Deformable Image Registration Research Articles

Inter- and intrafractional 4D dose accumulation for evaluating ΔNTCP robustness in lung cancer

Background and purposeModel-based selection of proton therapy patients relies on a predefined reduction in normal tissue complication probability (NTCP) with respect to photon therapy. The decision is necessarily made based on the treatment plan, but NTCP can be affected when the delivered treatment deviates from the plan due to delivery inaccuracies. Especially for proton therapy of lung cancer, this can be important because of tissue density changes and, with pencil beam scanning, the interplay effect between the proton beam and breathing motion. Materials and methodsIn this work, we verified whether the expected benefit of proton therapy is retained despite delivery inaccuracies by reconstructing the delivered treatment using log-file based dose reconstruction and inter- and intrafractional accumulation. Additionally, the importance of two uncertain parameters for treatment reconstruction, namely deformable image registration (DIR) algorithm and α/β ratio, was assessed. ResultsThe expected benefit or proton therapy was confirmed in 97% of all studied cases, despite regular differences up to 2 percent point (p.p.) NTCP between the delivered and planned treatments. The choice of DIR algorithm affected NTCP up to 1.6 p.p., an order of magnitude higher than the effect of α/β ratio. ConclusionFor the patient population and treatment technique employed, the predicted clinical benefit for patients selected for proton therapy was confirmed for 97.0% percent of all cases, although the NTCP based proton selection was subject to 2 p.p. variations due to delivery inaccuracies.

Assessment of delivered dose in prostate cancer patients treated with ultra-hypofractionated radiotherapy on 1.5-Tesla MR-Linac.

To quantitatively characterize the dosimetric effects of long on-couch time in prostate cancer patients treated with adaptive ultra-hypofractionated radiotherapy (UHF-RT) on 1.5-Tesla magnetic resonance (MR)-linac. Seventeen patients consecutively treated with UHF-RT on a 1.5-T MR-linac were recruited. A 36.25 Gy dose in five fractions was delivered every other day with a boost of 40 Gy to the whole prostate. We collected data for the following stages: pre-MR, position verification-MR (PV-MR) in the Adapt-To-Shape (ATS) workflow, and 3D-MR during the beam-on phase (Bn-MR) and at the end of RT (post-MR). The target and organ-at-risk contours in the PV-MR, Bn-MR, and post-MR stages were projected from the pre-MR data by deformable image registration and manually adapted by the physician, followed by dose recalculation for the ATS plan. Overall, 290 MR scans were collected (85 pre-MR, 85 PV-MR, 49 Bn-MR and 71 post-MR scans). With a median on-couch time of 49 minutes, the mean planning target volume (PTV)-V95% of all scans was 97.83 ± 0.13%. The corresponding mean clinical target volume (CTV)-V100% was 99.93 ± 0.30%, 99.32 ± 1.20%, 98.59 ± 1.84%, and 98.69 ± 1.85%. With excellent prostate-V100% dose coverage, the main reason for lower CTV-V100% was slight underdosing of seminal vesicles (SVs). The median V29 Gy change in the rectal wall was -1% (-20%-17%). The V29 Gy of the rectal wall increased by >15% was observed in one scan. A slight increase in the high dose of bladder wall was noted due to gradual bladder growth during the workflow. This 3D-MR-based dosimetry analysis demonstrated clinically acceptable estimated dose coverage of target volumes during the beam-on period with adaptive ATS workflow on 1.5-T MR-linac, albeit with a relatively long on-couch time. The 3-mm CTV-PTV margin was adequate for prostate irradiation but occasionally insufficient for SVs. More attention should be paid to restricting high-dose RT to the rectal wall when optimizing the ATS plan.

Arbitrary Order Total Variation for Deformable Image Registration

In this work, we investigate image registration in a variational framework and focus on regularization generality and solver efficiency. We first propose a variational model combining the state-of-the-art sum of absolute differences (SAD) and a new arbitrary order total variation regularization term. The main advantage is that this variational model preserves discontinuities in the resultant deformation while being robust to outlier noise. It is however non-trivial to optimize the model due to its non-convexity, non-differentiabilities, and generality in the derivative order. To tackle these, we propose to first apply linearization to the model to formulate a convex objective function and then break down the resultant convex optimization into several point-wise, closed-form subproblems using a fast, over-relaxed alternating direction method of multipliers (ADMM). With this proposed algorithm, we show that solving higher-order variational formulations is similar to solving their lower-order counterparts. Extensive experiments show that our ADMM is significantly more efficient than both the subgradient and primal-dual algorithms particularly when higher-order derivatives are used, and that our new models outperform state-of-the-art methods based on deep learning and free-form deformation. Our code implemented in both Matlab and Pytorch is publicly available at https://github.com/j-duan/AOTV.

