Deep learning-based synthetic-CT generation in radiotherapy and PET: a review

Spadea, Maria Francesca; Maspero, Matteo; Zaffino, Paolo; Seco, Joao

doi:10.48550/arxiv.2102.02734

Cited by 3 publications

(7 citation statements)

References 141 publications

(231 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table III shows near minimum, near maximum, and average dose difference from CT based plan received by the planning target volume (PTV). The difference in average dose to the PTV relative to the prescribed dose was found to be 0.166 ± 0.18% which lies within a clinically acceptable range of dose difference of < 0.5% and comparable with multiple studies in literature [11], [12], [34], [35] IV. DISCUSSION…”

Section: F Dosimetric Evaluationsupporting

confidence: 83%

“…This can be attributed to the bone classification task which accurately differentiates air regions from bone as both appear dark in the input MR image. The accuracy of prediction by our proposed method is on par with or better than other U-Net and GAN methods compared in [11], [12], [15], [16], [33].…”

Section: E Quantitative Analysismentioning

confidence: 79%

“…Many recent works have proposed deep CNN learning methods for continuous value sCT generation using variants of U-Net or GAN based regression [13], [14]. Two of the recent surveys on the topic [11], [12] have extensively summarized related works. The methods summarized in these surveys work on one or multiple clinical MR sequences and generate sCT with comparable appearance and quantitative accuracy.…”

Section: A Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Region of Interest focused MRI to Synthetic CT Translation using Regression and Classification Multi-task Network

Kaushik¹,

Bylund²,

Cozzini³

et al. 2022

Preprint

View full text Add to dashboard Cite

In this work, we present a method for synthetic CT (sCT) generation from zero-echo-time (ZTE) MRI aimed at structural and quantitative accuracies of the image, with a particular focus on the accurate bone density value prediction. We propose a loss function that favors a spatially sparse region in the image. We harness the ability of a multi-task network to produce correlated outputs as a framework to enable localization of region of interest (RoI) via classification, emphasize regression of values within RoI and still retain the overall accuracy via global regression. The network is optimized by a composite loss function that combines a dedicated loss from each task. We demonstrate how the multi-task network with RoI focused loss offers an advantage over other configurations of the network to achieve higher accuracy of performance. This is relevant to sCT where failure to accurately estimate high Hounsfield Unit values of bone could lead to impaired accuracy in clinical applications. We compare the dose calculation maps from the proposed sCT and the real CT in a radiation therapy treatment planning setup.

show abstract

Section: F Dosimetric Evaluationsupporting

confidence: 83%

Section: E Quantitative Analysismentioning

confidence: 79%

Section: A Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Region of Interest focused MRI to Synthetic CT Translation using Regression and Classification Multi-task Network

Kaushik¹,

Bylund²,

Cozzini³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Although experimental results indicate the superior performance of HRU-Net approach, the promising results obtained by deep learning techniques have shown comparable performance of atlas-based techniques with relatively small dosimetric errors in the field of CT synthesis from magnetic resonance imaging (MRI) images, the deep learning-based methods show drift robustness to the outliers and high vulnerability (Arabi et al 2018, Lei et al 2019. The CNN model, however, is a more desirable alternative for clinical usage considering the computation time, as the synthetic CT generation process takes less than 1 min compared to atlas-based methods that take up to 2 h (Sjölund et al 2015, Spadea et al 2021. Moreover, in our experiment, MLACF took approximately 26 h for reconstruction of one brain model, whereas HRU-Net took less than 1 min Since there are not sufficient 3D models to generate a 3D dataset, all the data we used is 2D slices, which is one of the limitations of our study.…”

Section: Discussionmentioning

confidence: 99%

Generation of attenuation correction factors from time-of-flight PET emission data using high-resolution residual U-net

