Review badges
0 pre-pub reviews
1 post-pub reviews
Abstract

Purpose: Interests have been rapidly growing in the field of radiotherapy to replace CT with magnetic resonance imaging (MRI), due to superior soft tissue contrast offered by MRI and the desire to reduce unnecessary radiation dose. MR-only radiotherapy also simplifies clinical workflow and avoids uncertainties in aligning MR with CT. Methods, however, are needed to derive CT-equivalent representations, often known as synthetic CT (sCT), from patient MR images for dose calculation and DRR-based patient positioning. Synthetic CT estimation is also important for PET attenuation correction in hybrid PET-MR systems. We propose in this work a novel deep convolutional neural network (DCNN) method for sCT generation and evaluate its performance on a set of brain tumor patient images.Methods: The proposed method builds upon recent developments of deep learning and convolutional neural networks in the computer vision literature. The proposed DCNN model has 27 convolutional layers interleaved with pooling and unpooling layers and 35 million free parameters, which can be trained to learn a direct end-to-end mapping from MR images to their corresponding CTs. Training such a large model on our limited data is made possible through the principle of transfer learning and by initializing model weights from a pretrained model. Eighteen brain tumor patients with both CT and T1-weighted MR images are used as experimental data and a sixfold cross-validation study is performed. Each sCT generated is compared against the real CT image of the same patient on a voxel-by-voxel basis. Comparison is also made with respect to an atlas-based approach that involves deformable atlas registration and patch-based atlas fusion.Results: The proposed DCNN method produced a mean absolute error (MAE) below 85 HU for 13 of the 18 test subjects. The overall average MAE was 84.8 +/- 17.3 HU for all subjects, which was found to be significantly better than the average MAE of 94.5 +/- 17.8 HU for the atlas-based method. The DCNN method also provided significantly better accuracy when being evaluated using two other metrics: the mean squared error (188.6 +/- 33.7 versus 198.3 +/- 33.0) and the Pearson correlation coefficient(0.906 +/- 0.03 versus 0.896 +/- 0.03). Although training a DCNN model can be slow, training only need be done once. Applying a trained model to generate a complete sCT volume for each new patient MR image only took 9 s, which was much faster than the atlas-based approach.Conclusions: A DCNN model method was developed, and shown to be able to produce highly accurate sCT estimations from conventional, single-sequence MR images in near real time. Quantitative results also showed that the proposed method competed favorably with an atlas-based method, in terms of both accuracy and computation speed at test time. Further validation on dose computation accuracy and on a larger patient cohort is warranted. Extensions of the method are also possible to further improve accuracy or to handle multi-sequence MR images. (C) 2017 American Association of Physicists in Medicine

Authors

Han, Xiao

Publons users who've claimed - I am an author

No Publons users have claimed this paper.

Contributors on Publons
  • 1 reviewer
Followers on Publons
Metrics
Publons score (from 1 score)
7.5
Altmetric
Web of Science Core Collection Citations
  • Abstract, title and references ● Is the aim clear? Yes ● Is it clear what the study found and how they did it? Yes ● Is the title informative and relevant? Yes ● Are the references: ● Relevant? Yes ● Recent? Yes ● Referenced correctly? Yes ● Are appropriate key studies included? Yes Introduction/ background ● Is it clear what is already known about this topic? Yes ● Is the research question clearly outlined? Yes ● Is the research question justified given what is already known about the topic? Yes Methods ● Is the process of subject selection clear? Yes ● Are the variables defined and measured appropriately? Yes ● Are the study methods valid and reliable? To some Extent ● Is there enough detail in order to replicate the study? Yes Results ● Is the data presented in an appropriate way? Yes ● Tables and figures relevant and clearly presented? Yes ● Appropriate units, rounding, and number of decimals? Yes ● Titles, columns, and rows labelled correctly and clearly? Yes ● Categories grouped appropriately? Yes ● Does the text in the results add to the data or is it repetitive? repetitive ● Are you clear about what is a statistically significant result? Yes ● Are you clear about what is a practically meaningful result? Yes Discussion and Conclusions ● Are the results discussed from multiple angles and placed into context without being over interpreted? Yes ● Do the conclusions answer the aims of the study? To some extent ● Are the conclusions supported by references or results? Yes ● Are the limitations of the study fatal or are they opportunities to inform future research? Needs Future Research.

    Published in
    Ongoing discussion
All peer review content displayed here is covered by a Creative Commons CC BY 4.0 license.