Validation of a surface-based deformable MRI-3D ultrasound image registration algorithm toward clinical implementation for interstitial prostate brachytherapy

PURPOSEThe purpose of this study was to evaluate and clinically implement a deformable surface-based magnetic resonance imaging (MRI) to three-dimensional ultrasound (US) image registration algorithm for prostate brachytherapy (BT) with the aim to reduce operator dependence and facilitate dose escalation to an MRI-defined target. METHODS AND MATERIALSOur surface-based deformable image registration (DIR) algorithm first translates and scales to align the US- and MR-defined prostate surfaces, followed by deformation of the MR-defined prostate surface to match the US-defined prostate surface. The algorithm performance was assessed in a phantom using three deformation levels, followed by validation in three retrospective high-dose-rate BT clinical cases. For comparison, manual rigid registration and cognitive fusion by physician were also employed. Registration accuracy was assessed using the Dice similarity coefficient (DSC) and target registration error (TRE) for embedded spherical landmarks. The algorithm was then implemented intraoperatively in a prospective clinical case. RESULTSIn the phantom, our DIR algorithm demonstrated a mean DSC and TRE of 0.74 ± 0.08 and 0.94 ± 0.49 mm, respectively, significantly improving the performance compared to manual rigid registration with 0.64 ± 0.16 and 1.88 ± 1.24 mm, respectively. Clinical results demonstrated reduced variability compared to the current standard of cognitive fusion by physicians. CONCLUSIONSWe successfully validated a DIR algorithm allowing for translation of MR-defined target and organ-at-risk contours into the intraoperative environment. Prospective clinical implementation demonstrated the intraoperative feasibility of our algorithm, facilitating targeted biopsies and dose escalation to the MR-defined lesion. This method provides the potential to standardize the registration procedure between physicians, reducing operator dependence.

Prediction of adaptive strategies based on deformation vector field features for MR-guided adaptive radiotherapy of prostate cancer.

The selection of adaptive strategies in MR-guided adaptive radiotherapy (MRgART) usually relies on subjective review of anatomical changes. However, this kind of review may lead to improper selection of adaptive strategy for some fractions. The purpose of this study was to develop prediction models based on deformation vector field (DVF) features for automatic and accurate strategy selection, using prostate cancer as an example. 100 fractions of 20 prostate cancer patients were retrospectively selected in this study. Treatment plans using both adapt to position (ATP) strategy and adapt to shape (ATS) strategy were generated. Optimal adaptive strategy was determined according to dosimetric evaluation. DVFs of the deformable image registration (DIR) of daily MRI and CT simulation scans were extracted. The shape, first order statistics, and spatial features were extracted from the DVFs, subjected to further selection using the minimum redundancy maximum relevance (mRMR) method. The number of features (Fn ) was hyper-tuned using bootstrapping method, and then Fn indicating a peak area under the curve (AUC) value was used to construct three prediction models. According to subjective review, the ATS strategy was adopted for all 100 fractions. However, the evaluation results showed that the ATP strategy could have met the clinical requirements for 23 (23%) fractions. The three prediction models showed high prediction performance, with the best performing model achieving an AUC value of 0.896, corresponding accuracy (ACC), sensitivity (SEN) and specificity (SPC) of 0.9, 0.958, and 0.667, respectively. The features used to construct prediction models included four features extracted from y direction of DVF (DVFy ) and mask, one feature from z direction of DVF (DVFz ). It indicated that the deformation along the anterior-posterior direction had a greater impact on determining the adaptive strategy than other directions. DVF-feature-based models could accurately predict the adaptive strategy and avoid unnecessary selection of time-consuming ATS strategy, which consequently improves the efficiency of the MRgART process.

Dosimetric impact of deformable image registration using radiophotoluminescent glass dosimeters with a physical geometric phantom.

