Cone-beam Computed Tomography Image Registration Research Articles

Unsupervised computed tomography and cone-beam computed tomography image registration using a dual attention network.

The registration of computed tomography (CT) and cone-beam computed tomography (CBCT) plays a key role in image-guided radiotherapy (IGRT). However, the large intensity variation between CT and CBCT images limits the registration performance and its clinical application in IGRT. In this study, a learning-based unsupervised approach was developed to address this issue and accurately register CT and CBCT images by predicting the deformation field. A dual attention module was used to handle the large intensity variation between CT and CBCT images. Specifically, a scale-aware position attention block (SP-BLOCK) and a scale-aware channel attention block (SC-BLOCK) were employed to integrate contextual information from the image space and channel dimensions. The SP-BLOCK enhances the correlation of similar features by weighting and aggregating multi-scale features at different positions, while the SC-BLOCK handles the multiple features of all channels to selectively emphasize dependencies between channel maps. The proposed method was compared with existing mainstream methods on the 4D-LUNG data set. Compared to other mainstream methods, it achieved the highest structural similarity (SSIM) and dice similarity coefficient (DICE) scores of 86.34% and 89.74%, respectively, and the lowest target registration error (TRE) of 2.07 mm. The proposed method can register CT and CBCT images with high accuracy without the needs of manual labeling. It provides an effective way for high-accuracy patient positioning and target localization in IGRT.

Multi-Intensity Optimization-Based CT and Cone Beam CT Image Registration

Cancer is a highly lethal disease that is mainly treated by image-guided radiotherapy. Because the low dose of cone beam CT is less harmful to patients, cone beam CT images are often used for target delineation in image-guided radiotherapy of various cancers, especially in breast and lung cancer. However, breathing and heartbeat can cause position errors in images taken during different periods, and the low dose of cone beam CT also results in insufficient imaging clarity, rendering existing registration methods unable to meet the CT and cone beam CT registration tasks. In this paper, we propose a novel multi-intensity optimization-based CT and cone beam CT registration method. First, we use a multi-weighted mean curvature filtering algorithm to preserve the multi-intensity details of the input image pairs. Then, the strong edge retention results are registered using and intensity-based method to obtain the multi-intensity registration results. Next, a novel evaluation method called intersection mutual information is proposed to evaluate the registration accuracy of the different multi-intensity registration results. Finally, we determine the optimal registration transformation by intersection mutual information and apply it to the input image pairs to obtain the final registration results. The experimental results demonstrate the excellent performance of the proposed method, meeting the requirements of image-guided radiotherapy.

Enhancement of cone beam CT image registration by super-resolution pre-processing algorithm.

In order to enhance cone-beam computed tomography (CBCT) image information and improve the registration accuracy for image-guided radiation therapy, we propose a super-resolution (SR) image enhancement method. This method uses super-resolution techniques to pre-process the CBCT prior to registration. Three rigid registration methods (rigid transformation, affine transformation, and similarity transformation) and a deep learning deformed registration (DLDR) method with and without SR were compared. The five evaluation indices, the mean squared error (MSE), mutual information, Pearson correlation coefficient (PCC), structural similarity index (SSIM), and PCC + SSIM, were used to validate the results of registration with SR. Moreover, the proposed method SR-DLDR was also compared with the VoxelMorph (VM) method. In rigid registration with SR, the registration accuracy improved by up to 6% in the PCC metric. In DLDR with SR, the registration accuracy was improved by up to 5% in PCC + SSIM. When taking the MSE as the loss function, the accuracy of SR-DLDR is equivalent to that of the VM method. In addition, when taking the SSIM as the loss function, the registration accuracy of SR-DLDR is 6% higher than that of VM. SR is a feasible method to be used in medical image registration for planning CT (pCT) and CBCT. The experimental results show that the SR algorithm can improve the accuracy and efficiency of CBCT image alignment regardless of which alignment algorithm is used.

