1
|
Zanier O, Theiler S, Mutten RD, Ryu SJ, Regli L, Serra C, Staartjes VE. TomoRay: Generating Synthetic Computed Tomography of the Spine From Biplanar Radiographs. Neurospine 2024; 21:68-75. [PMID: 38317547 PMCID: PMC10992629 DOI: 10.14245/ns.2347158.579] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/01/2023] [Revised: 12/22/2023] [Accepted: 12/30/2023] [Indexed: 02/07/2024] Open
Abstract
OBJECTIVE Computed tomography (CT) imaging is a cornerstone in the assessment of patients with spinal trauma and in the planning of spinal interventions. However, CT studies are associated with logistical problems, acquisition costs, and radiation exposure. In this proof-of-concept study, the feasibility of generating synthetic spinal CT images using biplanar radiographs was explored. This could expand the potential applications of x-ray machines pre-, post-, and even intraoperatively. METHODS A cohort of 209 patients who underwent spinal CT imaging from the VerSe2020 dataset was used to train the algorithm. The model was subsequently evaluated using an internal and external validation set containing 55 from the VerSe2020 dataset and a subset of 56 images from the CTSpine1K dataset, respectively. Digitally reconstructed radiographs served as input for training and evaluation of the 2-dimensional (2D)-to-3-dimentional (3D) generative adversarial model. Model performance was assessed using peak signal to noise ratio (PSNR), structural similarity index (SSIM), and cosine similarity (CS). RESULTS At external validation, the developed model achieved a PSNR of 21.139 ± 1.018 dB (mean ± standard deviation). The SSIM and CS amounted to 0.947 ± 0.010 and 0.671 ± 0.691, respectively. CONCLUSION Generating an artificial 3D output from 2D imaging is challenging, especially for spinal imaging, where x-rays are known to deliver insufficient information frequently. Although the synthetic CT scans derived from our model do not perfectly match their ground truth CT, our proof-of-concept study warrants further exploration of the potential of this technology.
Collapse
Affiliation(s)
- Olivier Zanier
- Machine Intelligence in Clinical Neuroscience (MICN) Laboratory, Department of Neurosurgery, Clinical Neuroscience Center, University Hospital Zürich, University of Zürich, Zürich, Switzerland
| | - Sven Theiler
- Machine Intelligence in Clinical Neuroscience (MICN) Laboratory, Department of Neurosurgery, Clinical Neuroscience Center, University Hospital Zürich, University of Zürich, Zürich, Switzerland
| | - Raffaele Da Mutten
- Machine Intelligence in Clinical Neuroscience (MICN) Laboratory, Department of Neurosurgery, Clinical Neuroscience Center, University Hospital Zürich, University of Zürich, Zürich, Switzerland
| | - Seung-Jun Ryu
- Department of Neurosurgery, Daejeon Eulji University Hospital, Eulji University Medical School, Daejeon, Korea
| | - Luca Regli
- Machine Intelligence in Clinical Neuroscience (MICN) Laboratory, Department of Neurosurgery, Clinical Neuroscience Center, University Hospital Zürich, University of Zürich, Zürich, Switzerland
| | - Carlo Serra
- Machine Intelligence in Clinical Neuroscience (MICN) Laboratory, Department of Neurosurgery, Clinical Neuroscience Center, University Hospital Zürich, University of Zürich, Zürich, Switzerland
| | - Victor E. Staartjes
- Machine Intelligence in Clinical Neuroscience (MICN) Laboratory, Department of Neurosurgery, Clinical Neuroscience Center, University Hospital Zürich, University of Zürich, Zürich, Switzerland
| |
Collapse
|
2
|
Vereecke E, Morbée L, Laloo F, Chen M, Jaremko JL, Herregods N, Jans L. Anatomical variation of the sacroiliac joints: an MRI study with synthetic CT images. Insights Imaging 2023; 14:30. [PMID: 36750489 PMCID: PMC9905396 DOI: 10.1186/s13244-023-01373-1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/17/2022] [Accepted: 01/09/2023] [Indexed: 02/09/2023] Open
Abstract