To study the dosimetry impact of deformable image registration (DIR) using radiophotoluminescent glass dosimeter (RPLD) and custom developed phantom with various inserts. The phantom was developed to facilitate simultaneous evaluation of geometric and dosimetric accuracy of DIR. Four computed tomography (CT) images of the phantom were acquired with four different configurations. Four volumetric modulated arc therapy (VMAT) plans were computed for different phantom. Two different patterns were applied to combination of four phantom configurations. RPLD dose measurement was combined between corresponding two phantom configurations. DIR-based dose accumulation was calculated between corresponding two CT images with two commercial DIR software and various DIR parameter settings, and an open source software. Accumulated dose calculated using DIR was then compared with measured dose using RPLD. The mean±standard deviation (SD) of dose difference was 2.71±0.23% (range, 2.22%-3.01%) for tumor-proxy and 3.74±0.79% (range, 1.56%-4.83%) for rectum-proxy. The mean±SD of target registration error (TRE) was 1.66±1.36mm (range, 0.03-4.43mm) for tumor-proxy and 6.87±5.49mm (range, 0.54-17.47mm) for rectum-proxy. These results suggested that DIR accuracy had wide range among DIR parameter setting. The dose difference observed in our study was 3% for tumor-proxy and within 5% for rectum-proxy. The custom developed physical phantom with inserts showed potential for accurate evaluation of DIR-based dose accumulation. The prospect of simultaneous evaluation of geometric and dosimetric DIR accuracy in a single phantom may be useful for validation of DIR for clinical use.

Use of Contrast Media During CT-guided Thermal Ablation of Colorectal Liver Metastasis for Procedure Planning is Associated with Improved Immediate Outcomes.

The aim of this study was to analyze the impact of using intra-procedural pre-ablation contrast-enhanced CT prior to percutaneous thermal ablation (pre-ablation CECT) of colorectal liver metastases (CLM) on local outcomes. This retrospective analysis of a prospectively collected liver ablation registry included 144 consecutive patients (median age 57years IQR [49, 65], 60% men) who underwent 173 CT-guided ablation sessions for 250 CLM between October 2015 and March 2020. In addition to oncologic outcomes, technical success was retrospectively evaluated using a biomechanical deformable image registration software for 3D-minimal ablative margin (3D-MAM) quantification. Bayesian regression was used to estimate effects of pre-ablation CECT on residual unablated tumor, 3D-MAM, and local tumor progression-free survival (LTPFS). Pre-ablation CECT was acquired in 71/173 (41%) sessions. Residual unablated tumor was present in one (0.9%) versus nine tumors (6.6%) ablated with versus without using pre-ablation CECT, respectively (p = 0.024). Pre-ablation CECT use decreased the odds of residual disease on first follow-up by 78% (CI95% [5, 86]) and incomplete ablation (3D-MAM ≤ 0mm) by 58% (CI95% [13, 122]). The odds ratio for residual unablated tumor for larger CLM was lower when pre-ablation CECT was used (odds ratio 1.0 with pre-ablation CECT vs. 2.52 without). Pre-ablation CECT use was not associated with improvements on LTPFS. Pre-ablation CECT is associated with improved immediate outcomes by significantly reducing the incidence of residual unablated tumor and by mitigating the risk of incomplete ablation for larger CLM. We recommend performing baseline intra-procedural pre-ablation CECT as a standard imaging protocol. Level 3 (retrospective cohort study).

Dosimetric comparison of deformable image registration and synthetic CT generation based on CBCT images for organs at risk in cervical cancer radiotherapy