Magnetic Resonance Imaging and Cone Beam CT Image Registration for Osseous and Soft Tissue Abnormalities of Temporomandibular Joint: A Systematic Review

Abstract Introduction Proper diagnosis of temporomandibular disorders (TMDs) is paramount for proper management. Cone beam computed tomography (CBCT) provides additional information about osseous components of TMJ, but diagnosis of soft tissue abnormalities is still a challenging issue for the clinicians. To overcome this, magnetic resonance imaging (MRI) and CBCT registration has shown to be a new promising technique that combines key features of both images into one single fusion image. The purpose of this systematic review was to compile available literature regarding the usefulness of CBCT and MRI-CBCT fusion imaging in proper diagnosis of temporomandibular joint (TMJ) abnormalities. Materials and Methods Computerized literature search from January 2010 to 2020 using five databases: PubMed (MEDLINE) database (National Library of Medicine, NCBI), Cochrane Database, Scopus, Embase, and Science Direct was conducted by use of medical specific headings (MeSH) keywords. Results Only 10 articles met the inclusion criteria. Eight articles showed that CBCT accurately detected osseous components of TMJ but has low reliability in diagnosing soft tissue articular disc disorders. Relationship between hard and soft tissues were clearly visualized in fused images. Conclusion MRI-CBCT image registration is a reliable diagnostic tool for simultaneous evaluation of osseous and soft tissue abnormalities of TMJ and in future more studies should be encouraged to validate its accuracy for TMJ analysis.

Patch-based field-of-view matching in multi-modal images for electroporation-based ablations

Various multi-modal imaging sensors are currently involved at different steps of an interventional therapeutic work-flow. Cone beam computed tomography (CBCT), computed tomography (CT) or Magnetic Resonance (MR) images thereby provides complementary functional and/or structural information of the targeted region and organs at risk. Merging this information relies on a correct spatial alignment of the observed anatomy between the acquired images. This can be achieved by the means of multi-modal deformable image registration (DIR), demonstrated to be capable of estimating dense and elastic deformations between images acquired by multiple imaging devices. However, due to the typically different field-of-view (FOV) sampled across the various imaging modalities, such algorithms may severely fail in finding a satisfactory solution.In the current study we propose a new fast method to align the FOV in multi-modal 3D medical images. To this end, a patch-based approach is introduced and combined with a state-of-the-art multi-modal image similarity metric in order to cope with multi-modal medical images. The occurrence of estimated patch shifts is computed for each spatial direction and the shift value with maximum occurrence is selected and used to adjust the image field-of-view. The performance of the proposed method – in terms of both registration accuracy and computational needs – is analyzed in the practical case of on-line irreversible electroporation procedures. In total, 30 pairs of pre-/per-operative IRE images are considered to illustrate the efficiency of our algorithm.We show that a regional registration approach using voxel patches provides a good structural compromise between the voxel-wise and “global shifts” approaches. The method was thereby beneficial for CT to CBCT and MRI to CBCT registration tasks, especially when highly different image FOVs are involved. Besides, the benefit of the method for CT to CBCT and MRI to CBCT image registration is analyzed, including the impact of artifacts generated by percutaneous needle insertions. Additionally, the computational needs using commodity hardware are demonstrated to be compatible with clinical constraints in the practical case of on-line procedures. The proposed patch-based workflow thus represents an attractive asset for DIR at different stages of an interventional procedure.

A graphical user interface for automated 2- or 3-dimensional image registration in dental treatment recovery planning: the DentIR application.

Cone-beam computed tomography (CBCT) scans enable quantification of interproximal bone loss after implant procedures in dental patients. In order for this quantification to be accurate, software is typically used to manipulate image sets captured before and after implantation to obtain their exact registration (i.e., alignment). However, no affordable CBCT image registration software is currently available for dental applications. Thus, the aim of the present study was to develop a freely available graphical user interface, called DentIR, that automates 2-dimensional (2-D) or 3-D image registration for use in planning dental treatment. The DentIR app was designed using the MATLAB environment, downloaded to a desktop personal computer (PC and Mac), and tested for its ease of use and alignment accuracy in the absence of the MATLAB environment. The DentIR app enabled previewing of the CBCT images in 3-D to allow for filtering of each frame to reduce noise and blurring before registration. The 2-D or 3-D registration was tested with four transformation methods. The accuracy of each method was assessed by comparing the mean squared error and the peak signal-to-noise ratio values that were provided by the DentIR app. The registered images could be saved as Portable Network Graphics (PNG) images. The free, user-friendly DentIR app was easily downloadable to Mac or PC platforms. It provided accurate image registration to aid in the planning of dental treatment. Future updates of the DentIR app include adding the ability to register more than two images at once, enhancing image editing options and enabling registration of a cropped portion of the image for more in-depth analyses.

"Dose of the day" based on cone beam computed tomography and deformable image registration for lung cancer radiotherapy.

