Robust Deformable Image Registration Research Articles

A robust deformable image registration enhancement method based on radial basis function.

To develop and evaluate a robust deformable image registration (DIR) enhancement method based on radial basis function (RBF) expansion. To improve DIR accuracy using sparsely available measured displacements, it is crucial to estimate the motion correlation between the voxels. In the proposed method, we chose to derive this correlation from the initial displacement vector fields (DVFs), and represent it in the form of RBF expansion coefficients of the voxels. The method consists of three steps: (I) convert an initial DVF to a coefficient matrix comprising expansion coefficients of the Wendland's RBF; (II) modify the coefficient matrix under the guidance of sparely distributed landmarks to generate the post-enhancement coefficient matrix; and (III) convert the post-enhancement coefficient matrix to the post-enhancement DVF. The method was tested on five DIR algorithms using a digital phantom. 3D registration errors were calculated for comparisons between the pre-/post-enhancement DVFs and the ground-truth DVFs. Effects of the number and locations of landmarks on DIR enhancement were evaluated. After applying the DIR enhancement method, the 3D registration errors per voxel (unit: mm) were reduced from pre-enhancement to post-enhancement by 1.3 (2.4 to 1.1, 54.2%), 0.0 (0.9 to 0.9, 0.0%), 6.1 (8.2 to 2.1, 74.4%), 3.2 (4.7 to 1.5, 68.1%), and 1.7 (2.9 to 1.2, 58.6%) for the five tested DIR algorithms respectively. The average DIR error reduction was 2.5±2.3 mm (percentage error reduction: 51.1%±29.1%). 3D registration errors decreased inverse-exponentially as the number of landmarks increased, and were insensitive to the landmarks' locations in relation to the down-sampling DVF grids. We demonstrated the feasibility of a robust RBF-based method for enhancing DIR accuracy using sparsely distributed landmarks. This method has been shown robust and effective in reducing DVF errors using different numbers and distributions of landmarks for various DIR algorithms.

P200] Deformable Image Registration performances in head and neck patients: Impact of daily imaging quality

Purpose To investigate the accuracy and robustness of commercial algorithms for Deformable Image Registration (DIR) to propagate regions of interest (ROIs), against image quality typical of Cone Beam CT (CBCT) and Megavoltage Cone-beam CT (MVCT) images, using computational phantoms based on real head and neck (HN) patient images. Methods Eleven institutions joined the study with five commercial DIR solutions. Three real patient data-sets from different on board imaging devices (OBI by Varian, XVI by Elekta, and MVCT by Tomotherapy) were used with different image quality. Two specific Deformation Vector Fields (DVF) with realistic level of deformations, were generated with a dedicated software and then applied to each data-set. The accuracy of the algorithms was assessed by comparing the DIR-mapped ROIs from each center with those of reference, using the Dice Similarity Coefficient (DSC) and Mean Distance to Conformity (MDC) metrics. Prognostic factors of DIR performances were carried out. Results Based on more than 700 DIR-mapped ROIs the 4-way ANOVA analysis states that DIR algorithm, image quality and organ type are significant predictors of DIR performances. A slightly dependence from applied DVF is detected (p = 0.035) with a post hoc test. Putting together all CI data of all institutions for the first DVF, the mean CI was 0.88 ± 0.05 (1 SD), and the mean MDC is 0.38 ± 0.96 (1 SD); for the second DVF, the mean CI was 0.86 ± 0.11 (1 SD), and the mean MDC is 0.43 ± 1.08 (1 SD). The thyroid, located near the FOV border, was the underperformer organ with large errors (unacceptable). All other organs were mapped with an acceptable accuracy (at voxel size level). One imaging system performs statistically better than the others, and one algorithm produced unsatisfactory results respect all the others. Conclusions This work studies the impact of CBCT and MVCT image quality on DIR performance in a ground truth provided scenario. Clinical issues like Adaptive Radiation Therapy (ART) need accurate and robust DIR software. The reported result highlight that in HN district the image quality and the implemented algorithm are strictly related to DIR performances.

