Refine
Document Type
- Article (3)
Language
- English (3)
Has Fulltext
- no (3)
Is part of the Bibliography
- no (3)
Keywords
- Maximum A Posteriori estimation (1)
- Patient positioning (1)
- Setup error (1)
- Short arc (1)
- Stochastic triangulation (1)
Institute
Begutachtungsstatus
- peer-reviewed (2)
Background and purpose:
Fast and reliable tumor localization is an important part of today's radiotherapy utilizing new delivery techniques. This proof-of-principle study demonstrates the use of a method called herein 'stochastic triangulation' for this purpose. Stochastic triangulation uses very short imaging arcs and a few projections.
Materials and methods:
A stochastic Maximum A Posteriori (MAP) estimator is proposed based on an uncertainty-driven model of the acquisition geometry and inter-/intra-fractional deformable anatomy. The application of this method was designed to use the available linac-mounted cone-beam computed tomography (CBCT) and/or electronic portal imaging devices (EPID) for the patient setup based on short imaging arcs. For the proof-of-principle clinical demonstration, the MAP estimator was applied to 5 CBCT scans of a prostate cancer patient with 2 implanted gold markers. Estimation was performed for several (18) very short imaging arcs of 5° with 10 projections resulting in 90 estimations.
Results:
Short-arc stochastic triangulation led to residual radial errors compared to manual inspection with a mean value of 1.4mm and a standard deviation of 0.9 mm (median 1.2mm, maximum 3.8mm) averaged over imaging directions all around the patient. Furthermore, abrupt intra-fractional motion of up to 10mm resulted in radial errors with a mean value of 1.8mm and a standard deviation of 1.1mm (median 1.5mm, maximum 5.6mm). Slow periodic intra-fractional motions in the range of 12 mm resulted in radial errors with a mean value of 1.8mm and a standard deviation of 1.1mm (median 1.6mm, maximum 4.7 mm).
Conclusion:
Based on this study, the proposed stochastic method is fast, robust and can be used for inter- as well as intra-fractional target localization using current CBCT units.
Purpose:
The authors present a stochastic framework for radiotherapy patient positioning directly utilizing radiographic projections. This framework is developed to be robust against anatomical nonrigid deformations and to cope with challenging imaging scenarios, involving only a few cone beam CT projections from short arcs.
Methods:
Specifically, a Bayesian estimator (BE) is explicitly derived for the given scanning geometry. This estimator is compared to reference methods such as chamfer matching (CM) and the minimization of the median absolute error adapted as tools of robust image processing and statistics. In order to show the performance of the stochastic short-arc patient positioning method, a CIRS IMRT thorax phantom study is presented with movable markers and the utilization of an Elekta Synergy® XVI system. Furthermore, a clinical prostate CBCT scan of a Varian® On-Board Imager® system is utilized to investigate the robustness of the method for large variations of image quality (anterior-posterior vs lateral views).
Results:
The results show that the BE shifts reduce the initial setup error of up to 3 cm down to 3 mm at maximum for an imaging arc as short as 10° while CM achieves residual errors of 7 mm at maximum only for arcs longer than 40°. Furthermore, the BE can compensate robustly for low image qualities using several low quality projections simultaneously.
Conclusions:
In conclusion, an estimation method for marker-based patient positioning for short imaging arcs is presented and shown to be robust and accurate for deformable anatomies.
Purpose
We present a systematic Bayesian formulation of the stochastic localization/triangulation problem close to constraining interfaces.
Methods
For this purpose, the terminology of Bayesian estimation is summarized suitably for applied researchers including the presentation of Maximum Likelihood (ML), Maximum A Posteriori (MAP), and Minimum Mean Square Error (MMSE) estimation. Explicit estimators for triangulation are presented for the linear 2D parallel beam and the nonlinear 3D cone beam model. The priors in MAP and MMSE optionally incorporate (A) the hard constraints about the interface and (B) knowledge about the probability of the object with respect to the interface. All presented estimators are compared in several simulation studies for live acquisition scenarios with 10,000 samples each.
Results
First, the presented application shows that MAP and MMSE perform considerably better, leading to lower Root Mean Square Errors (RMSEs) in the simulation studies compared to the ML approach by typically introducing a bias. Second, utilizing priors including (A) and (B) is very beneficial compared to just including (A). Third, typically MMSE leads to better results than MAP, by the cost of significantly higher computational effort.
Conclusion
Depending on the specific application and prior knowledge, MAP and MMSE estimators strongly increase the estimation accuracy for localization close to interfaces.