PurposeAdaptive radiotherapy (ART) has potential to reduce toxicity and facilitate safe dose escalation. Dose calculations with the planning CT deformed to cone beam CT (CBCT) have shown promise for estimating the “dose of the day”. The purpose of this study is to investigate the “dose of the day” calculation accuracy based on CBCT and deformable image registration (DIR) for lung cancer radiotherapy.MethodsA total of 12 lung cancer patients were identified, for which daily CBCT imaging was performed for treatment positioning. A re‐planning CT (rCT) was acquired after 20 Gy for all patients. A virtual CT (vCT) was created by deforming initial planning CT (pCT) to the simulated CBCT that was generated from deforming CBCT to rCT acquired on the same day. Treatment beams from the initial plan were copied to the vCT and rCT for dose calculation. Dosimetric agreement between vCT‐based and rCT‐based accumulated doses was evaluated using the Bland‐Altman analysis.ResultsMean differences in dose‐volume metrics between vCT and rCT were smaller than 1.5%, and most discrepancies fell within the range of ± 5% for the target volume, lung, esophagus, and heart. For spinal cord Dmax, a large mean difference of −5.55% was observed, which was largely attributed to very limited CBCT image quality (e.g., truncation artifacts).ConclusionThis study demonstrated a reasonable agreement in dose‐volume metrics between dose accumulation based on vCT and rCT, with the exception for cases with poor CBCT image quality. These findings suggest potential utility of vCT for providing a reasonable estimate of the “dose of the day”, and thus facilitating the process of ART for lung cancer.

Feasibility study of intraoperative cone‐beam CT navigation for benign bone tumour surgery

BackgroundIntraoperative cone‐beam computed tomography (CBCT) offers the advantage of navigation on the current anatomical situation and the possibility to take a control scan. We assessed the feasibility of using intraoperative CBCT for navigated intralesional curettage.MethodsNine benign bone tumour patients were studied. Feasibility was assessed by describing the workflow and indications for navigation, scoring CBCT image quality and registration accuracy, and measuring scan and navigation set‐up times. Short‐term follow‐up was described.ResultsCBCT navigation was successful in all patients. Median tumour visibility, tumour delineation, and vital structure visibility scores were good. Median registration accuracy score was very good. Median scan and verification times were 5 and 3 minutes, respectively. One patient had a tumour recurrence after 6 months.ConclusionsIntraoperative CBCT navigation is feasible and safe. Indications for use of navigation in clinical practice are closeness to vital structures, complexly shaped tumours or bone, minimally invasive surgery, and repeated surgery.

Evidence-based region of interest matching guidelines for sarcoma volumetric image-guided radiation therapy.

Region-of-interest (ROI) guidelines for online cone-beam computed tomography (CBCT) radiotherapy may improve matching reproducibility and reduce inter-user variability of soft tissue sarcoma (STS) image guidance. The purpose of this work is to standardize ROI STS CBCT image registration guidelines using the plan-do-study-act (PDSA) cycle for the lower extremity, retroperitoneal, pelvis, and thorax. Based on anatomic bony surrogates, initial ROI matching guidelines for STS were developed by a team of radiation therapists, physicists and oncologists (Plan). Retrospective, qualitative evaluation of the guidelines was completed by the designated sarcoma lead therapist to determine clinical feasibility (Do). Validation of the ROI guidelines was performed through independent evaluation by radiation therapy CBCT imaging experts on a cohort of 10 patients per anatomic region (Study). Draft ROI guidelines were evaluated by 2 independent observers who registered weekly CBCT images to test their validity. Each observer assessed 5 patients per anatomic site, testing ROI options for accuracy of image registration and feasibility, while some ROI borders were adjusted based on algorithm matching performance. Validated ROI guidelines were presented to the sarcoma multidisciplinary site group, and an inter-professional committee of imaging experts for approval prior to clinical implementation (Act). ROI matching guidelines for STS IGRT were standardized for 4 anatomic sites using the PDSA cycle for change testing and implementation. IGRT guidelines are intended to improve STS image registration reproducibility, and in turn, are expected to improve the confidence of IGRT decision making and workflow efficiencies for a rare disease with diverse presentation.

Comparison of Online 6 Degree-of-Freedom Image Registration of Varian TrueBeam Cone-Beam CT and BrainLab ExacTrac X-Ray for Intracranial Radiosurgery.