TU‐H‐CAMPUS‐JeP1‐04: Deformable Image Registration Performances in Pelvis Patients: Impact of CBCT Image Quality

Purpose:To investigate the accuracy and robustness, against image noise and artifacts (typical of CBCT images), of a commercial algorithm for deformable image registration (DIR), to propagate regions of interest (ROIs) in computational phantoms based on real prostate patient images.Methods:The Anaconda DIR algorithm, implemented in RayStation was tested. Two specific Deformation Vector Fields (DVFs) were applied to the reference data set (CTref) using the ImSimQA software, obtaining two deformed CTs. For each dataset twenty‐four different level of noise and/or capping artifacts were applied to simulate CBCT images. DIR was performed between CTref and each deformed CTs and CBCTs. In order to investigate the relationship between image quality parameters and the DIR results (expressed by a logit transform of the Dice Index) a bilinear regression was defined.Results:More than 550 DIR‐mapped ROIs were analyzed. The Statistical analysis states that deformation strenght and artifacts were significant prognostic factors of DIR performances, while noise appeared to have a minor role in DIR process as implemented in RayStation as expected by the image similarity metric built in the registration algorithm. Capping artifacts reveals a determinant role for the accuracy of DIR results. Two optimal values for capping artifacts were found to obtain acceptable DIR results (DICE> 075/ 0.85). Various clinical CBCT acquisition protocol were reported to evaluate the significance of the study.Conclusion:This work illustrates the impact of image quality on DIR performance. Clinical issues like Adaptive Radiation Therapy (ART) and Dose Accumulation need accurate and robust DIR software. The RayStation DIR algorithm resulted robust against noise, but sensitive to image artifacts. This result highlights the need of robustness quality assurance against image noise and artifacts in the commissioning of a DIR commercial system and underlines the importance to adopt optimized protocols for CBCT image acquisitions in ART clinical implementation.

TU‐H‐CAMPUS‐JeP1‐03: Results From a Multi Institution Study On Image Quality Impact in Deformable Image Registration

Purpose:Various commercial algorithms for deformable image registration (DIR) were tested to investigate their accuracy and robustness, against image noise and artifacts.Methods:Ten institutions with five available commercial solutions provided data to assess the agreement of DIR‐propagated ROIs with automatically drawn ROIs considered as ground‐truth for the comparison. The DIR algorithms were tested on real patient data (pelvis). A Deformation was applied to the reference data set (CTref) using the ImSimQA software, a (CTeval) and seven CBCT images with increasing level of noise and capping artifacts were simulated. Every center performed DIR between CTref, and the deformed datasets. The noise was defined as standard deviation normalized to signal in an homogeneous medium, artefacts as the difference of mean HU values between the central and peripheral region of an homogeneous medium. To investigate the relationship between image quality parameters and the DIR results a three way ANOVA was performed on logit function of DICE index.Results:Based on 480 DIR‐mapped ROIs the ANOVA test states that centers, ROIs and Image Quality are significant predictors of DIR performances. DIR conducted on CBCT simulated images is significant worst of the ones obtained on the original images. Increasing noise and artefacts didn't affect significantly DIR performances. Considering a limit for Dice Index of 0.75 one center underperform this level. DIR resulted significantly more accurate in rectal contours propagation.Conclusion:This work illustrates the effect of image noise to DIR performances in a ground truth provided scenario. Clinical issues like ART or Dose Accumulation need accurate and robust DIR software using CBCT images. For the range of artefacts and noise explored in this experiment the commercial software appeared to be robust. One centre didn't satisfy the minimum accuracy requirement showing that QA is mandatory to implement clinically DIR for ROI propagation.

