Deformable Image Registration Uncertainty Research Articles

Uncertainty estimation and evaluation of deformation image registration based convolutional neural networks

Objective. Fast and accurate deformable image registration (DIR), including DIR uncertainty estimation, is essential for safe and reliable clinical deployment. While recent deep learning models have shown promise in predicting DIR with its uncertainty, challenges persist in proper uncertainty evaluation and hyperparameter optimization for these methods. This work aims to develop and evaluate a model that can perform fast DIR and predict its uncertainty in seconds. Approach. This study introduces a novel probabilistic multi-resolution image registration model utilizing convolutional neural networks to estimate a multivariate normal distributed dense displacement field (DDF) in a multimodal image registration problem. To assess the quality of the DDF distribution predicted by the model, we propose a new metric based on the Kullback–Leibler divergence. The performance of our approach was evaluated against three other DIR algorithms (VoxelMorph, Monte Carlo dropout, and Monte Carlo B-spline) capable of predicting uncertainty. The evaluation of the models included not only the quality of the deformation but also the reliability of the estimated uncertainty. Our application investigated the registration of a treatment planning computed tomography (CT) to follow-up cone beam CT for daily adaptive radiotherapy. Main results. The hyperparameter tuning of the models showed a trade-off between the estimated uncertainty’s reliability and the deformation’s accuracy. In the optimal trade-off, our model excelled in contour propagation and uncertainty estimation (p <0.05) compared to existing uncertainty estimation models. We obtained an average dice similarity coefficient of 0.89 and a KL-divergence of 0.15. Significance. By addressing challenges in DIR uncertainty estimation and evaluation, our work showed that both the DIR and its uncertainty can be reliably predicted, paving the way for safe deployment in a clinical environment.

Evaluation and mitigation of deformable image registration uncertainties for MRI-guided adaptive radiotherapy.

We evaluate the performance of a deformable image registration (DIR) software package in registering abdominal magnetic resonance images (MRIs) and then develop a mechanical modeling method to mitigate detected DIR uncertainties. Three evaluation metrics, namely mean displacement to agreement (MDA), DICE similarity coefficient (DSC), and standard deviation of Jacobian determinants (STD-JD), are used to assess the multi-modality (MM), contour-consistency (CC), and image-intensity (II)-based DIR algorithms in the MIM software package, as well as an in-house developed, contour matching-based finite element method (CM-FEM). Furthermore, we develop a hybrid FEM registration technique to modify the displacement vector field of each MIM registration. The MIM and FEM registrations were evaluated on MRIs obtained from 10 abdominal cancer patients. One-tailed Wilcoxon-Mann-Whitney (WMW) tests were conducted to compare the MIM registrations with their FEM modifications. For the registrations performed with the MIM-CC, MIM-MM, MIM-II, and CM-FEM algorithms, their average MDAs are 0.62±0.27, 2.39±1.30, 3.07±2.42, 1.04±0.72mm, and average DSCs are 0.94±0.03, 0.80±0.12, 0.77±0.15, 0.90±0.11, respectively. The p-values of the WMW tests between the MIM registrations and their FEM modifications are less than 0.0084 for STD-JDs and greater than 0.87 for MDA and DSC. Among the three MIM DIR algorithms, MIM-CC shows the smallest errors in terms of MDA and DSC but exhibits significant Jacobian uncertainties in the interior regions of abdominal organs. The hybrid FEM technique effectively mitigates the Jacobian uncertainties in these regions.

Deep learning based uncertainty prediction of deformable image registration for contour propagation and dose accumulation in online adaptive radiotherapy

Objective. Online adaptive radiotherapy aims to fully leverage the advantages of highly conformal therapy by reducing anatomical and set-up uncertainty, thereby alleviating the need for robust treatments. This requires extensive automation, among which is the use of deformable image registration (DIR) for contour propagation and dose accumulation. However, inconsistencies in DIR solutions between different algorithms have caused distrust, hampering its direct clinical use. This work aims to enable the clinical use of DIR by developing deep learning methods to predict DIR uncertainty and propagating it into clinically usable metrics. Approach. Supervised and unsupervised neural networks were trained to predict the Gaussian uncertainty of a given deformable vector field (DVF). Since both methods rely on different assumptions, their predictions differ and were further merged into a combined model. The resulting normally distributed DVFs can be directly sampled to propagate the uncertainty into contour and accumulated dose uncertainty. Main results. The unsupervised and combined models can accurately predict the uncertainty in the manually annotated landmarks on the DIRLAB dataset. Furthermore, for 5 patients with lung cancer, the propagation of the predicted DVF uncertainty into contour uncertainty yielded for both methods an expected calibration error of less than 3%. Additionally, the probabilisticly accumulated dose volume histograms (DVH) encompass well the accumulated proton therapy doses using 5 different DIR algorithms. It was additionally shown that the unsupervised model can be used for different DIR algorithms without the need for retraining. Significance. Our work presents first-of-a-kind deep learning methods to predict the uncertainty of the DIR process. The methods are fast, yield high-quality uncertainty estimates and are useable for different algorithms and applications. This allows clinics to use DIR uncertainty in their workflows without the need to change their DIR implementation.

