Algorithm-enabled exploration of image-quality potential of cone-beam CT in image-guided radiation therapy

Han, Xiao; Pearson, Erik; Pelizzari, Charles A.; Al‐Hallaq, Hania; Sidky, Emil Y.; Bian, Junguo; Pan, Xiaochuan

doi:10.1088/0031-9155/60/12/4601

Cited by 27 publications

(27 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different strategies can be potentially used for the performance evaluation. For example, Han et al evaluated the constrained TV minimization-based reconstruction with parameters selected to have the highest performance according to different utility metrics charactering either reconstruction contrast or spatial resolution [20]. Thus the selected optimal parameters are strongly metric-dependent [20].…”

Section: Discussionmentioning

confidence: 99%

“…For example, Han et al evaluated the constrained TV minimization-based reconstruction with parameters selected to have the highest performance according to different utility metrics charactering either reconstruction contrast or spatial resolution [20]. Thus the selected optimal parameters are strongly metric-dependent [20]. In this work, we tuned the parameters for different algorithms such that the reconstructed images were at the same noise level and then compared algorithm performance at the matched noise level [49].…”

Section: Discussionmentioning

confidence: 99%

“…In this work, we tuned the parameters for different algorithms such that the reconstructed images were at the same noise level and then compared algorithm performance at the matched noise level [49]. Therefore, the parameters we used for TV reconstruction may not be optimal according to different metrics defined in [20]. We will compare the proposed penalties with TV by choosing metric-specific optimal parameters in a future work similar to that in [20].…”

Section: Discussionmentioning

confidence: 99%

“…Therefore, the parameters we used for TV reconstruction may not be optimal according to different metrics defined in [20]. We will compare the proposed penalties with TV by choosing metric-specific optimal parameters in a future work similar to that in [20]. As the parameters were not adjusted to be optimal according to certain specific evaluation metrics in our experiments, undesired artifacts existed in the reconstructed images in some cases.…”

Section: Discussionmentioning

confidence: 99%

“…Sidky and Pan proposed to minimize the TV penalty subject to a non-negative constraint for sparse-view projection reconstruction, where the TV objective function was minimized by steepest descent with an adaptive step size, and the non-negative constraint was enforced by projection onto convex sets (POCS) [13], [18], [19]. Han et al studied the selection of parameter values in the constrained TV minimization-based reconstruction and showed, with both visualization and quantitative assessments, that the reconstructions were greatly improved, relative to the clinical FDK reconstruction, in terms of the devised utility metrics [20]. Bian et al showed that the TV-based iterative algorithm was more robust to different data conditions, such as number of views and exposure levels, than the FDK algorithm for CBCT reconstruction [21].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Low-Dose CBCT Reconstruction Using Hessian Schatten Penalties

Liu

Xiang

et al. 2017

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

Cone-beam computed tomography (CBCT) has been widely used in radiation therapy. For accurate patient setup and treatment target localization, it is important to obtain high-quality reconstruction images. The total variation (TV) penalty has shown state-of-the-art performance in suppressing noise and preserving edges for statistical iterative image reconstruction, but it sometimes leads to the so-called staircase effect. In this study, we proposed to use a new family of penalties — the Hessian Schatten (HS) penalties — for CBCT reconstruction. Consisting of the second-order derivatives, the HS penalties are able to reflect the smooth intensity transitions of the underlying image without introducing the staircase effect. We discussed and compared the behaviors of several convex HS penalties with orders 1, 2, and +∞ for CBCT reconstruction. We used the majorization-minimization approach with a primal-dual formulation for the corresponding optimization problem. Experiments on two digital phantoms and two physical phantoms demonstrated the proposed penalty family’s outstanding performance over TV in suppressing the staircase effect, and the HS penalty with order 1 had the best performance among the HS penalties tested.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Low-Dose CBCT Reconstruction Using Hessian Schatten Penalties

Liu

Xiang

et al. 2017

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

show abstract

Patient‐specific calibration of cone‐beam computed tomography data sets for radiotherapy dose calculations and treatment plan assessment

MacFarlane

Wong

Hoover

et al. 2018

J Applied Clin Med Phys

View full text Add to dashboard Cite

PurposeIn this work, we propose a new method of calibrating cone beam computed tomography (CBCT) data sets for radiotherapy dose calculation and plan assessment. The motivation for this patient‐specific calibration (PSC) method is to develop an efficient, robust, and accurate CBCT calibration process that is less susceptible to deformable image registration (DIR) errors.MethodsInstead of mapping the CT numbers voxel‐by‐voxel with traditional DIR calibration methods, the PSC methods generates correlation plots between deformably registered planning CT and CBCT voxel values, for each image slice. A linear calibration curve specific to each slice is then obtained by least‐squares fitting, and applied to the CBCT slice's voxel values. This allows each CBCT slice to be corrected using DIR without altering the patient geometry through regional DIR errors.A retrospective study was performed on 15 head‐and‐neck cancer patients, each having routine CBCTs and a middle‐of‐treatment re‐planning CT (reCT). The original treatment plan was re‐calculated on the patient's reCT image set (serving as the gold standard) as well as the image sets produced by voxel‐to‐voxel DIR, density‐overriding, and the new PSC calibration methods. Dose accuracy of each calibration method was compared to the reference reCT data set using common dose‐volume metrics and 3D gamma analysis. A phantom study was also performed to assess the accuracy of the DIR and PSC CBCT calibration methods compared with planning CT.ResultsCompared with the gold standard using reCT, the average dose metric differences were ≤ 1.1% for all three methods (PSC: −0.3%; DIR: −0.7%; density‐override: −1.1%). The average gamma pass rates with thresholds 3%, 3 mm were also similar among the three techniques (PSC: 95.0%; DIR: 96.1%; density‐override: 94.4%).ConclusionsAn automated patient‐specific calibration method was developed which yielded strong dosimetric agreement with the results obtained using a re‐planning CT for head‐and‐neck patients.

show abstract

Regularization strategies in statistical image reconstruction of low‐dose x‐ray CT: A review

et al. 2018

View full text Add to dashboard Cite

Statistical image reconstruction (SIR) methods have shown potential to substantially improve the image quality of low-dose x-ray computed tomography (CT) as compared to the conventional filtered back-projection (FBP) method. According to the maximum a posteriori (MAP) estimation, the SIR methods are typically formulated by an objective function consisting of two terms: (a) a data-fidelity term that models imaging geometry and physical detection processes in projection data acquisition, and (b) a regularization term that reflects prior knowledge or expectations of the characteristics of the to-be-reconstructed image. SIR desires accurate system modeling of data acquisition, while the regularization term also has a strong influence on the quality of reconstructed images. A variety of regularization strategies have been proposed for SIR in the past decades, based on different assumptions, models, and prior knowledge. In this paper, we review the conceptual and mathematical bases of these regularization strategies and briefly illustrate their efficacies in SIR of low-dose CT.

show abstract

Algorithm-enabled exploration of image-quality potential of cone-beam CT in image-guided radiation therapy

Cited by 27 publications

References 43 publications

Low-Dose CBCT Reconstruction Using Hessian Schatten Penalties

Low-Dose CBCT Reconstruction Using Hessian Schatten Penalties

Patient‐specific calibration of cone‐beam computed tomography data sets for radiotherapy dose calculations and treatment plan assessment

Regularization strategies in statistical image reconstruction of low‐dose x‐ray CT: A review

Contact Info

Product

Resources

About