Purpose:The study was aimed to compare online 6 degree-of-freedom image registrations of TrueBeam cone-beam computed tomography and BrainLab ExacTrac X-ray imaging systems for intracranial radiosurgery.Methods:Phantom and patient studies were performed on a Varian TrueBeam STx linear accelerator (version 2.5), which is integrated with a BrainLab ExacTrac imaging system (version 6.1.1). The phantom study was based on a Rando head phantom and was designed to evaluate isocenter location dependence of the image registrations. Ten isocenters at various locations representing clinical treatment sites were selected in the phantom. Cone-beam computed tomography and ExacTrac X-ray images were taken when the phantom was located at each isocenter. The patient study included 34 patients. Cone-beam computed tomography and ExacTrac X-ray images were taken at each patient’s treatment position. The 6 degree-of-freedom image registrations were performed on cone-beam computed tomography and ExacTrac, and residual errors calculated from cone-beam computed tomography and ExacTrac were compared.Results:In the phantom study, the average residual error differences (absolute values) between cone-beam computed tomography and ExacTrac image registrations were 0.17 ± 0.11 mm, 0.36 ± 0.20 mm, and 0.25 ± 0.11 mm in the vertical, longitudinal, and lateral directions, respectively. The average residual error differences in the rotation, roll, and pitch were 0.34° ± 0.08°, 0.13° ± 0.09°, and 0.12° ± 0.10°, respectively. In the patient study, the average residual error differences in the vertical, longitudinal, and lateral directions were 0.20 ± 0.16 mm, 0.30 ± 0.18 mm, 0.21 ± 0.18 mm, respectively. The average residual error differences in the rotation, roll, and pitch were 0.40°± 0.16°, 0.17° ± 0.13°, and 0.20° ± 0.14°, respectively. Overall, the average residual error differences were <0.4 mm in the translational directions and <0.5° in the rotational directions. ExacTrac X-ray image registration is comparable to TrueBeam cone-beam computed tomography image registration in intracranial treatments.

Evaluation of Elekta 4D cone beam CT-based automatic image registration for radiation treatment of lung cancer.

The study was aimed to evaluate the precision of Elekta four-dimensional (4D) cone beam CT (CBCT)-based automatic dual-image registrations using different landmarks for clipbox for radiation treatment of lung cancer. 30 4D CBCT scans from 15 patients were studied. 4D CBCT images were registered with reference CT images using dual-image registration: a clipbox registration and a mask registration. The image registrations performed in clinic using a physician-defined clipbox, were reviewed by physicians, and were taken as the standard. Studies were conducted to evaluate the automatic dual registrations using three kinds of landmarks for clipbox: spine, spine plus internal target volume (ITV) and lung (including as much of the lung as possible). Translational table shifts calculated from the automatic registrations were compared with those of the standard. The mean of the table shift differences in the lateral direction were 0.03, 0.03 and 0.03 cm, for clipboxes based on spine, spine plus ITV and lung, respectively. The mean of the shift differences in the longitudinal direction were 0.08, 0.08 and 0.08 cm, respectively. The mean of the shift differences in the vertical direction were 0.03, 0.03 and 0.03 cm, respectively. The automatic registrations using three different landmarks for clipbox showed similar results. One can use any of the three landmarks in 4D CBCT dual-image registration. Advance in knowledge: The study provides knowledge and recommendations for application of Elekta 4D CBCT image registration in radiation therapy of lung cancer.

SU-E-J-89: Comparative Analysis of MIM and Velocity's Image Deformation Algorithm Using Simulated KV-CBCT Images for Quality Assurance

Purpose: Deformable image registration (DIR) is used routinely in the clinic without a formalized quality assurance (QA) process. Using simulated deformations to digitally deform images in a known way and comparing to DIR algorithm predictions is a powerful technique for DIR QA. This technique must also simulate realistic image noise and artifacts, especially between modalities. This study developed an algorithm to create simulated daily kV cone-beam computed-tomography (CBCT) images from CT images for DIR QA between these modalities. Methods: A Catphan and physical head-and-neck phantom, with known deformations, were used. CT and kV-CBCT images of the Catphan were utilized to characterize the changes in Hounsfield units, noise, and image cupping that occur between these imaging modalities. The algorithm then imprinted these changes onto a CT image of the deformed head-and-neck phantom, thereby creating a simulated-CBCT image. CT and kV-CBCT images of the undeformed and deformed head-and-neck phantom were also acquired. The Velocity and MIM DIR algorithms were applied between the undeformed CT image and each of the deformed CT, CBCT, and simulated-CBCT images to obtain predicted deformations. The error between the known and predicted deformations was used as a metric to evaluate the quality of the simulated-CBCT image. Ideally, themore » simulated-CBCT image registration would produce the same accuracy as the deformed CBCT image registration. Results: For Velocity, the mean error was 1.4 mm for the CT-CT registration, 1.7 mm for the CT-CBCT registration, and 1.4 mm for the CT-simulated-CBCT registration. These same numbers were 1.5, 4.5, and 5.9 mm, respectively, for MIM. Conclusion: All cases produced similar accuracy for Velocity. MIM produced similar values of accuracy for CT-CT registration, but was not as accurate for CT-CBCT registrations. The MIM simulated-CBCT registration followed this same trend, but overestimated MIM DIR errors relative to the CT-CBCT registration.« less