Yin¹,

Obi²

2021

Biomed. Phys. Eng. Express

View full text Add to dashboard Cite

Attenuation correction of annihilation photons is essential in PET image reconstruction for providing accurate quantitative activity maps. In the absence of an aligned CT device to obtain attenuation information, we propose the high-resolution residual U-net (HRU-Net) to extract attenuation correction factors (ACF) directly from time-of-flight (TOF) PET emission data. HRU-Net is built upon the U-Net encoding-decoding architecture and it utilizes four blocks of modified residual connections in each stage. In each residual block, concatenation is performed to incorporate input and output feature vectors. In addition, flexible and efficient elements of convolutional neural network (CNN) such as dilated convolutions, pre-activation order of a batch normalization (BN) layer, a rectified linear unit (ReLU) layer and a convolution layer, and residual connections are utilized to extract high resolution features. To illustrate the effectiveness of the proposed method, HRU-Net estimated ACF, attenuation maps and activity maps are compared with maximum likelihood ACF (MLACF) algorithm, U-Net, and HC-Net. An ablation study is conducted using non-TOF and TOF sinograms as inputs of networks. The experimental results show that HRU-Net with TOF projections as inputs leads to normalized root mean square error (NRMSE) of 4.84%1.58%, outperforming MLACF, U-Net and HC-Net with NRMSE of 47.82%13.62%, 6.92%1.94%, and 7.99%2.49%, respectively.

show abstract

“…Many existing approaches to this task have been summarized as belonging to three general classes: atlas-based, voxel-based, and learning-based methods (Edmund andNyholm 2017, Johnstone et al 2018). Recent investigations have focused primarily on approaches belonging to the last category, namely deep learning (DL)-based approaches in which convolutional neural networks are used to approximate a mapping between MRI and CT images (Spadea et al 2021).…”

Section: Introductionmentioning

confidence: 99%

Abdominal synthetic CT reconstruction with intensity projection prior for MRI-only adaptive radiotherapy

et al. 2021

View full text Add to dashboard Cite

Objective. Owing to the superior soft tissue contrast of MRI, MRI-guided adaptive radiotherapy (ART) is well-suited to managing interfractional changes in anatomy. An MRI-only workflow is desirable, but producing synthetic CT (sCT) data through paired data-driven deep learning (DL) for abdominal dose calculations remains a challenge due to the highly variable presence of intestinal gas. We present the preliminary dosimetric evaluation of our novel approach to sCT reconstruction that is well suited to handling intestinal gas in abdominal MRI-only ART. Approach. We utilize a paired data DL approach enabled by the intensity projection prior, in which well-matching training pairs are created by propagating air from MRI to corresponding CT scans. Evaluations focus on two classes: patients with (1) little involvement of intestinal gas, and (2) notable differences in intestinal gas presence between corresponding scans. Comparisons between sCT-based plans and CT-based clinical plans for both classes are made at the first treatment fraction to highlight the dosimetric impact of the variable presence of intestinal gas. Main results. Class 1 patients (n = 13) demonstrate differences in prescribed dose coverage of the PTV of 1.3 ± 2.1% between clinical plans and sCT-based plans. Mean DVH differences in all structures for Class 1 patients are found to be statistically insignificant. In Class 2 (n = 20), target coverage is 13.3 ± 11.0% higher in the clinical plans and mean DVH differences are found to be statistically significant. Significance. Significant deviations in calculated doses arising from the variable presence of intestinal gas in corresponding CT and MRI scans result in uncertainty in high-dose regions that may limit the effectiveness of adaptive dose escalation efforts. We have proposed a paired data-driven DL approach to sCT reconstruction for accurate dose calculations in abdominal ART enabled by the creation of a clinically unavailable training data set with well-matching representations of intestinal gas.

show abstract

Deep learning-based synthetic-CT generation in radiotherapy and PET: a review

Cited by 3 publications

References 141 publications

Region of Interest focused MRI to Synthetic CT Translation using Regression and Classification Multi-task Network

Region of Interest focused MRI to Synthetic CT Translation using Regression and Classification Multi-task Network

Generation of attenuation correction factors from time-of-flight PET emission data using high-resolution residual U-net

Abdominal synthetic CT reconstruction with intensity projection prior for MRI-only adaptive radiotherapy

Contact Info

Product

Resources

About