Thomas R. Mazur

SIFT-based dense pixel tracking on 0.35 T cine-MR images acquired during image-guided radiation therapy with application to gating optimization

PURPOSE To first demonstrate the viability of applying an image processing technique for tracking regions on low-contrast cine-MR images acquired during image-guided radiation therapy, and then outline a scheme that uses tracking data for optimizing gating results in a patient-specific manner. METHODS A first-generation MR-IGRT system-treating patients since January 2014-integrates a 0.35 T MR scanner into an annular gantry consisting of three independent Co-60 sources. Obtaining adequate frame rates for capturing relevant patient motion across large fields-of-view currently requires coarse in-plane spatial resolution. This study initially (1) investigate the feasibility of rapidly tracking dense pixel correspondences across single, sagittal plane images (with both moderate signal-to-noise and spatial resolution) using a matching objective for highly descriptive vectors called scale-invariant feature transform (SIFT) descriptors associated to all pixels that describe intensity gradients in local regions around each pixel. To more accurately track features, (2) harmonic analysis was then applied to all pixel trajectories within a region-of-interest across a short training period. In particular, the procedure adjusts the motion of outlying trajectories whose relative spectral power within a frequency bandwidth consistent with respiration (or another form of periodic motion) does not exceed a threshold value that is manually specified following the training period. To evaluate the tracking reliability after applying this correction, conventional metrics-including Dice similarity coefficients (DSCs), mean tracking errors (MTEs), and Hausdorff distances (HD)-were used to compare target segmentations obtained via tracking to manually delineated segmentations. Upon confirming the viability of this descriptor-based procedure for reliably tracking features, the study (3) outlines a scheme for optimizing gating parameters-including relative target position and a tolerable margin about this position-derived from a probability density function that is constructed using tracking results obtained just prior to treatment. RESULTS The feasibility of applying the matching objective for SIFT descriptors toward pixel-by-pixel tracking on cine-MR acquisitions was first retrospectively demonstrated for 19 treatments (spanning various sites). Both with and without motion correction based on harmonic analysis, sub-pixel MTEs were obtained. A mean DSC value spanning all patients of 0.916 ± 0.001 was obtained without motion correction, with DSC values exceeding 0.85 for all patients considered. While most patients show accurate tracking without motion correction, harmonic analysis does yield substantial gain in accuracy (defined using HDs) for three particularly challenging subjects. An application of tracking toward a gating optimization procedure was then demonstrated that should allow a physician to balance beam-on time and tissue sparing in a patient-specific manner by tuning several intuitive parameters. CONCLUSIONS Tracking results show high fidelity in assessing intrafractional motion observed on cine-MR acquisitions. Incorporating harmonic analysis during a training period improves the robustness of the tracking for challenging targets. The concomitant gating optimization procedure should allow for physicians to quantitatively assess gating effectiveness quickly just prior to treatment in a patient-specific manner.

An integrated model-driven method for in-treatment upper airway motion tracking using cine MRI in head and neck radiation therapy.

PURPOSE For the first time, MRI-guided radiation therapy systems can acquire cine images to dynamically monitor in-treatment internal organ motion. However, the complex head and neck (H&N) structures and low-contrast/resolution of on-board cine MRI images make automatic motion tracking a very challenging task. In this study, the authors proposed an integrated model-driven method to automatically track the in-treatment motion of the H&N upper airway, a complex and highly deformable region wherein internal motion often occurs in an either voluntary or involuntary manner, from cine MRI images for the analysis of H&N motion patterns. METHODS Considering the complex H&N structures and ensuring automatic and robust upper airway motion tracking, the authors firstly built a set of linked statistical shapes (including face, face-jaw, and face-jaw-palate) using principal component analysis from clinically approved contours delineated on a set of training data. The linked statistical shapes integrate explicit landmarks and implicit shape representation. Then, a hierarchical model-fitting algorithm was developed to align the linked shapes on the first image frame of a to-be-tracked cine sequence and to localize the upper airway region. Finally, a multifeature level set contour propagation scheme was performed to identify the upper airway shape change, frame-by-frame, on the entire image sequence. The multifeature fitting energy, including the information of intensity variations, edge saliency, curve geometry, and temporal shape continuity, was minimized to capture the details of moving airway boundaries. Sagittal cine MR image sequences acquired from three H&N cancer patients were utilized to demonstrate the performance of the proposed motion tracking method. RESULTS The tracking accuracy was validated by comparing the results to the average of two manual delineations in 50 randomly selected cine image frames from each patient. The resulting average dice similarity coefficient (93.28%  ±  1.46%) and margin error (0.49  ±  0.12 mm) showed good agreement between the automatic and manual results. The comparison with three other deformable model-based segmentation methods illustrated the superior shape tracking performance of the proposed method. Large interpatient variations of swallowing frequency, swallowing duration, and upper airway cross-sectional area were observed from the testing cine image sequences. CONCLUSIONS The proposed motion tracking method can provide accurate upper airway motion tracking results, and enable automatic and quantitative identification and analysis of in-treatment H&N upper airway motion. By integrating explicit and implicit linked-shape representations within a hierarchical model-fitting process, the proposed tracking method can process complex H&N structures and low-contrast/resolution cine MRI images. Future research will focus on the improvement of method reliability, patient motion pattern analysis for providing more information on patient-specific prediction of structure displacements, and motion effects on dosimetry for better H&N motion management in radiation therapy.