ObjectiveAnatomical variations existing in cervical cancer radiotherapy treatment can be monitored by cone-beam computed tomography (CBCT) images. Deformable image registration (DIR) from planning CT (pCT) to CBCT images and synthetic CT (sCT) image generation based on CBCT are two methods for improving the quality of CBCT images. This study aims to compare the accuracy of these two approaches geometrically and dosimetrically in cervical cancer radiotherapy.MethodsIn this study, 40 paired pCT-CBCT images were collected to evaluate the accuracy of DIR and sCT generation. The DIR method was based on a 3D multistage registration network that was trained with 150 paired pCT-CBCT images, and the sCT generation method was performed based on a 2D cycle-consistent adversarial network (CycleGAN) with 6000 paired pCT-CBCT slices for training. Then, the doses were recalculated with the CBCT, pCT, deformed pCT (dpCT) and sCT images by a GPU-based Monte Carlo dose code, ArcherQA, to obtain DoseCBCT, DosepCT, DosedpCT and DosesCT. Organs at risk (OARs) included small intestine, rectum, bladder, spinal cord, femoral heads and bone marrow, CBCT and pCT contours were delineated manually, dpCT contours were propagated through deformation vector fields, sCT contours were auto-segmented and corrected manually.ResultsThe global gamma pass rate of DosesCT and DosedpCT was 99.66% ± 0.34%, while that of DoseCBCT and DosedpCT was 85.92% ± 7.56% at the 1%/1 mm criterion and a low-dose threshold of 10%. Based on DosedpCT as uniform dose distribution, there were comparable errors in femoral heads and bone marrow for the dpCT and sCT contours compared with CBCT contours, while sCT contours had lower errors in small intestine, rectum, bladder and spinal cord, especially for those with large volume difference of pCT and CBCT.ConclusionsFor cervical cancer radiotherapy, the DIR method and sCT generation could produce similar precise dose distributions, but sCT contours had higher accuracy when the difference in planning CT and CBCT was large.

Comparison of Lower Extremity Computed Tomography Angiography by Using Different Subtraction Techniques

To compare the diagnostic capabilities of orbital synchronized helical scanning at lower extremity computed tomography angiography between the Add/Sub software and the deformable image registration. From March 2015 to December 2016, 100 dialysis patients underwent orbital synchronized lower limb CT subtraction angiography and lower limb endovascular treatment within 4 months. For the visual evaluation of blood vessels in the lower extremities, a stenosis rate of 50% or more was considered to be stenosis. It was classified into two areas: above-knee (AK) region (superficial femoral artery and popliteal artery) and below-knee (BK) region (anterior tibial artery, posterior tibial artery, and fibula artery). Considering angiography for the lower limb endovascular treatment as the golden standard, we calculated the sensitivity, specificity, positive-predictive value, negative-predictive value, and diagnostic capabilities. Receiver operating characteristic curve (ROC) analysis was performed to calculate the area under curve (AUC). Calcification subtraction failure was observed to be 11% in the AK region and 2% in the BK region using the Add/Sub software. The specificity, positive-predictive value, diagnostic capabilities, and AUC of the deformable image registration were lower than those of the Add/Sub software. Add/Sub software and deformable image registration have high diagnostic capability to remove calcification. On the other hand, the specificity and AUC of the deformable image registration were lower than those of the Add/Sub software. Also, even if the same deformable image registration is used, caution is required because the diagnostic performance varies depending on the site.

DragNet: Learning-based deformable registration for realistic cardiac MR sequence generation from a single frame.

Deformable image registration (DIR) can be used to track cardiac motion. Conventional DIR algorithms aim to establish a dense and non-linear correspondence between independent pairs of images. They are, nevertheless, computationally intensive and do not consider temporal dependencies to regulate the estimated motion in a cardiac cycle. In this paper, leveraging deep learning methods, we formulate a novel hierarchical probabilistic model, termed DragNet, for fast and reliable spatio-temporal registration in cine cardiac magnetic resonance (CMR) images and for generating synthetic heart motion sequences. DragNet is a variational inference framework, which takes an image from the sequence in combination with the hidden states of a recurrent neural network (RNN) as inputs to an inference network per time step. As part of this framework, we condition the prior probability of the latent variables on the hidden states of the RNN utilised to capture temporal dependencies. We further condition the posterior of the motion field on a latent variable from hierarchy and features from the moving image. Subsequently, the RNN updates the hidden state variables based on the feature maps of the fixed image and the latent variables. Different from traditional methods, DragNet performs registration on unseen sequences in a forward pass, which significantly expedites the registration process. Besides, DragNet enables generating a large number of realistic synthetic image sequences given only one frame, where the corresponding deformations are also retrieved. The probabilistic framework allows for computing spatio-temporal uncertainties in the estimated motion fields. Our results show that DragNet performance is comparable with state-of-the-art methods in terms of registration accuracy, with the advantage of offering analytical pixel-wise motion uncertainty estimation across a cardiac cycle and being a motion generator. We will make our code publicly available.

