Accelerating 4D image reconstruction for magnetic resonance-guided radiotherapy
Background and purpose: Physiological motion impacts the dose delivered to tumours and vital organs in external beam radiotherapy and particularly in particle therapy. The excellent soft-tissue demarcation of 4D magnetic resonance imaging (4D-MRI) could inform on intra-fractional motion, but long im...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-07-01
|
| Series: | Physics and Imaging in Radiation Oncology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405631623000751 |
| _version_ | 1827824132256956416 |
|---|---|
| author | Bastien Lecoeur Marco Barbone Jessica Gough Uwe Oelfke Wayne Luk Georgi Gaydadjiev Andreas Wetscherek |
| author_facet | Bastien Lecoeur Marco Barbone Jessica Gough Uwe Oelfke Wayne Luk Georgi Gaydadjiev Andreas Wetscherek |
| author_sort | Bastien Lecoeur |
| collection | DOAJ |
| description | Background and purpose: Physiological motion impacts the dose delivered to tumours and vital organs in external beam radiotherapy and particularly in particle therapy. The excellent soft-tissue demarcation of 4D magnetic resonance imaging (4D-MRI) could inform on intra-fractional motion, but long image reconstruction times hinder its use in online treatment adaptation. Here we employ techniques from high-performance computing to reduce 4D-MRI reconstruction times below two minutes to facilitate their use in MR-guided radiotherapy. Material and methods: Four patients with pancreatic adenocarcinoma were scanned with a radial stack-of-stars gradient echo sequence on a 1.5T MR-Linac. Fast parallelised open-source implementations of the extra-dimensional golden-angle radial sparse parallel algorithm were developed for central processing unit (CPU) and graphics processing unit (GPU) architectures. We assessed the impact of architecture, oversampling and respiratory binning strategy on 4D-MRI reconstruction time and compared images using the structural similarity (SSIM) index against a MATLAB reference implementation. Scaling and bottlenecks for the different architectures were studied using multi-GPU systems. Results: All reconstructed 4D-MRI were identical to the reference implementation (SSIM > 0.99). Images reconstructed with overlapping respiratory bins were sharper at the cost of longer reconstruction times. The CPU + GPU implementation was over 17 times faster than the reference implementation, reconstructing images in 60 ± 1 s and hyper-scaled using multiple GPUs. Conclusion: Respiratory-resolved 4D-MRI reconstruction times can be reduced using high-performance computing methods for online workflows in MR-guided radiotherapy with potential applications in particle therapy. |
| first_indexed | 2024-03-12T02:21:37Z |
| format | Article |
| id | doaj.art-dbb7e8c0102743dd8cb22445868b5d13 |
| institution | Directory Open Access Journal |
| issn | 2405-6316 |
| language | English |
| last_indexed | 2024-03-12T02:21:37Z |
| publishDate | 2023-07-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Physics and Imaging in Radiation Oncology |
| spelling | doaj.art-dbb7e8c0102743dd8cb22445868b5d132023-09-06T04:52:31ZengElsevierPhysics and Imaging in Radiation Oncology2405-63162023-07-0127100484Accelerating 4D image reconstruction for magnetic resonance-guided radiotherapyBastien Lecoeur0Marco Barbone1Jessica Gough2Uwe Oelfke3Wayne Luk4Georgi Gaydadjiev5Andreas Wetscherek6Joint Department of Physics at The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, 15 Cotswold Rd, London SM2 5NG, United Kingdom; Department of Computing, Imperial College London, Exhibition Rd, South Kensington, London SW7 2BX, United Kingdom; Corresponding author at: Joint Department of Physics at The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, 15 Cotswold Rd, London SM2 5NG, United Kingdom.Joint Department of Physics at The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, 15 Cotswold Rd, London SM2 5NG, United Kingdom; Department of Computing, Imperial College London, Exhibition Rd, South Kensington, London SW7 2BX, United KingdomDepartment of Radiotherapy at the Royal Marsden NHS Foundation Trust, Downs Rd, London SM2 5PT, United KingdomJoint Department of Physics at The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, 