Assessment of intrafractional prostate motion and its dosimetric impact in MRI-guided online adaptive radiotherapy with gating.

PURPOSE: This study aimed to evaluate the intrafractional prostate motion captured during gated magnetic resonance imaging (MRI)-guided online adaptive radiotherapy for prostate cancer and analyze its impact on the delivered dose as well as the effect of gating. METHODS: Sagittal 2D cine-MRI scans were acquired at 4 Hz during treatment at a ViewRay MRIdian (ViewRay Inc., Oakwood Village, OH, USA) MR linac. Prostate shifts in anterior-posterior (AP) and superior-inferior (SI) directions were extracted separately. Using the static dose cloud approximation, the planned fractional dose was shifted according to the 2D gated motion (residual motion in gating window) to estimate the delivered dose by superimposing and averaging the shifted dose volumes. The dose of a hypothetical non-gated delivery was reconstructed similarly using the non-gated motion. For the clinical target volume (CTV), rectum, and bladder, dose-volume histogram parameters of the planned and reconstructed doses were compared. RESULTS: In total, 174 fractions (15.7 h of cine-MRI) from 10 patients were evaluated. The average (±1 σ) non-gated prostate motion was 0.6 ± 1.0 mm in the AP and 0.0 ± 0.6 mm in the SI direction with respect to the centroid position of the gating boundary. 95% of the shifts were within [-3.5, 2.7] mm in the AP and [-2.9, 3.2] mm in the SI direction. For the gated treatment and averaged over all fractions, CTV D98% decreased by less than 2% for all patients. The rectum and the bladder D2% increased by less than 3% and 0.5%, respectively. Doses reconstructed for gated and non-gated delivery were similar for most fractions. CONCLUSION: A pipeline for extraction of prostate motion during gated MRI-guided radiotherapy based on 2D cine-MRI was implemented. The 2D motion data enabled an approximate estimation of the delivered dose. For the majority of fractions, the benefit of gating was negligible, and clinical dosimetric constraints were met, indicating safety of the currently adopted gated MRI-guided treatment workflow.

Accuracy of low-dose proton CT image registration for pretreatment alignment verification in reference to planning proton CT.

PURPOSE: Proton CT (pCT) has the ability to reduce inherent uncertainties in proton treatment by directly measuring the relative proton stopping power with respect to water, thereby avoiding the uncertain conversion of X-ray CT Hounsfield unit to relative stopping power and the deleterious effect of X- ray CT artifacts. The purpose of this work was to further evaluate the potential of pCT for pretreatment positioning using experimental pCT data of a head phantom. METHODS: The performance of a 3D image registration algorithm was tested with pCT reconstructions of a pediatric head phantom. A planning pCT simulation scan of the phantom was obtained with 200 MeV protons and reconstructed with a 3D filtered back projection (FBP) algorithm followed by iterative reconstruction and a representative pretreatment pCT scan was reconstructed with FBP only to save reconstruction time. The pretreatment pCT scan was rigidly transformed by prescribing random errors with six degrees of freedom or deformed by the deformation field derived from a head and neck cancer patient to the pretreatment pCT reconstruction, respectively. After applying the rigid or deformable image registration algorithm to retrieve the original pCT image before transformation, the accuracy of the registration was assessed. To simulate very low-dose imaging for patient setup, the proton CT images were reconstructed with 100%, 50%, 25%, and 12.5% of the total number of histories of the original planning pCT simulation scan, respectively. RESULTS: The residual errors in image registration were lower than 1 mm and 1° of magnitude regardless of the anatomic directions and imaging dose. The mean residual errors ranges found for rigid image registration were from -0.29 ± 0.09 to 0.51 ± 0.50 mm for translations and from -0.05 ± 0.13 to 0.08 ± 0.08 degrees for rotations. The percentages of sub-millimetric errors found, for deformable image registration, were between 63.5% and 100%. CONCLUSION: This experimental head phantom study demonstrated the potential of low-dose pCT imaging for 3D image registration. Further work is needed to confirm the value pCT for pretreatment image-guided proton therapy.

