Evaluation of deformable image registration for contour propagation between CT and cone-beam CT images in adaptive head and neck radiotherapy

Li, X.; Zhang, Y.Y.; Shi, Yao Wu; Zhou, Linghong; Zhen, Xin

doi:10.3233/thc-161204

Cited by 15 publications

(13 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A lower HD from the in‐house hybrid method is found in most of the analyzed OARs. These DSC statistics are in agreement with results in previous studies 23,30,33,34 . Bastien showed that the mean DSC for the parotid gland of 15 patients was 0.75 30 .…”

Section: Discussionsupporting

confidence: 92%

Comparison of an in‐house hybrid DIR method to NiftyReg on CBCT and CT images for head and neck cancer

Jiang

Huang

Ding

et al. 2022

J Applied Clin Med Phys

View full text Add to dashboard Cite

An in‐house hybrid deformable image registration (DIR) method, which combines free‐form deformation (FFD) and the viscous fluid registration method, is proposed. Its results on the planning computed tomography (CT) and the day 1 treatment cone‐beam CT (CBCT) image from 68 head and neck cancer patients are compared with the results of NiftyReg, which uses B‐spline FFD alone. Several similarity metrics, the target registration error (TRE) of annotated points, as well as the Dice similarity coefficient (DSC) and Hausdorff distance (HD) of the propagated organs at risk are employed to analyze their registration accuracy. According to quantitative analysis on mutual information, normalized cross‐correlation, and the absolute pixel value differences, the results of the proposed DIR are more similar to the CBCT images than the NiftyReg results. Smaller TRE of the annotated points is observed in the proposed method, and the overall mean TRE for the proposed method and NiftyReg was 2.34 and 2.98 mm, respectively (p < 0.001). The mean DSC in the larynx, spinal cord, oral cavity, mandible, and parotid given by the proposed method ranged from 0.78 to 0.91, significantly higher than the NiftyReg results (ranging from 0.77 to 0.90), and the HD was significantly lower compared to NiftyReg. Furthermore, the proposed method did not suffer from unrealistic deformations as the NiftyReg did in the visual evaluation. Meanwhile, the execution time of the proposed method was much higher than NiftyReg (96.98 ± 11.88 s vs. 4.60 ± 0.49 s). In conclusion, the in‐house hybrid method gave better accuracy and more stable performance than NiftyReg.

show abstract

Section: Discussionsupporting

confidence: 92%

Comparison of an in‐house hybrid DIR method to NiftyReg on CBCT and CT images for head and neck cancer

Jiang

Huang

Ding

et al. 2022

J Applied Clin Med Phys

View full text Add to dashboard Cite

show abstract

“…This, in turn, leads to the images containing so-called streaking artifacts while at the same time being affected by a low contrast and a low signal-to-noise ratio (SNR) (Ahmad et al 2009, Schulze et al 2011. Streaking artifacts become especially problematic when CT-CBCT intensity-based registration is of interest, due to the intrinsic intensity inconsistencies they introduce between the CT and the CBCT image (Li et al 2016). Previous studies address this issue by incorporating intensity correction/matching procedures into existing intensity-based registration algorithms such as demon registration (Nithiananthan et al 2011, Zhen et al 2012, Park et al 2017 or optical flow (Ostergaard et al 2008, Li et al 2016.…”

Section: Introductionmentioning

confidence: 99%

Non-rigid CT/CBCT to CBCT registration for online external beam radiotherapy guidance

Zachiu¹,

Tijssen²,

Kotte³

et al. 2017

Phys. Med. Biol.

