain Sciences, University College London, London, United KingdomDepartment of Brain and Behavioural Sciences, University of Pavia, Pavia, Italy0Brain MRI 3T Center, IRCCS Mondino Foundation, Pavia, ItalyCentre for Medical Image Computing, Department of Computer Science, University College London, London, United KingdomCentre for Medical Image Computing, Department of Computer Science, University College London, London, United KingdomCentre for Medical Image Computing, Department of Computer Science, University College London, London, United KingdomPurpose: We investigate the feasibility of data-driven, model-free quantitative MRI (qMRI) protocol design on in vivo brain and prostate diffusion-relaxation imaging (DRI).Methods: We select subsets of measurements within lengthy pilot scans, without identifying tissue parameters for which to optimise for. We use the “select and retrieve via direct upsampling” (SARDU-Net) algorithm, made of a selector, identifying measurement subsets, and a predictor, estimating fully-sampled signals from the subsets. We implement both using artificial neural networks, which are trained jointly end-to-end. We deploy the algorithm on brain (32 diffusion-/T1-weightings) and prostate (16 diffusion-/T2-weightings) DRI scans acquired on three healthy volunteers on two separate 3T Philips systems each. We used SARDU-Net to identify sub-protocols of fixed size, assessing reproducibility and testing sub-protocols for their potential to inform multi-contrast analyses via the T1-weighted spherical mean diffusion tensor (T1-SMDT, brain) and hybrid multi-dimensional MRI (HM-MRI, prostate) models, for which sub-protocol selection was not optimised explicitly.Results: In both brain and prostate, SARDU-Net identifies sub-protocols that maximise information content in a reproducible manner across training instantiations using a small number of pilot scans. The sub-protocols support T1-SMDT and HM-MRI multi-contrast modelling for which they were not optimised explicitly, providing signal quality-of-fit in the top 5% against extensive sub-protocol comparisons.Conclusions: Identifying economical but informative qMRI protocols from subsets of rich pilot scans is feasible and potentially useful in acquisition-time-sensitive applications in which there is not a qMRI model of choice. SARDU-Net is demonstrated to be a robust algorithm for data-driven, model-free protocol design.https://www.frontiersin.org/articles/10.3389/fphy.2021.752208/fullquantitative MRI (qMRI)protocol designartificial neural network (ANN)diffusion-relaxationbrainprostate |
| spellingShingle | Francesco Grussu Francesco Grussu Francesco Grussu Stefano B. Blumberg Marco Battiston Lebina S. Kakkar Hongxiang Lin Andrada Ianuş Torben Schneider Torben Schneider Saurabh Singh Roger Bourne Shonit Punwani David Atkinson Claudia A. M. Gandini Wheeler-Kingshott Claudia A. M. Gandini Wheeler-Kingshott Claudia A. M. Gandini Wheeler-Kingshott Eleftheria Panagiotaki Thomy Mertzanidou Daniel C. Alexander Feasibility of Data-Driven, Model-Free Quantitative MRI Protocol Design: Application to Brain and Prostate Diffusion-Relaxation Imaging Frontiers in Physics quantitative MRI (qMRI) protocol design artificial neural network (ANN) diffusion-relaxation brain prostate |
| title | Feasibility of Data-Driven, Model-Free Quantitative MRI Protocol Design: Application to Brain and Prostate Diffusion-Relaxation Imaging |
| title_full | Feasibility of Data-Driven, Model-Free Quantitative MRI Protocol Design: Application to Brain and Prostate Diffusion-Relaxation Imaging |
| title_fullStr | Feasibility of Data-Driven, Model-Free Quantitative MRI Protocol Design: Application to Brain and Prostate Diffusion-Relaxation Imaging |
| title_full_unstemmed | Feasibility of Data-Driven, Model-Free Quantitative MRI Protocol Design: Application to Brain and Prostate Diffusion-Relaxation Imaging |
| title_short | Feasibility of Data-Driven, Model-Free Quantitative MRI Protocol Design: Application to Brain and Prostate Diffusion-Relaxation Imaging |
| title_sort | feasibility of data driven model free quantitative mri protocol design application to brain and prostate diffusion relaxation imaging |
| topic | quantitative MRI (qMRI) protocol design artificial neural network (ANN) diffusion-relaxation brain prostate |
| url | https://www.frontiersin.org/articles/10.3389/fphy.2021.752208/full |
| work_keys_str_mv | AT francescogrussu feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT francescogrussu feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT francescogrussu feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT stefanobblumberg feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT marcobattiston feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT lebinaskakkar feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT hongxianglin feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT andradaianus feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT torbenschneider feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT torbenschneider feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT saurabhsingh feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT rogerbourne feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT shonitpunwani feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT davidatkinson feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT claudiaamgandiniwheelerkingshott feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT claudiaamgandiniwheelerkingshott feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT claudiaamgandiniwheelerkingshott feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT eleftheriapanagiotaki feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT thomymertzanidou feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging AT danielcalexander feasibilityofdatadrivenmodelfreequantitativemriprotocoldesignapplicationtobrainandprostatediffusionrelaxationimaging |