Development of a multi-layer quality assurance program to evaluate the uncertainty of deformable dose accumulation in adaptive radiotherapy.

Deformable dose accumulation (DDA) has uncertainties which impede the implementation of DDA-based adaptive radiotherapy (ART) in clinic. The purpose of this study is to develop a multi-layer quality assurance (MLQA) program to evaluate uncertainties in DDA. A computer program is developed to generate a pseudo-inverse displacement vector field (DVF) for each deformable image registration (DIR) performed in Accuray's PreciseART. The pseudo-inverse DVF is first used to calculate a pseudo-inverse consistency error (PICE) and then implemented in an energy and mass congruent mapping (EMCM) method to reconstruct a deformed dose. The PICE is taken as a metric to estimate DIR uncertainties. A pseudo-inverse dose agreement rate (PIDAR) is used to evaluate the consequence of the DIR uncertainties in DDA and the principle of energy conservation is used to validate the integrity of dose mappings. The developed MLQA program was tested using the data collected from five representative cancer patients treated with tomotherapy. DIRs were performed in PreciseART to generate primary DVFs for the five patients. The fidelity index and PICE of these DVFs on average are equal to 0.028mm and 0.169mm, respectively. With the criteria of 3mm/3% and 5mm/5%, the PIDARs of the PreciseART-reconstructed doses are 73.9±4.4% and 87.2±3.3%, respectively. The PreciseART and EMCM-based dose reconstructions have their deposited energy changed by 5.6±3.9% and 2.6±1.5% in five GTVs, and by 9.2±7.8% and 4.7±3.6% in 30 OARs, respectively. A pseudo-inverse map-based EMCM program has been developed to evaluate DIR and dose mapping uncertainties. This program could also be used as a sanity check tool for DDA-based ART.

Evaluation of Repeated Lung Stereotactic Body Radiotherapy Using a Radiotherapy Dose Accumulation Routine

<h3>Purpose/Objective(s)</h3> Repeated lung SBRT has been used for recurrent or metachronous non-small-cell lung cancer. Comprehensive evaluation of cumulative dose is essential for safe reirradiation to minimize the risk of pneumonitis and esophageal toxicity. We evaluated an in-house script named RADAR (Radiotherapy Dose Accumulation Routine) designed to provide accurate dose assessment that accounts for registration uncertainties, and improve the efficiency of clinical workflow. <h3>Materials/Methods</h3> In this pilot study, we analyzed a cohort of 20 patients who underwent 2-5 courses of repeated lung SBRT. Lung volume changed 8±8% between courses. Planning CT images from earlier treatment courses were registered onto the final course in two steps: first, a rigid registration based on alignment of spine and rib cage, then a deformable image registration (DIR) using a ‘demons' algorithm in a planning system. The resulting deformable vector fields were used by RADAR to transform voxels for automated dose accumulation. User can choose to output either physical dose or equivalent dose in 2Gy fractions (EQD2, α/β=3) with a discount factor (Paradis, ARO 2019). RADAR provides three approaches for dose accumulation: 1) direct sum that outputs voxel level dose summation; 2) unrestricted search that incorporates DIR uncertainty by conservatively assigning the maximum dose within a search radius (3mm) to the voxel of interest for dose accumulation; 3) restricted search, which performs the same dose accumulation as in 2) but imposes structure label matching in the search radius, therefore including only voxels within the same structure. Accumulated dose distribution was written back to the planning system for evaluation. We extracted mean lung dose (MLD) and esophagus D5cc (EsoD5cc) with respect to various accumulation approaches for comparison. <h3>Results</h3> MLD and EsoD5cc were significantly higher when accounting for DIR uncertainties (<i>p</i><0.001 two-tailed t-test). MLD has less uncertainty than EsoD5cc due to its averaging nature. For radiotherapy with >2 courses, RADAR considerably shortened the processing time by auto-accumulating dose from multiple courses simultaneously, as opposed to the cumbersome manual approach of creating plan sums iteratively. Typical time for dose accumulation and evaluation using RADAR is < 10 mins. <h3>Conclusion</h3> RADAR is a convenient and efficient tool to calculate and evaluate cumulative dose for lung reirradiation cases. It identifies higher OAR doses due to dose accumulation uncertainties, for example coming from the registration. Our study provides initial data to establish clinical guidance for design and optimization of repeated lung SBRT.