A GPU-accelerated Monte Carlo dose calculation platform and its application toward validating an MRI-guided radiation therapy beam model.

PURPOSE The clinical commissioning of IMRT subject to a magnetic field is challenging. The purpose of this work is to develop a GPU-accelerated Monte Carlo dose calculation platform based on penelope and then use the platform to validate a vendor-provided MRIdian head model toward quality assurance of clinical IMRT treatment plans subject to a 0.35 T magnetic field. METHODS penelope was first translated from fortran to c++ and the result was confirmed to produce equivalent results to the original code. The c++ code was then adapted to cuda in a workflow optimized for GPU architecture. The original code was expanded to include voxelized transport with Woodcock tracking, faster electron/positron propagation in a magnetic field, and several features that make gpenelope highly user-friendly. Moreover, the vendor-provided MRIdian head model was incorporated into the code in an effort to apply gpenelope as both an accurate and rapid dose validation system. A set of experimental measurements were performed on the MRIdian system to examine the accuracy of both the head model and gpenelope. Ultimately, gpenelope was applied toward independent validation of patient doses calculated by MRIdians kmc. RESULTS An acceleration factor of 152 was achieved in comparison to the original single-thread fortran implementation with the original accuracy being preserved. For 16 treatment plans including stomach (4), lung (2), liver (3), adrenal gland (2), pancreas (2), spleen(1), mediastinum (1), and breast (1), the MRIdian dose calculation engine agrees with gpenelope with a mean gamma passing rate of 99.1% ± 0.6% (2%/2 mm). CONCLUSIONS A Monte Carlo simulation platform was developed based on a GPU- accelerated version of penelope. This platform was used to validate that both the vendor-provided head model and fast Monte Carlo engine used by the MRIdian system are accurate in modeling radiation transport in a patient using 2%/2 mm gamma criteria. Future applications of this platform will include dose validation and accumulation, IMRT optimization, and dosimetry system modeling for next generation MR-IGRT systems.

Development of a fast Monte Carlo dose calculation system for online adaptive radiation therapy quality assurance

Online adaptive radiation therapy (ART) based on real-time magnetic resonance imaging represents a paradigm-changing treatment scheme. However, conventional quality assurance (QA) methods based on phantom measurements are not feasible with the patient on the treatment couch. The purpose of this work is to develop a fast Monte Carlo system for validating online re-optimized tri-60Co IMRT adaptive plans with both high accuracy and speed. The Monte Carlo system is based on dose planning method (DPM) code with further simplification of electron transport and consideration of external magnetic fields. A vendor-provided head model was incorporated into the code. Both GPU acceleration and variance reduction were implemented. Additionally, to facilitate real-time decision support, a C++ GUI was developed for visualizing 3D dose distributions and performing various analyses in an online adaptive setting. A thoroughly validated Monte Carlo code (gPENELOPE) was used to benchmark the new system, named GPU-accelerated DPM with variance reduction (gDPMvr). The comparison using 15 clinical IMRT plans demonstrated that gDPMvr typically runs 43 times faster with only 0.5% loss in accuracy. Moreover, gDPMvr reached 1% local dose uncertainty within 2.3 min on average, and thus is well-suited for ART QA.