Development of Evidence-Based Region of Interest Matching Guidelines to Standardize Volumetric Image-Guided Radiation Therapy Practice for Sarcoma

Development of Evidence-Based Region of Interest Matching Guidelines to Standardize Volumetric Image-Guided Radiation Therapy Practice for Sarcoma

Investigating CT to CBCT image registration for head and neck proton therapy as a tool for daily dose recalculation.

Intensity modulated proton therapy (IMPT) of head and neck (H&N) cancer patients may be improved by plan adaptation. The decision to adapt the treatment plan based on a dose recalculation on the current anatomy requires a diagnostic quality computed tomography (CT) scan of the patient. As gantry-mounted cone beam CT (CBCT) scanners are currently being offered by vendors, they may offer daily or weekly updates of patient anatomy. CBCT image quality may not be sufficient for accurate proton dose calculation and it is likely necessary to perform CBCT CT number correction. In this work, the authors investigated deformable image registration (DIR) of the planning CT (pCT) to the CBCT to generate a virtual CT (vCT) to be used for proton dose recalculation. Datasets of six H&N cancer patients undergoing photon intensity modulated radiation therapy were used in this study to validate the vCT approach. Each dataset contained a CBCT acquired within 3 days of a replanning CT (rpCT), in addition to a pCT. The pCT and rpCT were delineated by a physician. A Morphons algorithm was employed in this work to perform DIR of the pCT to CBCT following a rigid registration of the two images. The contours from the pCT were deformed using the vector field resulting from DIR to yield a contoured vCT. The DIR accuracy was evaluated with a scale invariant feature transform (SIFT) algorithm comparing automatically identified matching features between vCT and CBCT. The rpCT was used as reference for evaluation of the vCT. The vCT and rpCT CT numbers were converted to stopping power ratio and the water equivalent thickness (WET) was calculated. IMPT dose distributions from treatment plans optimized on the pCT were recalculated with a Monte Carlo algorithm on the rpCT and vCT for comparison in terms of gamma index, dose volume histogram (DVH) statistics as well as proton range. The DIR generated contours on the vCT were compared to physician-drawn contours on the rpCT. The DIR accuracy was better than 1.4 mm according to the SIFT evaluation. The mean WET differences between vCT (pCT) and rpCT were below 1 mm (2.6 mm). The amount of voxels passing 3%/3 mm gamma criteria were above 95% for the vCT vs rpCT. When using the rpCT contour set to derive DVH statistics from dose distributions calculated on the rpCT and vCT the differences, expressed in terms of 30 fractions of 2 Gy, were within [-4, 2 Gy] for parotid glands (D(mean)), spinal cord (D(2%)), brainstem (D(2%)), and CTV (D(95%)). When using DIR generated contours for the vCT, those differences ranged within [-8, 11 Gy]. In this work, the authors generated CBCT based stopping power distributions using DIR of the pCT to a CBCT scan. DIR accuracy was below 1.4 mm as evaluated by the SIFT algorithm. Dose distributions calculated on the vCT agreed well to those calculated on the rpCT when using gamma index evaluation as well as DVH statistics based on the same contours. The use of DIR generated contours introduced variability in DVH statistics.