Evaluation of RayStation's delivered dose and accumulated dose features for spine stereotactic radiotherapy

<h2>Abstract</h2> We investigated delivered dose and dose accumulation features in the dose tracking module of RayStation version 11B before potential integration into the spine stereotactic radiosurgery (SSRS) program at our institution. End-to-end testing with a rigid Rando phantom was performed, and 10 retrospective clinical cases were selected for evaluation. Pre-treatment cone beam CTs (pCBCT) were corrected for Hounsfield unit (HU) integrity and contours were rigidly copied from the reference plan. We then calculated the delivered dose to the corrected cone beam CT (cCBCT). A deformable image registration (DIR) was generated between cCBCT and reference planning CT (rCT) using controlling region of interests. Deformed delivered dose to the rCT was summed to generate the accumulated dose for multiple fractions. The end-to-end tests of the phantom study revealed an improvement of cCBCT HU information by > 100 HU compared to the pCBCT. When compared to the reference plan, the delivered dose and deformed delivered dose were within 1.0% for the GTV and CTV and 3.0% for the spinal cord, respectively. Nearly all 10 clinical cases demonstrated delivered dose and accumulated dose deviations < 3.0% from the reference plan. However, 2 patients rendered delivered dose deviations between 3.0% and 4.0%, showing the effectiveness of the module. The dose tracking module in RayStation version 11B could potentially be utilized to aid clinical decision-making for external body shape change or positional deviation in SSRS for rigid target and critical structures. Evaluation before clinical application under one's specific practice is required, and results must be carefully analyzed specially near the high dose gradient area.

Deformable Image Registration Using Vision Transformers for Cardiac Motion Estimation from Cine Cardiac MRI Images.

Accurate cardiac motion estimation is a crucial step in assessing the kinematic and contractile properties of the cardiac chambers, thereby directly quantifying the regional cardiac function, which plays an important role in understanding myocardial diseases and planning their treatment. Since the cine cardiac magnetic resonance imaging (MRI) provides dynamic, high-resolution 3D images of the heart that depict cardiac motion throughout the cardiac cycle, cardiac motion can be estimated by finding the optical flow representation between the consecutive 3D volumes from a 4D cine cardiac MRI dataset, thereby formulating it as an image registration problem. Therefore, we propose a hybrid convolutional neural network (CNN) and Vision Transformer (ViT) architecture for deformable image registration of 3D cine cardiac MRI images for consistent cardiac motion estimation. We compare the image registration results of our proposed method with those of the VoxelMorph CNN model and conventional B-spline free form deformation (FFD) non-rigid image registration algorithm. We conduct all our experiments on the open-source Automated Cardiac Diagnosis Challenge (ACDC) dataset. Our experiments show that the deformable image registration results obtained using the proposed method outperform the CNN model and the traditional FFD image registration method.

Automated Learning for Deformable Medical Image Registration by Jointly Optimizing Network Architectures and Objective Functions.

Deformable image registration plays a critical role in various tasks of medical image analysis. A successful registration algorithm, either derived from conventional energy optimization or deep networks, requires tremendous efforts from computer experts to well design registration energy or to carefully tune network architectures with respect to medical data available for a given registration task/scenario. This paper proposes an automated learning registration algorithm (AutoReg) that cooperatively optimizes both architectures and their corresponding training objectives, enabling non-computer experts to conveniently find off-the-shelf registration algorithms for various registration scenarios. Specifically, we establish a triple-level framework to embrace the searching for both network architectures and objectives with a cooperating optimization. Extensive experiments on multiple volumetric datasets and various registration scenarios demonstrate that AutoReg can automatically learn an optimal deep registration network for given volumes and achieve state-of-the-art performance. The automatically learned network also improves computational efficiency over the mainstream UNet architecture from 0.558 to 0.270 seconds for a volume pair on the same configuration.