15 Cotswold Rd, London SM2 5NG, United KingdomDepartment of Computing, Imperial College London, Exhibition Rd, South Kensington, London SW7 2BX, United KingdomDepartment of Computing, Imperial College London, Exhibition Rd, South Kensington, London SW7 2BX, United Kingdom; Bernoulli Institute, University of Groningen, Nijenborgh 9, Groningen 9747 AG, The NetherlandsJoint Department of Physics at The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, 15 Cotswold Rd, London SM2 5NG, United KingdomBackground and purpose: Physiological motion impacts the dose delivered to tumours and vital organs in external beam radiotherapy and particularly in particle therapy. The excellent soft-tissue demarcation of 4D magnetic resonance imaging (4D-MRI) could inform on intra-fractional motion, but long image reconstruction times hinder its use in online treatment adaptation. Here we employ techniques from high-performance computing to reduce 4D-MRI reconstruction times below two minutes to facilitate their use in MR-guided radiotherapy. Material and methods: Four patients with pancreatic adenocarcinoma were scanned with a radial stack-of-stars gradient echo sequence on a 1.5T MR-Linac. Fast parallelised open-source implementations of the extra-dimensional golden-angle radial sparse parallel algorithm were developed for central processing unit (CPU) and graphics processing unit (GPU) architectures. We assessed the impact of architecture, oversampling and respiratory binning strategy on 4D-MRI reconstruction time and compared images using the structural similarity (SSIM) index against a MATLAB reference implementation. Scaling and bottlenecks for the different architectures were studied using multi-GPU systems. Results: All reconstructed 4D-MRI were identical to the reference implementation (SSIM > 0.99). Images reconstructed with overlapping respiratory bins were sharper at the cost of longer reconstruction times. The CPU + GPU implementation was over 17 times faster than the reference implementation, reconstructing images in 60 ± 1 s and hyper-scaled using multiple GPUs. Conclusion: Respiratory-resolved 4D-MRI reconstruction times can be reduced using high-performance computing methods for online workflows in MR-guided radiotherapy with potential applications in particle therapy.http://www.sciencedirect.com/science/article/pii/S24056316230007514D-MRIMR-guided RadiotherapyMR-integrated Proton TherapyIntrafraction motionHigh-performance computing |
| spellingShingle | Bastien Lecoeur Marco Barbone Jessica Gough Uwe Oelfke Wayne Luk Georgi Gaydadjiev Andreas Wetscherek Accelerating 4D image reconstruction for magnetic resonance-guided radiotherapy Physics and Imaging in Radiation Oncology 4D-MRI MR-guided Radiotherapy MR-integrated Proton Therapy Intrafraction motion High-performance computing |
| title | Accelerating 4D image reconstruction for magnetic resonance-guided radiotherapy |
| title_full | Accelerating 4D image reconstruction for magnetic resonance-guided radiotherapy |
| title_fullStr | Accelerating 4D image reconstruction for magnetic resonance-guided radiotherapy |
| title_full_unstemmed | Accelerating 4D image reconstruction for magnetic resonance-guided radiotherapy |
| title_short | Accelerating 4D image reconstruction for magnetic resonance-guided radiotherapy |
| title_sort | accelerating 4d image reconstruction for magnetic resonance guided radiotherapy |
| topic | 4D-MRI MR-guided Radiotherapy MR-integrated Proton Therapy Intrafraction motion High-performance computing |
| url | http://www.sciencedirect.com/science/article/pii/S2405631623000751 |
| work_keys_str_mv | AT bastienlecoeur accelerating4dimagereconstructionformagneticresonanceguidedradiotherapy AT marcobarbone accelerating4dimagereconstructionformagneticresonanceguidedradiotherapy AT jessicagough accelerating4dimagereconstructionformagneticresonanceguidedradiotherapy AT uweoelfke accelerating4dimagereconstructionformagneticresonanceguidedradiotherapy AT wayneluk accelerating4dimagereconstructionformagneticresonanceguidedradiotherapy AT georgigaydadjiev accelerating4dimagereconstructionformagneticresonanceguidedradiotherapy AT andreaswetscherek accelerating4dimagereconstructionformagneticresonanceguidedradiotherapy |