Phantom based evaluation of CT to CBCT image registration for proton therapy dose recalculation

The ability to perform dose recalculation on the anatomy of the day is important in the context of adaptive proton therapy. The objective of this study was to investigate the use of deformable image registration (DIR) and cone beam CT (CBCT) imaging to generate the daily stopping power distribution of the patient. We investigated the deformation of the planning CT scan (pCT) onto daily CBCT images to generate a virtual CT (vCT) using a deformable phantom designed for the head and neck (H & N) region.The phantom was imaged at a planning CT scanner in planning configuration, yielding a pCT and in deformed, treatment day configuration, yielding a reference CT (refCT). The treatment day configuration was additionally scanned at a CBCT scanner. A Morphons DIR algorithm was used to generate a vCT. The accuracy of the vCT was evaluated by comparison to the refCT in terms of corresponding features as identified by an adaptive scale invariant feature transform (aSIFT) algorithm. Additionally, the vCT CT numbers were compared to those of the refCT using both profiles and regions of interest and the volumes and overlap (DICE coefficients) of various phantom structures were compared. The water equivalent thickness (WET) of the vCT, refCT and pCT were also compared to evaluate proton range differences. Proton dose distributions from the same initial fluence were calculated on the refCT, vCT and pCT and compared in terms of proton range. The method was tested on a clinical dataset using a replanning CT scan acquired close in time to a CBCT scan as reference using the WET evaluation.Results from the aSIFT investigation suggest a deformation accuracy of 2–3 mm. The use of the Morphon algorithm did not distort CT number intensity in uniform regions and WET differences between vCT and refCT were of the order of 2% of the proton range. This result was confirmed by proton dose calculations. The patient results were consistent with phantom observations. In conclusion, our phantom study suggests the vCT approach is adequate for proton dose recalculation on the basis of CBCT imaging.

How Long Does It Take? An Analysis Of Volumetric Image Assessment Time

To measure the time required to register and assess volumetric cone-beam computed tomography (CBCT) images for image-guided radiation therapy (IGRT). Under ethics approval, CBCT-IGRT data from 4592 patients treated from 2007 to 2010 across 8 linear accelerators were retrospectively analyzed. From the CBCT imaging database, translational discrepancies in the left/right (L/R), cranial/caudal (C/C), and anterior/posterior (A/P) directions recorded from daily localization CBCT images were extracted to quantify setup displacements for each patient. A 3-dimensional (3D) vector was calculated as (x2 + y2 + z2)½, where x = L/R, y = C/C, and z = A/P. The time stamps associated with each CBCT's reconstruction and image registration were also extracted. The “IGRT Decision Time” (IGRT-DT) is defined as the difference between the timestamp of image registration approval and timestamp of reconstruction acceptance by the end user (i.e. radiation therapists). 4592 patients from 13 treatment sites were included for analysis. From 117301 localization CBCTs, the mean ± standard deviation (SD) of the 3D vector displacement was 4.8 ± 4.1 mm. The mean(SD) IGRT-DT was 79.1(51.2) seconds, with no differences observed between year of treatment or treatment site. A plot of IGRT-DT versus vector displacement yield a positive correlation (slope = 3.1), indicating in clinical practice, a larger observed discrepancy between the planned and CBCT images resulted in a longer decision time. Registrations resulting in vector displacements exceeding 4 cm were further investigated (n = 38, 16 patients). Two CBCTs had a true vector displacement of > 4 cm: one case was due to a large patient with unreliable skin setup marks, and the other due to poor immobilization despite repeated resetups and repositioning. The causality of the other CBCTs with large offsets include 24 CBCTs that had offsets introduced into the setup to visualize adjacent soft tissue anatomy, and 12 CBCTs were imaged at an incorrect setup point (i.e. incorrect pre-treatment setup shift implemented, or setup on an incorrect tattoo). The average time for radiation therapists to register, assess and approve volumetric images at our institution is approximately 1.5 minutes. The CBCT image registration and assessment process is efficient, as shown by the consistent IGRT-DTs over 4 years and 8 linear accelerators. CBCT images exhibiting larger offsets at image registration are associated with an increase in decision time, reflective of the time required for critical assessment and decision making.