BACKGROUND Synthetic computed tomography (sCT) images are magnetic resonance imaging (MRI)-based images, generated using artificial intelligence. This study aimed to determine the prevalence of anatomical variants of sacroiliac joints (SIJ) on sCT images and the correlation with age, sex and body weight. METHODS MRI of the SIJ including sCT images of 215 patients clinically suspected for sacroiliitis were retrospectively analyzed. The presence of anatomical variants of the SIJ was assessed. Age, sex and body mass index at the time of the MRI were recorded. RESULTS SIJ variants were found in 82.8% (356/430) of the evaluated joints. The most frequent variants were iliosacral complex (27.7%), bipartite iliac bony plate (27.2%) and crescent iliac bony plate (27%). One new variant was identified, consisting of an accessory facet of the SIJ on the superior side. Overall, SIJ variants were slightly more frequent in women (85.8% vs. 77.8%), but iliosacral complex was significantly more frequent in men. Isolated synostosis was more prevalent with advancing age, in contrast to semicircular defect and unfused ossification center. The occurrence of iliosacral complex was associated with higher BMI, while crescent iliac bony plate occurred more in patients with lower BMI. CONCLUSION Over 80% of patients in this study, who were all suspected of sacroiliitis, had at least one SIJ variant. These variants may actually represent subtypes of the normal SIJ. sCT enables detection of very small or subtle findings including SIJ variants.
Collapse
Affiliation(s)
- Elke Vereecke
- Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000, Ghent, Belgium.
| | - Lieve Morbée
- grid.410566.00000 0004 0626 3303Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000 Ghent, Belgium
| | - Frederiek Laloo
- grid.410566.00000 0004 0626 3303Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000 Ghent, Belgium
| | - Min Chen
- grid.440601.70000 0004 1798 0578Department of Radiology, Peking University Shenzhen Hospital, Shenzhen, 518036 China
| | - Jacob L. Jaremko
- grid.241114.30000 0004 0459 7625Department of Radiology and Diagnostic Imaging, University of Alberta Hospital, Edmonton, AB T6G 2B7 Canada
| | - Nele Herregods
- grid.410566.00000 0004 0626 3303Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000 Ghent, Belgium
| | - Lennart Jans
- grid.410566.00000 0004 0626 3303Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000 Ghent, Belgium
| |
Collapse
|
3
|
Morbée L, Vereecke E, Laloo F, Chen M, Herregods N, Jans LBO. Common incidental findings on sacroiliac joint MRI: Added value of MRI-based synthetic CT. Eur J Radiol 2023; 158:110651. [PMID: 36535080 DOI: 10.1016/j.ejrad.2022.110651] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/08/2022] [Revised: 12/01/2022] [Accepted: 12/11/2022] [Indexed: 12/15/2022]
Abstract
PURPOSE To determine the prevalence of incidental findings on sacroiliac joint MRI and to determine the added value of MRI-based synthetic CT in the detection and evaluation of these incidental findings. METHOD In this retrospective study 210 patients clinically suspected of spondyloarthritis who underwent MRI of the sacroiliac joint with synthetic CT sequence were included. The images were reviewed by two radiologists in consensus for the prevalence of sacroiliitis, incidental findings, and the ability of synthetic CT and the conventional MRI to detect and diagnose these findings. RESULTS In 44.7% of patients sacroiliitis was present. In 89.0% of patients MRI showed at least one incidental finding other than sacroiliitis. Degeneration of the sacroiliac joint was the most prevalent finding (140 patients, 66.6%). The most frequent incidental findings outside the sacroiliac joint were facet joint degeneration (29.0%), disc degeneration (25.2%), enostosis (19.5%) and lumbosacral transitional vertebrae (14.3%). A total of 788 lesions was recorded and synthetic CT was found to be problem solving or necessary for diagnosis in 543 (68.9%) of these lesions. 42.1% of lesions were not visible on conventional MRI (T1 TSE and STIR), most often degenerative osteophytes in the sacroiliac joint or lower lumbar spine. CONCLUSION Incidental findings are seen more frequently on sacroiliac joint MRI than sacroiliitis, which is relevant as some will have clinical significance or require treatment. Nearly half of these incidental lesions were only visible on synthetic CT, which additionally has been shown to be problem solving for diagnosis in many other cases.