PD-0400 Quantitative deformable image registration uncertainty framework for proton head and neck treatment

PD-0400 Quantitative deformable image registration uncertainty framework for proton head and neck treatment

A method using 4D dose accumulation to quantify the interplay effect in lung stereotactic body radiation therapy

The purpose of this study was to devise and evaluate a method to quantify the dosimetric uncertainty produced by the interplay between the movement of multileaf collimator and respiratory motion in lung stereotactic body radiation therapy. The method calculates the dose distribution for all control points from a dynamic treatment in all respiratory phases. The methodology includes some characteristics of a patient’s irregular breathing patterns. It selects, for each control point, the phases with maximum and minimum mean dose over the tumor and their corresponding adjacent phases, whenever necessary. According to this selection, the dose matrices from each control point are summed up to obtain two dose distributions in each phase, which are accumulated in the reference phase subsequently by deformable image registration (DIR). D 95 and were calculated over those accumulated dose distributions for Gross Tumor Volume (GTV), Planning Target Volume—based on Internal Target Volume approach—and Evaluation Target Volume (ETV), a novel concept that applies to 4D dose accumulation. With the ETV, DIR and interplay uncertainties are separated. The methodology also evaluated how variations in the breathing rate and field size affects the mean dose received by the GTV. The method was applied retrospectively in five patients treated with intensity modulated radiotherapy—minimum area defined by the leaves configuration at any control point was at least 4 cm2. Uncertainties in tumor coverage were small (in most patients, changes on D 95 and were below 2% for GTV and ETV) but significant over- and under-dosages near ETV, which can be accentuated by highly irregular breathing. Uncertainties in mean dose for GTV tended to decrease exponentially with increasing field size and were reduced by an increase of breathing rate. The implementation of this method would be helpful to assess treatment quality in patients with irregular breathing. Furthermore, it could be used to study interplay uncertainties when small field sizes are used.

Accuracy of Deformable Image Registration for Reirradiation of Lung Cancer Following Stereotactic Body Radiotherapy

Stereotactic body radiotherapy (SBRT) for lung cancer typically causes radiographic lung changes that can affect the registration accuracy at the time of reirradiation. In this study, we investigate the accuracy of deformable image registration (DIR) in the presence of post SBRT lung changes using anatomical landmarks. We hypothesize that DIR accuracy is worse in regions near radiographic changes than in normal lung tissue. Planning, 3-, 6-, and 12-month follow-up CT images were deformably registered for 10 patients exhibiting radiographic lung changes following SBRT. Several multi-pass b-spline based DIRs were performed using a commercially available algorithm by varying the target intensity window and the structures with overridden image intensities. The structures included: the planning target volume (PTV) and ground glass (GG), and consolidated (C) radiographic changes, while the intensity windows investigated included: full image contrast, lung window (-1024HU to 200 HU), and mediastinum window (-135HU to 215HU). Landmark points were manually located in both the treated lung (mean 26.5 points, range 17-30 per scan within 5cm of the PTV) and contralateral normal lung (mean 29.7 points, range 29-30). Accuracy was evaluated by calculating the average 3D landmark registration error. The Pearson correlation coefficient was used to investigate the influence of the PTV, GG, and C lung volumes and normal lung registration error on the treated region registration error. For all time points considered together, the most accurate registration near treated tissue involved masking both the PTV and GG lung changes and using full image contrast. The average landmark error near the treated region with this registration was 7.7 +/- 4.9 mm and 5.2 +/- 3.5mm in the normal lung. The landmark error for the individual time points, 3-, 6-, and 12-month respectively, were 3.7mm, 9.8mm, and 9.8mm near the treated region and 4.8mm, 5.7 and 5mm in the normal lung tissue. Landmark errors at 3-months were strongly correlated with the volume of the GG (rˆ2 = 0.84) and C (rˆ2 = 0.84) lung changes. The 6- and 12-month landmark errors were moderately correlated with GG (rˆ2 = 0.29, rˆ2 = 0.43) and C (rˆ2 = 0.31, rˆ2 = 0.47) volumes. There was moderate correlation between landmark errors near the treated region and normal tissue that increases over time from rˆ2 = 0.28 at 3 months to rˆ2 = 0.69 at 12 months. No strong correlation was found with the PTV volume. Landmark registration errors in the treated region can be a few millimeters larger than in normal tissue and appear to increase with time, before stabilizing. There is also an increasing correlation over time between the normal tissue registration errors and those in the treated region. DIR uncertainty in the treated and normal lung needs to be considered during retreatment composite planning.