SU‐FF‐I‐103: A Segmentation‐Based Deformable Registration Algorithm for Cone Beam CT and Helical CT Images

Purpose: Optical flow‐based deformable registration algorithm, which assumes that the image intensity does not change during the deformation, is not suitable for cone beam CT (CBCT) and Helical CT (HCT) image registration because CBCT image possesses a intensity inhomogeneity due to Compton scatter effect. The objective of this study is to develop a new deformable registration algorithm for CBCT and HCT image registration. Method and Materials: The proposed deformable registration algorithm for CBCT and HCT image registration is divided into two steps. First, a maximum a posteriori probability (MAP) based adaptive segmentation method is utilized to accurately segment the voxels of CBCT images and HCT images into four tissue types: air, fat, muscle and bone, even in the presence of severe image intensity inhomogeneity due to Compton scatter. The intensities of the segmented images represent the probability of each voxel belonging to different tissue types and are globally uniform. Thus, the image intensity inhomogeneity effect can be eliminated. Second, an optical flow based deformable registration algorithm is applied to the segmented images, instead of original CBCT and HCT image, to determine the deformation between CBCT and HCT images. Thus, the CBCT image can be deformed to match the HCT images. Results: The performance of the proposed deformable registration algorithm was tested using both digital phantoms and clinical CBCT images. Our studies showed that the conventional optical flow‐based method failed to converge to expected solution due to intensity inhomogeneity effect, and its registration results showed a severe mismatch on many detailed structures between CBCT and HCT images. In contrary, the proposed method matched CBCT and HCT image very well. Conclusion: The proposed segmentation‐based deformable registration algorithm is capable of CBCT and HCT image registration.

Comparison of Spine, Carina, and Tumor as Registration Landmarks for Volumetric Image-Guided Lung Radiotherapy

To assess the feasibility, reproducibility, and accuracy of volumetric lung image guidance using different thoracic landmarks for image registration. In 30 lung patients, four independent observers conducted automated and manual image registrations on Day 1 cone-beam computed tomography data sets using the spine, carina, and tumor (720 image registrations). The image registration was timed, and the couch displacements were recorded. The intraclass correlation was used to assess reproducibility, and the Bland-Altman analysis was used to compare the automatic and manual matching methods. Tumor coverage (accuracy) was assessed through grading the tumor position after image matching against the internal target volume and planning target volume. The image-guided process took an average of 1 min for all techniques, with the exception of manual tumor matching, which took 4 min. Reproducibility was greatest for automatic carina matching (intraclass correlation, 0.90-0.93) and lowest for manual tumor matching (intraclass correlation, 0.07-0.43) in the left-right, superoinferior, and anteroposterior directions, respectively. The Bland-Altman analysis showed no significant difference between the automatic and manual registration methods. The tumor was within the internal target volume 62% and 60% of the time and was outside the internal target volume, but within the planning target volume, 38% and 40% of the time after automatic spine and automatic carina matching, respectively. For advanced lung cancer, the spine or carina can be used equally for cone-beam computed tomography image registration without compromising target coverage. The carina was more reproducible than the spine, but additional analysis is required to confirm its validation as a tumor surrogate. Soft-tissue registration is unsuitable at present, given the limitations in contrast resolution and the high interobserver variability.

Comparison of Kilovoltage Cone-Beam Computed Tomography With Megavoltage Projection Pairs for Paraspinal Radiosurgery Patient Alignment and Position Verification

Implanted gold markers and megavoltage (MV) portal imaging are commonly used for setup verification of paraspinal tumors treated with high-dose, single-fraction radiotherapy. We investigated whether the use of kilovoltage cone-beam computed tomography (CBCT) imaging eliminates the need for marker implantation. Patients with paraspinal disease who were eligible for single-fraction stereotactic body radiotherapy were accrued to an institutional review board-approved protocol. Each of 16 patients underwent implantation of fiducial markers near the target. The markers were visible on the MV images. Three MV image pairs were acquired for each patient (initial, verification, and final) and were registered to the reference images. Every MV pair was complemented by a CBCT scan. CBCT image registration was performed automatically by maximizing the mutual information using a region of interest that excluded the markers. The corrections, as determined from the MV images, were compared with these from CBCT and were used for actual patient setup. The mean and standard deviation of the absolute values of the differences between the CBCT and MV corrections were 1.0 +/- 0.7, 1.0 +/- 0.6, and 1.0 +/- 0.8 mm for the left-right, anteroposterior, and superoinferior directions, respectively. The absolute differences between the corresponding pre- and post-treatment kilovoltage CBCT image registration were 0.6 +/- 0.5, 0.6 +/- 0.5, and 1.0 +/- 0.8 mm. The setup corrections found using CBCT without the use of implanted markers were consistent with the marker registration on MV projections. CBCT has additional advantages, including better positioning precision and robust automatic three-dimensional registration, as well as eliminating the need for invasive marker implantation. We have adopted CBCT for the setup of all single-fraction paraspinal patients. Our data have also demonstrated that target displacements during treatment are insignificant.