Cited 6 times in
Ensemble learning and personalized training for the improvement of unsupervised deep learning-based synthetic CT reconstruction
DC Field | Value | Language |
---|---|---|
dc.contributor.author | 김진성 | - |
dc.date.accessioned | 2024-01-05T05:39:09Z | - |
dc.date.available | 2024-01-05T05:39:09Z | - |
dc.date.issued | 2023-03 | - |
dc.identifier.issn | 0094-2405 | - |
dc.identifier.uri | https://ir.ymlib.yonsei.ac.kr/handle/22282913/197697 | - |
dc.description.abstract | Background: The growing adoption of magnetic resonance imaging (MRI)-guided radiation therapy (RT) platforms and a focus on MRI-only RT workflows have brought the technical challenge of synthetic computed tomography (sCT) reconstruction to the forefront. Unpaired-data deep learning-based approaches to the problem offer the attractive characteristic of not requiring paired training data, but the gap between paired- and unpaired-data results can be limiting. Purpose: We present two distinct approaches aimed at improving unpaired-data sCT reconstruction results: a cascade ensemble that combines multiple models and a personalized training strategy originally designed for the paired-data setting. Methods: Comparisons are made between the following models: (1) the paired-data fully convolutional DenseNet (FCDN), (2) the FCDN with the Intentional Deep Overfit Learning (IDOL) personalized training strategy, (3) the unpaired-data CycleGAN, (4) the CycleGAN with the IDOL training strategy, and (5) the CycleGAN as an intermediate model in a cascade ensemble approach. Evaluation of the various models over 25 total patients is carried out using a five-fold cross-validation scheme, with the patient-specific IDOL models being trained for the five patients of fold 3, chosen at random. Results: In both the paired- and unpaired-data settings, adopting the IDOL training strategy led to improvements in the mean absolute error (MAE) between true CT images and sCT outputs within the body contour (mean improvement, paired- and unpaired-data approaches, respectively: 38%, 9%) and in regions of bone (52%, 5%), the peak signal-to-noise ratio (PSNR; 15%, 7%), and the structural similarity index (SSIM; 6%, <1%). The ensemble approach offered additional benefits over the IDOL approach in all three metrics (mean improvement over unpaired-data approach in fold 3; MAE: 20%; bone MAE: 16%; PSNR: 10%; SSIM: 2%), and differences in body MAE between the ensemble approach and the paired-data approach are statistically insignificant. Conclusions: We have demonstrated that both a cascade ensemble approach and a personalized training strategy designed initially for the paired-data setting offer significant improvements in image quality metrics for the unpaired-data sCT reconstruction task. Closing the gap between paired- and unpaired-data approaches is a step toward fully enabling these powerful and attractive unpaired-data frameworks. | - |
dc.description.statementOfResponsibility | restriction | - |
dc.language | English | - |
dc.publisher | Published for the American Assn. of Physicists in Medicine by the American Institute of Physics. | - |
dc.relation.isPartOf | MEDICAL PHYSICS | - |
dc.rights | CC BY-NC-ND 2.0 KR | - |
dc.subject.MESH | Deep Learning* | - |
dc.subject.MESH | Humans | - |
dc.subject.MESH | Image Processing, Computer-Assisted / methods | - |
dc.subject.MESH | Magnetic Resonance Imaging | - |
dc.subject.MESH | Radiotherapy, Image-Guided* | - |
dc.subject.MESH | Tomography, X-Ray Computed | - |
dc.title | Ensemble learning and personalized training for the improvement of unsupervised deep learning-based synthetic CT reconstruction | - |
dc.type | Article | - |
dc.contributor.college | College of Medicine (의과대학) | - |
dc.contributor.department | Dept. of Radiation Oncology (방사선종양학교실) | - |
dc.contributor.googleauthor | Sven Olberg | - |
dc.contributor.googleauthor | Byong Su Choi | - |
dc.contributor.googleauthor | Inkyung Park | - |
dc.contributor.googleauthor | Xiao Liang | - |
dc.contributor.googleauthor | Jin Sung Kim | - |
dc.contributor.googleauthor | Jie Deng | - |
dc.contributor.googleauthor | Yulong Yan | - |
dc.contributor.googleauthor | Steve Jiang | - |
dc.contributor.googleauthor | Justin C Park | - |
dc.identifier.doi | 10.1002/mp.16087 | - |
dc.contributor.localId | A04548 | - |
dc.relation.journalcode | J02206 | - |
dc.identifier.eissn | 2473-4209 | - |
dc.identifier.pmid | 36336718 | - |
dc.identifier.url | https://aapm.onlinelibrary.wiley.com/doi/10.1002/mp.16087 | - |
dc.subject.keyword | MR-only RT | - |
dc.subject.keyword | deep learning | - |
dc.subject.keyword | synthetic CT | - |
dc.contributor.alternativeName | Kim, Jinsung | - |
dc.contributor.affiliatedAuthor | 김진성 | - |
dc.citation.volume | 50 | - |
dc.citation.number | 3 | - |
dc.citation.startPage | 1436 | - |
dc.citation.endPage | 1449 | - |
dc.identifier.bibliographicCitation | MEDICAL PHYSICS, Vol.50(3) : 1436-1449, 2023-03 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.