Collapse
Affiliation(s)
- Lieve Morbée
- Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000 Ghent, Belgium.
| | - Elke Vereecke
- Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000 Ghent, Belgium
| | - Frederiek Laloo
- Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000 Ghent, Belgium
| | - Min Chen
- Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000 Ghent, Belgium
| | - Nele Herregods
- Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000 Ghent, Belgium
| | - Lennart B O Jans
- Department of Radiology, Ghent University Hospital, Corneel Heymanslaan 10, 9000 Ghent, Belgium
| |
Collapse
|
4
|
Qi M, Li Y, Wu A, Lu X, Zhou L, Song T. Multi-sequence MR generated sCT is promising for HNC MR-only RT: a comprehensive evaluation of previously developed sCT generation networks. Med Phys 2022; 49:2150-2158. [PMID: 35218040 DOI: 10.1002/mp.15572] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/30/2021] [Revised: 02/01/2022] [Accepted: 02/20/2022] [Indexed: 11/11/2022] Open
Abstract
PURPOSE To verify the feasibility of our in-house developed multi-sequence magnetic resonance (MR)-generated synthetic computed tomography (sCT) for the accurate dose calculation and fractional positioning for head and neck MR-only radiation therapy (RT). MATERIALS AND METHODS Forty-five patients with nasopharyngeal carcinoma were retrospectively studied. By applying our previously in-house developed network, a patient's sCT can rapidly be generated with respect to feeding the sole T1 image, T1C image, T1DixonC image, T2 image, and their combination respectively (five pipelines in total). A k(5)-fold strategy was implemented during model establishment. Dose recalculation was performed for each pipeline generation to attain a dosimetric feasibility evaluation. Fractional positioning evaluation was performed by calculating the digitally reconstructed radiograph (DRR) of the sCT and planning CT and their offset to the portal image. RESULTS The dose mean absolute error values are (0.47±0.16)%, (0.48±0.15)% (p<0.05), (0.50±0.16)% (p<0.05), (0.50±0.15)% (p<0.05), and (0.45±0.16)% (p<0.05) for the T1, T1C, T1Dixon C, T2, and 4-channel generated sCT to the prescription dose, respectively. The 4-channel-generated sCT outperforms any other single-sequence pipelines. Among the single-sequence MR imaging-generated sCTs, the T1-generated shows the most accurate HU image quality and provide a reliable dose result. Quantified positioning errors with calculation of the difference to the planning CT offsets are (-0.26±0.50)mm, (-0.58±0.52)mm (p<0.05), (-0.27±0.57)mm (p>0.05), (-0.31±0.44)mm (p>0.05), and (-0.19±0.37)mm (p>0.05) at LNG and (0.34±0.53)mm, (0.48±0.56)mm (p>0.05), (0.55±0.56)mm (p>0.05), (0.37±0.61)mm (p>0.05), and (0.24±0.43)mm (p>0.05) at LAT of the anterior-posterior direction for the five pipelines. CONCLUSION Multi-sequence MR-generated sCT allows for accurate dose calculation and fractional positioning for head and neck MR-only RT. This article is protected by copyright. All rights reserved.
Collapse
Affiliation(s)
- Mengke Qi
- School of Biomedical Engineering, Southern Medical University, Guangzhou, Guangdong, 510515, China
| | - Yongbao Li
- Department of Radiation Oncology, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, Guangdong, 510060, China
| | - Aiqian Wu
- School of Biomedical Engineering, Southern Medical University, Guangzhou, Guangdong, 510515, China.,Department of Radiation Oncology, The First Affiliated Hospital of Guangzhou University of Traditional Chinese Medicine, Guangzhou, Guangdong, 510405, China
| | - Xingyu Lu
- School of Biomedical Engineering, Southern Medical University, Guangzhou, Guangdong, 510515, China
| | - Linghong Zhou
- School of Biomedical Engineering, Southern Medical University, Guangzhou, Guangdong, 510515, China
| | - Ting Song
- School of Biomedical Engineering, Southern Medical University, Guangzhou, Guangdong, 510515, China
| |
Collapse
|