SU-F-T-275: A Correlation Study On 3D Fluence-Based QA and 2D Dose Measurement-Based QA

PURPOSE The aim of this paper was to demonstrate the feasibility and creditability of computing and verifying 3D fluencies to assure IMRT and VMAT treatment deliveries, by correlating the passing rates of the 3D fluence-based QA (P(ά)) to the passing rates of 2D dose measurementbased QA (P(Dm)). METHODS 3D volumetric primary fluencies are calculated by forward-projecting the beam apertures and modulated by beam MU values at all gantry angles. We first introduce simulated machine parameter errors (MU, MLC positions, jaw, gantry and collimator) to the plan. Using passing rates of voxel intensity differences (P(Ir)) and 3D gamma analysis (P(γ)), calculated 3D fluencies, calculated 3D delivered dose, and measured 2D planar dose in phantom from the original plan are then compared with those from corresponding plans with errors, respectively. The correlations of these three groups of resultant passing rates, i.e. 3D fluence-based QA (P(ά,Ir) and P(ά,γ)), calculated 3D dose (P(Dc,Ir) and P(Dc,γ)), and 2D dose measurement-based QA (P(Dm,Ir) and P(Dm,γ)), will be investigated. RESULTS 20 treatment plans with 5 different types of errors were tested. Spearmans correlations were found between P(ά,Ir) and P(Dc,Ir), and also between P(ά,γ) and P(Dc,γ), with averaged p-value 0.037, 0.065, and averaged correlation coefficient ρ-value 0.942, 0.871 respectively. Using Matrixx QA for IMRT plans, Spearmans correlations were also obtained between P(ά,Ir) and P(Dm,Ir) and also between P(ά,γ) and P(Dm,γ), with p-value being 0.048, 0.071 and ρ-value being 0.897, 0.779 respectively. CONCLUSION The demonstrated correlations improve the creditability of using 3D fluence-based QA for assuring treatment deliveries for IMRT/VMAT plans. Together with advantages of high detection sensitivity and better visualization of machine parameter errors, this study further demonstrates the accuracy and feasibility of 3D fluence based-QA in pre-treatment QA and daily QA. Research reported in this study is supported by the Agency for Healthcare Research and Quality (AHRQ) under award 1R01HS0222888. The senior author received research grants from ViewRay Inc. and Varian Medical System.

Quantitative FDG-PET/CT predicts local recurrence and survival for squamous cell carcinoma of the anus

Purpose 18F-fluorodeoxyglucose (FDG) positron emission tomography–(PET)/computed tomography (CT) imaging is used for staging and treatment planning of patients with anal cancer. Quantitative pre- and posttreatment metrics that are predictive of recurrence are unknown. We evaluated the association between pre- and posttreatment FDG-PET/CT parameters and outcomes for patients with squamous cell carcinoma of the anus (SCCA). Methods and materials The records of 110 patients treated between 2003 and 2013 with definitive radiation therapy for SCCA were reviewed under an institutional review board–approved protocol. The median radiation therapy dose was 50.4 Gy (range, 35-60 Gy). Concurrent chemotherapy was administered for 109 of 110 patients and generally consisted of 5-fluorouracil and mitomycin C (n = 94). All patients underwent pretreatment FDG-PET/CT and 101 of 110 underwent posttreatment FDG-PET/CT 3 months after completion of radiation therapy. The maximum standard uptake value (SUVmax) was analyzed, in addition to multiple patient and treatment factors, by univariate and multivariate Cox regression for correlation with local recurrence (LR) and overall survival (OS). Results The median follow-up was 28.6 months. LR occurred in 1 of 15 (6.7%), 5 of 47 (10.6%), and 6 of 48 (12.5%) patients with stage I, II, and III disease, respectively. On univariate analysis, a significant association was observed between reduced LR and posttreatment SUVmax <6.1 (P = .0095) and between increased OS and posttreatment SUVmax <6.1 (P = .0086). On multivariate analysis, a significant association was observed between reduced LR and posttreatment SUVmax <6.1 (P = .0013) and the use of intensity modulated radiation therapy (P < .001). A significant multivariate association was observed between increased OS and posttreatment SUVmax <6.1 (P = .0373) and the use of 5-fluorouracil/mitomycin C chemotherapy (P = .001). Conclusion Posttreatment SUVmax <6.1 is associated with reduced LR and increased OS after chemoradiation therapy for SCCA independent of T and N stage on multivariate analysis. Greater follow-up is required to confirm this association with late patterns of failure.

A Method to Recognize Anatomical Site and Image Acquisition View in X-ray Images

A method was developed to recognize anatomical site and image acquisition view automatically in 2D X-ray images that are used in image-guided radiation therapy. The purpose is to enable site and view dependent automation and optimization in the image processing tasks including 2D-2D image registration, 2D image contrast enhancement, and independent treatment site confirmation. The X-ray images for 180 patients of six disease sites (the brain, head-neck, breast, lung, abdomen, and pelvis) were included in this study with 30 patients each site and two images of orthogonal views each patient. A hierarchical multiclass recognition model was developed to recognize general site first and then specific site. Each node of the hierarchical model recognized the images using a feature extraction step based on principal component analysis followed by a binary classification step based on support vector machine. Given two images in known orthogonal views, the site recognition model achieved a 99% average F1 score across the six sites. If the views were unknown in the images, the average F1 score was 97%. If only one image was taken either with or without view information, the average F1 score was 94%. The accuracy of the site-specific view recognition models was 100%.