Effect of uncertainties in quantitative 18 F-FDG PET/CT imaging feedback for intratumoral dose-response assessment and dose painting by number.

Intratumoral dose response can be detected using serial fluoro-2-deoxyglucose-(FDG) positron emission tomography (PET)/computed tomography (CT) imaging feedback during treatment and used to guide adaptive dose painting by number (DPbN). However, to reliably implement this technique, the effect of uncertainties in quantitative PET/CT imaging feedback on tumor voxel dose-response assessment and DPbN needs to be determined and reduced. Three major uncertainties, induced by (a) PET imaging partial volume effect (PVE) and (b) tumor deformable image registration (DIR), and (c) variation of the time interval between FDG injection and PET image acquisition (TI), were determined using serial FDG-PET/CT images acquired during chemoradiotherapy of 18 head and neck cancer patients. PET imaging PVE was simulated using the discrepancy between with and without iterative deconvolution-based PVE corrections. Effect of tumor DIR uncertainty was simulated using the discrepancy between two DIR algorithms, including one with and one without soft-tissue mechanical correction for the voxel displacement. The effect of TI variation was simulated using linear interpolation on the dual-point PET/CT images. Tumor voxel pretreatment metabolic activity (SUV0 ) and dose-response matrix (DRM) discrepancies induced by each of the three uncertainties were quantified, respectively. Adverse effects of tumor voxel SUV0 and DRM discrepancies on tumor control probability (TCP) in DPbN were assessed. Partial volume effect and TI variations of 10 mins induced a mean±standard deviation (SD) of tumor voxel SUV0 discrepancies to be -0.7%±9.2% and 0%±4.8%, respectively. Tumor voxel DRM discrepancies induced by PVE, tumor DIR discrepancy, and TI variations were 0.6%±8.9%, 1.7%±9.1%, and 0%±7%, respectively. Partial volume effect induced SUV0 and DRM discrepancies correlated significantly with the tumor shape and FDG uptake heterogeneity. Tumor DIR uncertainty-induced DRM discrepancy correlated significantly with the tumor volume and shrinkage during treatment. Among the three uncertainties, PVE dominated the adverse effects on the TCP, with a mean±SD of TCP reduction to be 12.7%±9.8% for all tumors if no compensation was applied for. Effect of uncertainties in quantitative FDG-PET/CT imaging feedback on intratumoral dose-response quantification was not negligible. These uncertainties primarily caused by PVE and tumor DIR were highly dependent on individual tumor shape, volume, shrinkage during treatment, and pretreatment SUV heterogeneity, which can be managed individually. The adverse effects of these uncertainties could be minimized by using proper PVE corrections and DIR methods and compensated for in the clinical implementation of DPbN.

Deformable image registration uncertainty for inter-fractional dose accumulation of lung cancer proton therapy

Background and purposeNon-small cell lung cancer (NSCLC) patients show typically large anatomical changes during treatment, making recalculation or adaption necessary. For report and review, the applied treatment dose can be accumulated on the reference planning CT using deformable image registration (DIR). We investigated the dosimetric impact of using six different clinically available DIR algorithms for dose accumulation in presence of inter-fractional anatomy variations. Materials and methodsFor seven NSCLC patients, proton treatment plans with 66 Gy-RBE to the planning target volume (PTV) were optimised. Nine repeated CTs were registered to the planning CT using six DIR algorithms each. All CTs were acquired in visually guided deep-inspiration breath-hold. The plans were recalculated on the repeated CTs and warped back to the planning CT using the corresponding DIRs. Fraction doses warped with the same DIR were summed up to six different accumulated dose distributions per patient, and compared to the initial dose. ResultsThe PTV-V95 of accumulated doses decreased by 16% on average over all patients, with variations due to DIR selection of 8.7%. A separation of the dose effects caused by anatomical changes and DIR uncertainty showed a good agreement between the dose degradation caused by anatomical changes and the dose predicted from the average of all DIRs (differences of only 1.6%). ConclusionThe dose degradation caused by anatomical changes was more pronounced than the uncertainty of employing different DIRs for dose accumulation, with averaged results from several DIRs providing a good representation of dose degradation caused by anatomy. However, accumulated dose variations between DIRs can be substantial, leading to an additional dose uncertainty.