GMCNet: A Generative Multi-Resolution Framework for Cardiac Registration

Deformable image registration plays a crucial role in estimating cardiac deformation from a sequence of images. However, existing registration methods primarily process images as pairs instead of processing all images in a sequence together. This study proposes a novel end-to-end learning-free generative multi-resolution convolutional neural network (GMCNet) with the primary focus of registering images in a sequence. Even though learning-based methods have yielded high performance for image registration, their performance depends on their ability to learn information from a large number of samples which are difficult to obtain and might bias the framework to the specific domain of data. The proposed learning-free method eliminates the need for a dedicated training set while exploiting the capabilities of neural networks to achieve accurate deformation fields. Due to its capability of parameter sharing through the architecture, the GMCNet can be used as a groupwise registration as well as pairwise registration. The proposed method was evaluated on three different clinical cardiac magnetic resonance imaging datasets and compared quantitatively against nine other state-of-the-art learning and optimization-based algorithms. The proposed method outperformed other methods in all comparisons and yielded average Dice metric values ranging from 0.85 to 0.88 for the datasets. Different aspects of the GMCNet are also explored by assessing 1) the robustness; 2) performance on pairwise registration; 3) the influence of spatial transformation in a controlled environment; and 4) the impact of different multi-resolution structures. The results demonstrate that using temporal information to estimate the deformation fields leads to more accurate registration results and improved robustness under different noise levels. Moreover, the proposed method does not need images for training, and therefore, its prediction is not domain-specific and can be applied to any sequence of images.

Determining Combined Modality Dosimetric Constraints by Integration of IMRT and LDR Prostate Brachytherapy Dosimetry and Correlation with Toxicity

Intermediate- and high-risk prostate cancer patients undergoing combination external beam radiation therapy (EBRT) and low dose rate (LDR) brachytherapy have demonstrated increased genitourinary (GU) toxicity. We have previously demonstrated a method to combine EBRT and LDR dosimetry. In this work, we use this technique for a sample of patients with intermediate- and high-risk prostate cancer, correlate with clinical toxicity, and suggest preliminary summed organ-at-risk constraints for future investigation. Intensity modulated EBRT and 103Pd-based LDR treatment plans were combined for 138 patients using biological effective dose (BED) and deformable image registration. GU and gastrointestinal (GI) toxicity were compared with combined dosimetry for the urethra, bladder, and rectum. Differences between doses in each toxicity grade were assessed by analysis of variance (α=0.05). Combined dosimetric constraints are proposed using the mean organ-at-risk dose, subtracting 1 standard deviation for a conservative recommendation. The majority of our 138-patient cohort experienced grade 0 to 2 GU or GI toxicity. Six grade 3 toxicities were noted. Mean prostate BED D90 (± 1 standard deviation) was 165.5±11.1 Gy. Mean urethra BED D10 was 230.3±33.9 Gy. Mean bladder BED was 35.2±11.0 Gy. Mean rectum BED D2cc was 85.6±24.3 Gy. Significant dosimetric differences between toxicity grades were found for mean bladder BED, bladder D15, and rectum D50, but differences between individual means were not statistically significant. Given the low incidence of grade 3 GU and GI toxicity, we propose urethra D10 <200 Gy, rectum D2cc <60 Gy, and bladder D15 <45 Gy as preliminary dose constraints for combined modality therapy. We successfully applied our dose integration technique to a sample of patients with intermediate- and high-risk prostate cancer. Incidence of grade 3 toxicity was low, suggesting that combined doses observed in this study were safe. We suggest preliminary dose constraints as a conservative starting point to investigate and escalate prospectively in a future study.

Deformed dose restoration to account for tumor deformation and position changes for adaptive proton therapy.