SU-F-303-11: Implementation and Applications of Rapid, SIFT-Based Cine MR Image Binning and Region Tracking

Purpose: To develop a novel and rapid, SIFT-based algorithm for assessing feature motion on cine MR images acquired during MRI-guided radiotherapy treatments. In particular, we apply SIFT descriptors toward both partitioning cine images into respiratory states and tracking regions across frames. Methods: Among a training set of images acquired during a fraction, we densely assign SIFT descriptors to pixels within the images. We cluster these descriptors across all frames in order to produce a dictionary of trackable features. Associating the best-matching descriptors at every frame among the training images to these features, we construct motion traces for the features. We use these traces to define respiratory bins for sorting images in order to facilitate robust pixel-by-pixel tracking. Instead of applying conventional methods for identifying pixel correspondences across frames we utilize a recently-developed algorithm that derives correspondences via a matching objective for SIFT descriptors. Results: We apply these methods to a collection of lung, abdominal, and breast patients. We evaluate the procedure for respiratory binning using target sites exhibiting high-amplitude motion among 20 lung and abdominal patients. In particular, we investigate whether these methods yield minimal variation between images within a bin by perturbing the resulting image distributions among bins. Moreover, we compare the motion between averaged images across respiratory states to 4DCT data for these patients. We evaluate the algorithm for obtaining pixel correspondences between frames by tracking contours among a set of breast patients. As an initial case, we track easily-identifiable edges of lumpectomy cavities that show minimal motion over treatment. Conclusions: These SIFT-based methods reliably extract motion information from cine MR images acquired during patient treatments. While we performed our analysis retrospectively, the algorithm lends itself to prospective motion assessment. Applications of these methods include motion assessment, identifying treatment windows for gating, and determining optimal margins for treatment.

A deep Boltzmann machine-driven level set method for heart motion tracking using cine MRI images

HighlightsThe DBM needs small‐sized data set to train, but imposes strong modeling ability.A three‐layer DBM can capture both local and global properties of heart contours.An efficient layer‐wise block‐Gibbs sampling is used to infer heart shape priors.The DBM‐induced heart shape priors are used as constraints of DRLSE evolution. Graphical abstract Figure. No caption available. ABSTRACT Heart motion tracking for radiation therapy treatment planning can result in effective motion management strategies to minimize radiation‐induced cardiotoxicity. However, automatic heart motion tracking is challenging due to factors that include the complex spatial relationship between the heart and its neighboring structures, dynamic changes in heart shape, and limited image contrast, resolution, and volume coverage. In this study, we developed and evaluated a deep generative shape model‐driven level set method to address these challenges. The proposed heart motion tracking method makes use of a heart shape model that characterizes the statistical variations in heart shapes present in a training data set. This heart shape model was established by training a three‐layered deep Boltzmann machine (DBM) in order to characterize both local and global heart shape variations. During the tracking phase, a distance regularized level‐set evolution (DRLSE) method was applied to delineate the heart contour on each frame of a cine MRI image sequence. The trained shape model was embedded into the DRLSE method as a shape prior term to constrain an evolutional shape to reach the desired heart boundary. Frame‐by‐frame heart motion tracking was achieved by iteratively mapping the obtained heart contour for each frame to the next frame as a reliable initialization, and performing a level‐set evolution. The performance of the proposed motion tracking method was demonstrated using thirty‐eight coronal cine MRI image sequences.

Magnetic resonance image guided radiation therapy for primary splenic diffuse large B-cell lymphoma: A teaching case.

Malignant neoplasms of the spleen are rare and most commonly non-Hodgkin lymphomas.1 The most frequent primary non-Hodgkin lymphoma of the spleen is a marginal zone lymphoma, which accounts for approximately 80% of cases, whereas primary splenic diffuse largeB-cell lymphoma (PS-DLBCL) is much more rare.2 Given the paucity of data regarding PS-DLBCL, optimal workup and management strategies are not well defined.2 Extrapolated from the standard of care from other extranodal sites of DLBCL, patients with limited-stage PS-DLBCL can be treated with combined modality therapy consisting of systemic chemotherapy with rituximab followed by involved site radiation therapy.3 The safety and efficacy of involved-site radiation therapy is heavily dependent on accurate image guidance that is most commonly accomplished using on-board cone beam computed tomography (CBCT). However, treatment of splenic tumors poses a significant challenge for CBCTguided RT because of its significant respiratory motion and poor soft-tissue resolution. We describe, to the best of our knowledge, the first reported case of a patient treated with