TH-C-BRF-01: The Promise and Potential Pitfalls of Deformable Image Registration in Clinical Practice

Accurate and robust deformable image registration (DIR) is a key enabling technique in the clinical realization of two approaches for advancing radiation therapy treatment efficacy: adaptive radiation therapy and treatment response assessment. Currently there are a wide variety of DIR methods including the categories of splines, optical/diffusion, free-form, and biomechanical algorithms. All methods aim to translate information between image sets (including multi-modal data) in the presence of spatial deformation of tissues. However, recent research has shown that different DIR algorithms can yield substantially different results for the same reference deformation, and that DIR performance can be site and application dependent. As a result, errors can occur, and subsequent patient treatment can be compromised. There is a clear need for greater understanding of appropriate use of DIR techniques, as well as effective methods of validation, evaluation, and improvement. In this session, we will review the state-of-the-art concerning DIR development, clinical application, and performance evaluation. Novel DIR methods and evaluating technologies will be reviewed. Learning Objectives: 1. To understand the underlying principles and physics of current DIR techniques 2. To explore potential clinical applications and areas of high impact for DIR 3. To investigate sources of uncertainty, appropriate usage, and methods for validating and evaluating DIR performance.

MO‐C‐17A‐11: A Segmentation and Point Matching Enhanced Deformable Image Registration Method for Dose Accumulation Between HDR CT Images

Purpose:To propose and validate a novel and accurate deformable image registration (DIR) scheme to facilitate dose accumulation among treatment fractions of high‐dose‐rate (HDR) gynecological brachytherapy.Method:We have developed a method to adapt DIR algorithms to gynecologic anatomies with HDR applicators by incorporating a segmentation step and a point‐matching step into an existing DIR framework. In the segmentation step, random walks algorithm is used to accurately segment and remove the applicator region (AR) in the HDR CT image. A semi‐automatic seed point generation approach is developed to obtain the incremented foreground and background point sets to feed the random walks algorithm. In the subsequent point‐matching step, a feature‐based thin‐plate spline‐robust point matching (TPS‐RPM) algorithm is employed for AR surface point matching. With the resulting mapping, a DVF characteristic of the deformation between the two AR surfaces is generated by B‐spline approximation, which serves as the initial DVF for the following Demons DIR between the two AR‐free HDR CT images. Finally, the calculated DVF via Demons combined with the initial one serve as the final DVF to map doses between HDR fractions.Results:The segmentation and registration accuracy are quantitatively assessed by nine clinical HDR cases from three gynecological cancer patients. The quantitative results as well as the visual inspection of the DIR indicate that our proposed method can suppress the interference of the applicator with the DIR algorithm, and accurately register HDR CT images as well as deform and add interfractional HDR doses.Conclusions:We have developed a novel and robust DIR scheme that can perform registration between HDR gynecological CT images and yield accurate registration results. This new DIR scheme has potential for accurate interfractional HDR dose accumulation.This work is supported in part by the National Natural ScienceFoundation of China (no 30970866 and no 81301940).

SU-E-J-86: A Deformable Image Registration Method for Dose Accumulation Between IMRT and HDR CT Images

Purpose: To develop and validate a method to improve existing deformable image registration (DIR) algorithms for the registration between the high-dose-rate (HDR) brachytherapy CT image and the intensity-modulated radiation therapy (IMRT) CT image of gynecologic cancer patients for dose accumulation. Methods: Deformable registration of brachytherapy CT images and external beam CT images is challenging due to the presence of the applicator in the brachytherapy image. We have developed a method to adapt DIR algorithms to this situation. A segmentation step is used to remove the applicator, which is imaged in the HDR CT image but not in the IMRT image, leaving an empty virtual cavity. An artificial deformation vector field is then generated in the tissue around the applicator to deflate this cavity toward the center at each slice and restore the vagina to a state without applicator. Different schemes are employed for two different registration tasks, i.e., deforming HDR CT image to IMRT CT image, or vice versa. In both cases, demons algorithm is used as the main DIR method to perform registration. The generated final deformation vector fields are used to deform the dose grid from one CT anatomy to another for dose accumulation. Results: Visual inspection of the DIR results indicates that our proposed method can suppress the interference of the applicator to the DIR algorithm, and successfully and accurately register the HDR and IMRT CT images as well as deforming and adding doses from both treatment modalities. Conclusion: We have developed a novel and robust DIR scheme that can perform registration between IMRT CT image and HDR CT image for dose accumulation, and can yield accurate registration results. This work was supported in part by the National Natural Science Foundation of China (no 30970866), Major Program of Natural Science Foundation of Guangdong Province of China (no 10251051501000007)