"Patient-specific validation of deformable image registration in radiation therapy: Overview and caveats".

Over the last few decades, deformable image registration (DIR) has gained popularity in image-guided radiation therapy for a number of applications, such as contour propagation, dose warping, and accumulation. Although this raises promising perspectives for the improvement of treatment outcomes and quality of radiotherapy clinical practice, the variety of proposed DIR algorithms, combined with the lack of an effective quantitative quality control metric of the registration, is slowing the transfer of DIR into the clinical routine. Recently, a task group (AAPM TG132) report was published outlining the essential aspects of DIR for image guidance in radiotherapy. However, an accurate and efficient patient-specific validation is not yet defined, and appropriate metrics should be identified to achieve the definition of both geometric and dosimetric accuracy. In this respect, the use of a dense set of anatomical landmarks, along with additional evaluations on contours or deformation field analysis, are likely to drive patient-specific DIR validation in clinical image-guided radiotherapy applications to account for geometric inaccuracies. Automatic and efficient strategies able to provide spatial information of DIR uncertainties and to evaluate monomodal and multimodal image registration, as well as to describe homogenous and un-contrasted regions are believed to represent the future direction in DIR validation. But especially in the case of DIR applications for dose mapping and accumulation, the need of accurate patient-specific validation is not only limited to the evaluation of geometric accuracy. In fact, the need to account for dosimetric inaccuracies due to DIR represents another important area in the field of adaptive treatments. Different approaches are currently being investigated to quantify the effect of DIR error on dose analysis, mainly relying on clinically relevant dose metrics, or on the study of deformation field properties for a voxel-by-voxel evaluation. However, novel research is required for the definition of dedicated and personalized measures capable to relate the geometric and dosimetric inaccuracies, thus bearing useful information for a safe use of DIR by clinical end users. In this paper we provide insights on DIR results evaluation on a patient-specific basis, facing the issues of both geometric and dosimetric paradigms. Challenges on DIR validation are overviewed and discussed, in order to push preliminary clinical guidelines forward on this fundamental topic and boost the implementation of more robust and reliable patient-specific evaluation metrics.

An automated, quantitative, and case-specific evaluation of deformable image registration in computed tomography images

A prerequisite for adaptive dose-tracking in radiotherapy is the assessment of the deformable image registration (DIR) quality. In this work, various metrics that quantify DIR uncertainties are investigated using realistic deformation fields of 26 head and neck and 12 lung cancer patients. Metrics related to the physiologically feasibility (the Jacobian determinant, harmonic energy (HE), and octahedral shear strain (OSS)) and numerically robustness of the deformation (the inverse consistency error (ICE), transitivity error (TE), and distance discordance metric (DDM)) were investigated. The deformable registrations were performed using a B-spline transformation model. The DIR error metrics were log-transformed and correlated (Pearson) against the log-transformed ground-truth error on a voxel level. Correlations of r ⩾ 0.5 were found for the DDM and HE. Given a DIR tolerance threshold of 2.0 mm and a negative predictive value of 0.90, the DDM and HE thresholds were 0.49 mm and 0.014, respectively. In conclusion, the log-transformed DDM and HE can be used to identify voxels at risk for large DIR errors with a large negative predictive value. The HE and/or DDM can therefore be used to perform automated quality assurance of each CT-based DIR for head and neck and lung cancer patients.

Effect of DIR uncertainty on prostate passive-scattering proton therapy dose accumulation

Deformable image registration (DIR) is important in dose accumulation. Currently, the impact of DIR-algorithm-associated uncertainties in proton therapy is unclear. Here, we quantify the effect of DIR uncertainties on prostate passive-scattering proton therapy (PSPT) dose accumulation. Ten patients with an intermediate risk for prostate cancer formerly treated by PSPT (PTV D95=78GyE) were studied. Dose distributions from all verification CT images (five images per patient) were warped and accumulated in the planning CT geometries with DIR. The dose-volume histogram parameters (Dmean, V40, and V70) for rectum and bladder were calculated. Two commercially available DIR software packages were employed: Velocity AI (Varian Medical Systems) and RayStation (RaySearch Laboratories). The dice similarity coefficient (DSC) and surface distance, which were calculated between planning CT contours and deformed contours, were used for DIR validation, with the relationship between the dose parameter and DIR uncertainty ultimately investigated. On average, when using RayStation, the DSC increased by 0.14 and surface distance decreased by 6.4mm, as compared to Velocity. For Dmean, V40, and V70 to the rectum, the average differences between the RayStation and Velocity were 3.9GyE, 5.5%, and 3.2%, respectively. For the bladder, the differences were 5.2GyE, 5.8%, and 5.4%, respectively. The maximum differences in V40 between RayStation and Velocity were 14.4% and 22.8% for the rectum and bladder, respectively, when the average DSC and surface distance differences were more than 0.14 and 6.4mm, respectively. Such results suggest that DIR uncertainties might significantly affect prostate PSPT dose accumulations.