Online adaptation during intensity-modulated proton therapy (IMPT) can minimize the effect of inter-fractional anatomical changes, but remains challenging because of the complex workflow. One approach for fast and automated online IMPT adaptation is dose restoration, which restores the initial dose distribution on the updated anatomy. However, this method may fail in cases where tumor deformation or position changes occur. To develop a fast and robust IMPT online adaptation method named "deformed dose restoration (DDR)" that can adjust for inter-fractional tumor deformation and position changes. THE DDR METHOD COMPRISES TWO STEPS: : (1) calculation of the deformed dose distribution, and (2) restoration of the deformed dose distribution. First, the deformable image registration (DIR) between the initial clinical target volume (CTV) and the new CTV were performed to calculate the vector field. To ensure robustness for setup and range uncertainty and the ability to restore the deformed dose distribution, an expanded CTV-based registration to maintain the dose gradient outside the CTV was developed. The deformed dose distribution was obtained by applying the vector field to the initial dose distribution. Then, the voxel-by-voxel dose difference optimization was performed to calculate beam parameters that restore the deformed dose distribution on the updated anatomy. The optimization function was the sum of total dose differences and dose differences of each field to restore the initial dose overlap of each field. This method only requires target contouring, which eliminates the need for organs at risk (OARs) contouring. Six clinical cases wherein the tumor deformation and/or position changed on repeated CTs were selected. DDR feasibility was evaluated by comparing the results with those from three other strategies, namely, not adapted (continuing the initial plan), adapted by previous dose restoration, and fully optimized. In all cases, continuing the initial plan was largely distorted on the repeated CTs and the dose-volume histogram (DVH) metrics for the target were reduced due to the tumor deformation or position changes. On the other hand, DDR improved DVH metrics for the target to the same level as the initial dose distribution. Dose increase was seen for some OARs because tumor growth had reduced the relative distance between CTVs and OARs. Robustness evaluation for setup and range uncertainty (3 mm/3.5%) showed that deviation in DVH-bandwidth for CTV D95% from the initial plan was 0.4 ± 0.5% (Mean ± S.D.) for DDR. The calculation time was 8.1 ± 6.4 min. An online adaptation algorithm was developed that improved the treatment quality for inter-fractional anatomical changes and retained robustness for intra-fractional setup and range uncertainty. The main advantage of this method is that it only requires target contouring alone and saves the time for OARs contouring. The fast and robust adaptation method for tumor deformation and position changes described here can reduce the need for offline adaptation and improve treatment efficiency. This article is protected by copyright. All rights reserved.

Quality assurance assessment of intra-acquisition diffusion-weighted and T2-weighted magnetic resonance imaging registration and contour propagation for head and neck cancer radiotherapy.

Adequate image registration of anatomical and functional magnetic resonance imaging (MRI) scans is necessary for MR-guided head and neck cancer (HNC) adaptive radiotherapy planning. Despite the quantitative capabilities of diffusion-weighted imaging (DWI) MRI for treatment plan adaptation, geometric distortion remains a considerable limitation. Therefore, we systematically investigated various deformable image registration (DIR) methods to co-register DWI and T2-weighted (T2W) images. We compared three commercial (ADMIRE, Velocity, Raystation) and three open-source (Elastix with default settings [Elastix Default], Elastix with parameter set 23 [Elastix 23], Demons) post-acquisition DIR methods applied to T2W and DWI MRI images acquired during the same imaging session in twenty immobilized HNC patients. In addition, we used the non-registered images (None) as a control comparator. Ground-truth segmentations of radiotherapy structures (tumour and organs at risk) were generated by a physician expert on both image sequences. For each registration approach, structures were propagated from T2W to DWI images. These propagated structures were then compared with ground-truth DWI structures using the Dice similarity coefficient and mean surface distance. 19 left submandibular glands, 18 right submandibular glands, 20left parotid glands, 20 right parotid glands, 20 spinal cords, and 12 tumours were delineated. Most DIR methods took <30s to execute per case, with the exception of Elastix 23 which took ∼458s to execute per case. ADMIRE and Elastix 23 demonstrated improved performance over None for all metrics and structures (Bonferroni-corrected p<0.05), while the other methods did not. Moreover, ADMIRE and Elastix 23 significantly improved performance in individual and pooled analysis compared to all other methods. The ADMIRE DIR method offers improved geometric performance with reasonable execution time so should be favoured for registering T2W and DWI images acquired during the same scan session in HNC patients. These results are important to ensure the appropriate selection of registration strategies for MR-guided radiotherapy.

BIRGU Net: deformable brain magnetic resonance image registration using gyral-net map and 3D Res-Unet

Deformable image registration is a fundamental procedure in medical imaging. Recently, deep learning-based deformable image registrations have achieved fast registration by learning the spatial correspondence from image pairs. However, it remains challenging in brain image registration due to the structural complexity of individual brains and the lack of ground truth for anatomical correspondences between the brain image pairs. This work devotes to achieving an end-to-end unsupervised brain deformable image registration method using the gyral-net map and 3D Res-Unet (BIRGU Net). Firstly, the gyral-net map was introduced to represent the 3D global cortex complex information of the brain image since it was considered as one of the anatomical landmarks, which can help to extract the salient structural feature of individual brains for registration. Secondly, the variant of 3D U-net architecture involving dual residual strategies was designed to map the image into the deformation field effectively and to prevent the gradient from vanishing as well. Finally, double regularized terms were imposed on the deformation field to guide the network training for leveraging the smoothness and the topology preservation of the deformation field. The registration procedure was trained in an unsupervised manner, which addressed the lack of ground truth for anatomical correspondences between the brain image pairs. The experimental results on four public data sets demonstrate that the extracted gyral-net can be an auxiliary feature for registration and the proposed network with the designed strategies can improve the registration performance since the Dice similarity coefficient (DSC) and normalized mutual information (NMI) are higher and the time consumption is comparable than the state-of-the-art. The code is available at https://github.com/mynameiswode/BIRGU-Net .