TH-D-332-06: A Generalized Method for Achieving Realistic Deformable Image Registration (DIR) Using the Inverse Consistency Constraint

Purpose: To present a novel methodology for improving the performance of existing deformable image registration (DIR) algorithms by imposing an inverse consistency constraint (ICC), and achieving physiologically more realistic and robust DIR. Method and Materials: The proposed approach combines forward and reverse mappings of a registration algorithm by an inverse consistency constraint based on fixed point theory. Unlikely most ICC‐DIR algorithms, our method does not require explicitly computing inverse transformations, thus reducing both potential numerical errors and required computations. Here we applied the methodology to the Demons algorithm: inverse consistency Demons (iDemons). However the method is generalized to be applied to many current algorithms, such as accelerated demons and energy minimization methods. The validation and comparison of the iDemons and Demons was carried out on 4DCT clinical lungimaging based on: 1) cross‐correlation; 2) inverse consistency errors (ICE); 3) visual verification; 4) robustness. Results: The registration performance of iDemons was superior to Demons. While both methods achieved similar level of correlation coefficient after 200 iterations, 0.9908 for iDemons and 0.9891 for Demons, the average ICE is reduced by 91%: 0.17mm (iDemons) vs. 1.86mm (Demons), resulting in up to 3mm difference in corresponding contours. Since ICE mostly occurred around organ boundaries, minimizing ICE would improve clinical efficacy of DIR based auto‐contouring. Moreover, the iDemons algorithm was shown to be more robustness and less sensitive to parameter selection, which is significant if DIR is to be utilized clinically on routine basis by non‐specialists. Conclusion: The proposed methodology achieves physiologically more realistic DIR. iDemons achieved significant reduction of ICE compared to Demons, which was significant for organ boundaries. Registration results for iDemons were less sensitive to parameter selection, an important consideration for clinical implementation. This novel inverse consistency constraint represents a generalized formalism, which can be applied to other registration algorithms.

SU‐EE‐A3‐06: Deformable Image Registration in Cone‐Beam CT Images for Image‐Guided Adaptive Radiotherapy

Purpose: With the availability of on‐board imaging devices capable of constructing cone‐beam CT (CBCT) images, it is expected that there will be great interest in using volumetric CBCT for image‐guided adaptive radiotherapy. In order to fully utilize CBCT, automatic segmentation on CBCT images is one of key steps toward this goal. The purpose of this study is to implement a robust deformable image registration for auto‐segmentation. Method and Materials: In four head and neck cancer patients, we used our previously developed, image intensity‐based deformable image registration algorithm to register the planning CT with the 3–5 daily CBCT in three scenarios. First, the daily CBCT was directly used without modification. Second, we applied a generic look‐up‐table transformation to map the CBCT image intensity to the conventional CT intensity using the measured electron density calibration curves for both the conventional and CBCT scanners. In the third scenario, we proposed a wavelet‐based dynamic window/level histogram matching algorithm to map the CT number from CBCT image to the conventional CT image. Then the deformable image registration was performed in the modified CBCT images to map the anatomical structures from the planning CT to the corresponding CBCT images. Results: Without pre‐processing, we found that the CT numbers in CBCT images were inconsistent, especially in soft tissue regions and in patients with large body circumferences. The deformable image registration using the window/level histogram matching method performed the best with good consistency in delineating soft tissue structures. The algorithm is also computationally efficient. Conclusion: We implemented a wavelet‐based window/level histogram matching algorithm to pre‐process the CBCT to allow for more robust deformable image registration of with the reference planning CT. This implementation allows for volumetric CBCT‐guided adaptive radiotherapy.