A multiple-image-based method to evaluate the performance of deformable image registration in the pelvis

Deformable image registration (DIR) is essential for adaptive radiotherapy (RT) for tumor sites subject to motion, changes in tumor volume, as well as changes in patient normal anatomy due to weight loss. Several methods have been published to evaluate DIR-related uncertainties but they are not widely adopted. The aim of this study was, therefore, to evaluate intra-patient DIR for two highly deformable organs—the bladder and the rectum—in prostate cancer RT using a quantitative metric based on multiple image registration, the distance discordance metric (DDM). Voxel-by-voxel DIR uncertainties of the bladder and rectum were evaluated using DDM on weekly CT scans of 38 subjects previously treated with RT for prostate cancer (six scans/subject). The DDM was obtained from group-wise B-spline registration of each patient’s collection of repeat CT scans. For each structure, registration uncertainties were derived from DDM-related metrics. In addition, five other quantitative measures, including inverse consistency error (ICE), transitivity error (TE), Dice similarity (DSC) and volume ratios between corresponding structures from pre- and post- registered images were computed and compared with the DDM. The DDM varied across subjects and structures; DDMmean of the bladder ranged from 2 to 13 mm and from 1 to 11 mm for the rectum. There was a high correlation between DDMmean of the bladder and the rectum (Pearson’s correlation coefficient, Rp = 0.62). The correlation between DDMmean and the volume ratios post-DIR was stronger (Rp = 0.51; 0.68) than the correlation with the TE (bladder: Rp = 0.46; rectum: Rp = 0.47), or the ICE (bladder: Rp = 0.34; rectum: Rp = 0.37). There was a negative correlation between DSC and DDMmean of both the bladder (Rp = −0.23) and the rectum (Rp = −0.63). The DDM uncertainty metric indicated considerable DIR variability across subjects and structures. Our results show a stronger correlation with volume ratios and with the DSC using DDM compared to using ICE and TE. The DDM has the potential to quantitatively identify regions of large DIR uncertainties and consequently identify anatomical/scan outliers. The DDM can, thus, be applied to improve the adaptive RT process for tumor sites subject to motion.

TU‐AB‐202‐03: Prediction of PET Transfer Uncertainty by DIR Error Estimating Software, AUTODIRECT

Purpose:Deformable image registration (DIR) is a powerful tool, but DIR errors can adversely affect its clinical applications. To estimate voxel‐specific DIR uncertainty, a software tool, called AUTODIRECT (automated DIR evaluation of confidence tool), has been developed and validated. This work tests the ability of this software to predict uncertainty for the transfer of standard uptake values (SUV) from positron‐emission tomography (PET) with DIR.Methods:Virtual phantoms are used for this study. Each phantom has a planning computed tomography (CT) image and a diagnostic PET‐CT image set. A deformation was digitally applied to the diagnostic CT to create the planning CT image and establish a known deformation between the images. One lung and three rectum patient datasets were employed to create the virtual phantoms. Both of these sites have difficult deformation scenarios associated with them, which can affect DIR accuracy (lung tissue sliding and changes in rectal filling). The virtual phantoms were created to simulate these scenarios by introducing discontinuities in the deformation field at the lung rectum border. The DIR algorithm from Plastimatch software was applied to these phantoms. The SUV mapping errors from the DIR were then compared to that predicted by AUTODIRECT.Results:The SUV error distributions closely followed the AUTODIRECT predicted error distribution for the 4 test cases. The minimum and maximum PET SUVs were produced from AUTODIRECT at 95% confidence interval before applying gradient‐based SUV segmentation for each of these volumes. Notably, 93.5% of the target volume warped by the true deformation was included within the AUTODIRECT‐predicted maximum SUV volume after the segmentation, while 78.9% of the target volume was within the target volume warped by Plastimatch.Conclusion:The AUTODIRECT framework is able to predict PET transfer uncertainty caused by DIR, which enables an understanding of the associated target volume uncertainty.

Deformable image registration and interobserver variation in contour propagation for radiation therapy planning.