Effects of respiratory and cardiac motion on estimating radiation dose to the left ventricle during radiotherapy for lung cancer.

Establish a workflow to evaluate radiotherapy (RT) dose variation induced by respiratory and cardiac motion on the left ventricle (LV) and left ventricular myocardium (LVM). Eight lung cancer patients underwent 4D-CT, expiratory T1-volumetric-interpolated-breath-hold-examination (VIBE), and cine MRI scans in expiration. Treatment plans were designed on the average intensity projection (AIP) datasets from 4D-CTs. RT dose from AIP was transferred onto 4D-CT respiratory phases. About 50% 4D-CT dose was mapped onto T1-VIBE (following registration) and from there onto average cine MRI datasets. Dose from average cine MRI was transferred onto all cardiac phases. Cumulative cardiac dose was estimated by transferring dose from each cardiac phase onto a reference cine phase following deformable image registration. The LV was contoured on each 4D-CT breathing phase and was called clinical LV (cLV); this structure is blurred by cardiac motion. Additionally, LV, LVM, and an American Heart Association (AHA) model were contoured on all cardiac phases. Relative maximum/mean doses for contoured regions were calculated with respect to each patient's maximum/mean AIP dose. During respiration, relative maximum and mean doses on the cLV ranged from -4.5% to 5.6% and -14.2% to 16.5%, respectively, with significant differences in relative mean doses between inspiration and expiration (P<0.0145). During cardiac motion at expiration, relative maximum and mean doses on the LV ranged from 1.6% to 59.3%, 0.5% to 27.4%, respectively. Relative mean doses were significantly different between diastole and systole (P=0.0157). No significant differences were noted between systolic, diastolic, or cumulative cardiac doses compared to the expiratory 4D-CT (P>0.14). Significant differences were observed in AHA segmental doses depending on tumour proximity compared to global LV doses on expiratory 4D-CT (P<0.0117). In this study, the LV dose was highest during expiration and diastole. Segmental evaluation suggested that future cardiotoxicity evaluations may benefit from regional assessments of dose that account for cardiopulmonary motion.

A probability model for anatomical robust optimisation in head and neck cancer proton therapy

Objective. Develop an anatomical model based on the statistics of the population data and evaluate the model for anatomical robust optimisation in head and neck cancer proton therapy. Approach. Deformable image registration was used to build the probability model (PM) that captured the major deformation from patient population data and quantified the probability of each deformation. A cohort of 20 nasopharynx patients was included in this retrospective study. Each patient had a planning CT and 6 weekly CTs during radiotherapy. We applied the model to 5 test patients. Each test patient used the remaining 19 training patients to build the PM and estimate the likelihood of a certain anatomical deformation to happen. For each test patient, a spot scanning proton plan was created. The PM was evaluated using proton spot location deviation and dose distribution. Main results. Using the proton spot range, the PM can simulate small non-rigid variations in the first treatment week within 0.21 ± 0.13 mm. For overall anatomical uncertainty prediction, the PM can reduce anatomical uncertainty from 4.47 ± 1.23 mm (no model) to 1.49 ± 1.08 mm at week 6. The 95% confidence interval (CI) of dose metric variations caused by actual anatomical deformations in the first week is −0.59% ∼ −0.31% for low-risk CT D 95, and 0.84–3.04 Gy for parotid D mean. On the other hand, the 95% CI of dose metric variations simulated by the PM at the first week is −0.52 ∼ −0.34% for low-risk CTV D 95, and 0.58 ∼ 2.22 Gy for parotid D mean. Significance. The PM improves the estimation accuracy of anatomical uncertainty compared to the previous models and does not depend on the acquisition of the weekly CTs during the treatment. We also provided a solution to quantify the probability of an anatomical deformation. The potential of the model for anatomical robust optimisation is discussed.