View full text Add to dashboard Cite

Image-guided external beam radiotherapy (EBRT) allows radiation dose deposition with a high degree of accuracy and precision. Guidance is usually achieved by estimating the displacements, via image registration, between cone beam computed tomography (CBCT) and computed tomography (CT) images acquired at different stages of the therapy. The resulting displacements are then used to reposition the patient such that the location of the tumor at the time of treatment matches its position during planning. Moreover, ongoing research aims to use CBCT-CT image registration for online plan adaptation. However, CBCT images are usually acquired using a small number of x-ray projections and/or low beam intensities. This often leads to the images being subject to low contrast, low signal-to-noise ratio and artifacts, which ends-up hampering the image registration process. Previous studies addressed this by integrating additional image processing steps into the registration procedure. However, these steps are usually designed for particular image acquisition schemes, therefore limiting their use on a case-by-case basis. In the current study we address CT to CBCT and CBCT to CBCT registration by the means of the recently proposed EVolution registration algorithm. Contrary to previous approaches, EVolution does not require the integration of additional image processing steps in the registration scheme. Moreover, the algorithm requires a low number of input parameters, is easily parallelizable and provides an elastic deformation on a point-by-point basis. Results have shown that relative to a pure CT-based registration, the intrinsic artifacts present in typical CBCT images only have a sub-millimeter impact on the accuracy and precision of the estimated deformation. In addition, the algorithm has low computational requirements, which are compatible with online image-based guidance of EBRT treatments.

show abstract

“…Therefore, it is challenging for these simple intensity-based registration software tools or methods to accurately register CT and CBCT images. 7 To address this challenge, some researchers integrated CT-CBCT intensity correction/matching step into conventional intensity-based registration methods such as Demons, [8][9][10] optical flow, 11 or viscous fluid model. 12 Hou et al 8 and Nithiananthan et al 9 employed global matching strategy.…”

Section: Introductionmentioning

confidence: 99%

“…However, there is always a considerable intensity difference between CBCT and CT images even though they are both x‐ray tomography imaging. Therefore, it is challenging for these simple intensity‐based registration software tools or methods to accurately register CT and CBCT images 7 …”

Section: Introductionmentioning

confidence: 99%

Unsupervised learning for deformable registration of thoracic CT and cone‐beam CT based on multiscale features matching with spatially adaptive weighting

Duan

Liu

et al. 2020

Medical Physics

View full text Add to dashboard Cite

Purpose Cone‐beam computed tomography (CBCT) is a common on‐treatment imaging widely used in image‐guided radiotherapy. Fast and accurate registration between the on‐treatment CBCT and planning CT is significant for and precise adaptive radiotherapy treatment (ART). However, existing CT–CBCT registration methods, which are mostly affine or time‐consuming intensity‐ based deformation registration, still need further study due to the considerable CT–CBCT intensity discrepancy and the artifacts in low‐quality CBCT images. In this paper, we propose a deep learning‐based CT–CBCT registration model to promote rapid and accurate CT–CBCT registration for radiotherapy. Methods The proposed CT–CBCT registration model consists of a registration network and an innovative deep similarity metric network. The registration network is a novel fully convolution network adapted specially for patch‐wise CT–CBCT registration. The metric network, going beyond intensity, automatically evaluates the high‐dimensional attribute‐based dissimilarity between the registered CT and CBCT images. In addition, considering the artifacts in low‐quality CBCT images, we add spatial weighting (SW) block to adaptively attach more importance to those informative voxels while inhibit the interference of artifact regions. Such SW‐based metric network is expected to extract the most meaningful and discriminative deep features, and form a more reliable CT–CBCT similarity measure to train the registration network. Results We evaluate the proposed method on clinical thoracic CBCT and CT dataset, and compare the registration results with some other common image similarity metrics and some state‐of‐the‐art registration algorithms. The proposed method provides the highest Structural Similarity index (86.17 ± 5.09), minimum Target Registration Error of landmarks (2.37 ± 0.32 mm), and the best DSC coefficient (78.71 ± 10.95) of tumor volumes. Moreover, our model also obtains comparable distance error of lung surfaces (1.75 ± 0.35 mm). Conclusion The proposed model shows both efficiency and efficacy for reliable thoracic CT–CBCT registration, and can generate the matched CT and CBCT images within few seconds, which is of great significance to clinical radiotherapy.

show abstract

Evaluation of deformable image registration for contour propagation between CT and cone-beam CT images in adaptive head and neck radiotherapy

Cited by 15 publications

References 24 publications

Comparison of an in‐house hybrid DIR method to NiftyReg on CBCT and CT images for head and neck cancer

Comparison of an in‐house hybrid DIR method to NiftyReg on CBCT and CT images for head and neck cancer

Non-rigid CT/CBCT to CBCT registration for online external beam radiotherapy guidance

Unsupervised learning for deformable registration of thoracic CT and cone‐beam CT based on multiscale features matching with spatially adaptive weighting

Contact Info

Product

Resources

About