Deformable image registration (DIR) and interobserver variation inevitably introduce uncertainty into the treatment planning process. The purpose of the current work was to measure deformable image registration (DIR) errors and interobserver variability for regions of interest (ROIs) in the head and neck and pelvic regions. Measured uncertainties were combined to examine planning margin adequacy for contours propagated for adaptive therapy and to assess the trade‐off of DIR and interobserver uncertainty in atlas‐based automatic segmentation. Two experienced dosimetrists retrospectively contoured brainstem, spinal cord, anterior oral cavity, larynx, right and left parotids, optic nerves, and eyes on the planning CT (CT1) and attenuation‐correction CT of diagnostic PET/CT (CT2) for 30 patients who received radiation therapy for head and neck cancer. Two senior radiation oncology residents retrospectively contoured prostate, bladder, and rectum on the postseed‐implant CT (CT1) and planning CT (CT2) for 20 patients who received radiation therapy for prostate cancer. Interobserver variation was measured by calculating mean Hausdorff distances between the two observers' contours. CT2 was deformably registered to CT1 via commercially available multipass B‐spline DIR. CT2 contours were propagated and compared with CT1 contours via mean Hausdorff distances. These values were summed in quadrature with interobserver variation for margin analysis and compared with interobserver variation for statistical significance using two‐tailed t‐tests for independent samples (α=0.05). Combined uncertainty ranged from 1.5‐5.8 mm for head and neck structures and 3.1‐3.7 mm for pelvic structures. Conventional 5 mm margins may not be adequate to cover this additional uncertainty. DIR uncertainty was significantly less than interobserver variation for four head and neck and one pelvic ROI. DIR uncertainty was not significantly different than interobserver variation for four head and neck and one pelvic ROI. DIR uncertainty was significantly greater than interobserver variation for two head and neck and one pelvic ROI. The introduction of DIR errors may offset any reduction in interobserver variation by using atlas‐based automatic segmentation.PACS number(s): 87.57.nj, 87.55.D‐

Quantification of dose uncertainties in cumulated dose estimation compared to planned dose in prostate IMRT

Background and purposeIn prostate IMRT, the objectives were to quantify, for the bladder and the rectum: (i) the dose difference (DD) between the planned dose (PD) and the estimated cumulated dose (ECD) by deformable image registration (DIR); (ii) the dose accumulation uncertainties (DAU). Materials and methodsA series of 24 patients receiving 80Gy in the prostate was used to calculate the ECDpts and the DDpts. To evaluate the DAU, a numerical phantom (ph) simulating deformations of main pelvic organs was used to calculate the ECDph using the same DIR method. A reference cumulated dose (RCDph) was calculated, based on the simulated deformations. The DAUph was defined by the differences between RCDph and ECDph. ResultsFor the mean dose to the bladder, the standard deviation of DDpts was 6.9Gy (18.1Gy maximum) with a DAUph of 2.7Gy. For the rectum wall, it was 2.0Gy (4.2Gy maximum) with a DAUph of 1.2Gy. Volume differences between PDpts and ECDpts, along the dose–volume histogram, ranged from −30% to +37% and −14% to +14% for the bladder and rectum, respectively. The corresponding uncertainties ranged from −23% to +7% and −4% to +7% for the bladder and rectum, respectively. ConclusionsLarge differences between planned and delivered doses to the bladder have been quantified and are higher than the uncertainties of the DIR method. For the rectum, the differences are smaller and close to the DIR uncertainties.

Effect of deformable registration uncertainty on lung SBRT dose accumulation.

Deformable image registration (DIR) plays an important role in dose accumulation, such as incorporating breathing motion into the accumulation of the delivered dose based on daily 4DCBCT images. However, it is not yet well understood how the uncertainties associated with DIR methods affect the dose calculations and resulting clinical metrics. The purpose of this study is to evaluate the impact of DIR uncertainty on the clinical metrics derived from its use in dose accumulation. A biomechanical model based DIR method and a biomechanical-intensity-based hybrid method, which reduced the average registration error by 1.6 mm, were applied to ten lung cancer patients. A clinically relevant dose parameter [minimum dose to 0.5 cm(3) (Dmin)] was calculated for three dose scenarios using both algorithms. Dose scenarios included static (no breathing motion), predicted (breathing motion at the time of planning), and total accumulated (interfraction breathing motion). The relationship between the dose parameter and a combination of DIR uncertainty metrics, tumor volume, and dose heterogeneity of the plan was investigated. Depending on the dose heterogeneity, tumor volume, and DIR uncertainty, in over 50% of the patients, differences greater than 1.0 Gy were observed in the Dmin of the tumor in the static dose calculation on exhale phase of the 4DCT. Such differences were due to the errors in propagating the tumor contours from the reference planning 4DCT phase onto a subsequent 4DCT phase using each DIR algorithm and calculating the dose on that phase. The differences in predicted dose were more subtle when breathing motion was modeled explicitly at the time of planning with only one patient exhibiting a greater than 1.0 Gy difference in Dmin. Dmin differences of up to 2.5 Gy were found in the total accumulated delivered dose due to difference in quantifying the interfraction variations. Such dose uncertainties could potentially be clinically significant. Reductions in average uncertainty in DIR algorithms by 1.6 mm may have a clinically significant impact on the decision-making metrics used in dose planning and dose accumulation assessment.

SU-E-J-95: A Novel Objective Approach to Identify Scan Outliers in Deformable Image Registration for Longitudinal Datasets

Purpose: The current practice for evaluating deformable image registration (DIR) is commonly subjective as it involves input from an observer due to lack of an absolute ground truth. We have therefore developed an automated objective method to evaluate DIR in order to identify scan outliers for longitudinal datasets. Methods: The imaging dataset consisted of nine repeated CT scans (CT1–9) from four prostate patients. Six “similar” CT scans (CT1–6) from one patient (PT1) were used as ground truth meanwhile three other CT scans (CT7–9) from PT2–4 served as scan outliers. Voxel-by-voxel DIR-related uncertainties (Distance Discordance Metric, DDM) were evaluated on groups of five CT scans for the following ten scenarios: Scenario 1–5: five out of six CT scans (CT1–6) from PT1 using leave-one-out technique were used; Scenario 6–8: single scan from PT1 was replaced by another CT scan from PT2–4; and Scenario 9–10: two to three CT scans from PT1 were replaced by CT scans from PT2–4. For each scenario, the DDM map was superimposed on CT-1 and two-sample t-test was performed to compare the uncertainty distributions for all 10 scenarios. Results: The mean DDM values were 3.8–4.2 mm for the first five scenarios, 5.1–6.7 mm for scenario 6–8, and 8.3 mm and 8.9 mm for scenario 9 and scenario 10, respectively. The two-sample t-test showed that the DDM distributions for scenarios 1–5 have a similar mean while the distributions for scenarios 6–10 have statistically different means compared to scenario 1–5. Conclusion: The DIR uncertainty distributions as estimated from our method on a set of images from the same patient are of similar magnitude. When images for a patient are replaced by images from other patients, the distribution of the uncertainties changes considerably, motivating the use of DDM as a metric to automatically detect scan outliers.

SU-E-J-92: Validating Dose Uncertainty Estimates Produced by AUTODIRECT, An Automated Program to Evaluate Deformable Image Registration Accuracy

Purpose: Deformable image registration (DIR) is a powerful tool with the potential to deformably map dose from one computed-tomography (CT) image to another. Errors in the DIR, however, will produce errors in the transferred dose distribution. We have proposed a software tool, called AUTODIRECT (automated DIR evaluation of confidence tool), which predicts voxel-specific dose mapping errors on a patient-by-patient basis. This work validates the effectiveness of AUTODIRECT to predict dose mapping errors with virtual and physical phantom datasets. Methods: AUTODIRECT requires 4 inputs: moving and fixed CT images and two noise scans of a water phantom (for noise characterization). Then, AUTODIRECT uses algorithms to generate test deformations and applies them to the moving and fixed images (along with processing) to digitally create sets of test images, with known ground-truth deformations that are similar to the actual one. The clinical DIR algorithm is then applied to these test image sets (currently 4) . From these tests, AUTODIRECT generates spatial and dose uncertainty estimates for each image voxel based on a Student's t distribution. This work compares these uncertainty estimates to the actual errors made by the Velocity Deformable Multi Pass algorithm on 11 virtual and 1 physical phantom datasets. Results: For 11 of the 12 tests, the predicted dose error distributions from AUTODIRECT are well matched to the actual error distributions within 1–6% for 10 virtual phantoms, and 9% for the physical phantom. For one of the cases though, the predictions underestimated the errors in the tail of the distribution. Conclusion: Overall, the AUTODIRECT algorithm performed well on the 12 phantom cases for Velocity and was shown to generate accurate estimates of dose warping uncertainty. AUTODIRECT is able to automatically generate patient-, organ-, and voxel-specific DIR uncertainty estimates. This ability would be useful for patient-specific DIR quality assurance.