Target position reproducibility in left-breast irradiation with deep inspiration breath-hold using multiple optical surface control points.

The aim of this study was to investigate the use of 3D optical localization of multiple surface control points for deep inspiration breath-hold (DIBH) guidance in left-breast radiotherapy treatments. Ten left-breast cancer patients underwent whole-breast DIBH radiotherapy controlled by the Real-time Position Management (RPM) system. The reproducibility of the tumor bed (i.e., target) was assessed by the position of implanted clips, acquired through in-room kV imaging. Six to eight passive fiducials were positioned on the patients' thoraco-abdominal surface and localized intrafractionally by means of an infrared 3D optical tracking system. The point-based registration between treatment and planning fiducials coordinates was applied to estimate the interfraction variations in patients' breathing baseline and to improve target reproducibility. The RPM-based DIBH control resulted in a 3D error in target reproducibility of 5.8 ± 3.4 mm (median value ± interquartile range) across all patients. The reproducibility errors proved correlated with the interfraction baseline variations, which reached 7.7 mm for the single patient. The contribution of surface fiducials registration allowed a statistically significant reduction (p < 0.05) in target localization errors, measuring 3.4 ± 1.7 mm in 3D. The 3D optical monitoring of multiple surface control points may help to optimize the use of the RPM system for improving target reproducibility in left-breast DIBH irradiation, providing insights on breathing baseline variations and increasing the robustness of external surrogates for DIBH guidance.

MRI quantification of pancreas motion as a function of patient setup for particle therapy -a preliminary study.

Particle therapy (PT) has shown positive therapeutic results in local control of locally advanced pancreatic lesions. PT effectiveness is highly influenced by target localization accuracy both in space, since the pancreas is located in proximity to radiosensitive vital organs, and in time as it is subject to substantial breathing-related motion. The purpose of this preliminary study was to quantify pancreas range of motion under typical PT treatment conditions. Three common immobilization devices (vacuum cushion, thermoplastic mask, and compressor belt) were evaluated on five male patients in prone and supine positions. Retrospective four-dimensional magnetic resonance imaging data were reconstructed for each condition and the pancreas was manually segmented on each of six breathing phases. A k-means algorithm was then applied on the manually segmented map in order to obtain clusters representative of the three pancreas segments: head, body, and tail. Centers of mass (COM) for the pancreas and its segments were computed, as well as their displacements with respect to a reference breathing phase (beginning exhalation). The median three-dimensional COM displacements were in the range of 3 mm. Latero-lateral and superior-inferior directions had a higher range of motion than the anterior-posterior direction. Motion analysis of the pancreas segments showed slightly lower COM displacements for the head cluster compared to the tail cluster, especially in prone position. Statistically significant differences were found within patients among the investigated setups. Hence a patient-specific approach, rather than a general strategy, is suggested to define the optimal treatment setup in the frame of a millimeter positioning accuracy.

Optimal marker placement in hadrontherapy: intelligent optimization strategies with augmented Lagrangian pattern search.

PURPOSE: In high precision photon radiotherapy and in hadrontherapy, it is crucial to minimize the occurrence of geometrical deviations with respect to the treatment plan in each treatment session. To this end, point-based infrared (IR) optical tracking for patient set-up quality assessment is performed. Such tracking depends on external fiducial points placement. The main purpose of our work is to propose a new algorithm based on simulated annealing and augmented Lagrangian pattern search (SAPS), which is able to take into account prior knowledge, such as spatial constraints, during the optimization process. MATERIAL AND METHODS: The SAPS algorithm was tested on data related to head and neck and pelvic cancer patients, and that were fitted with external surface markers for IR optical tracking applied for patient set-up preliminary correction. The integrated algorithm was tested considering optimality measures obtained with Computed Tomography (CT) images (i.e. the ratio between the so-called target registration error and fiducial registration error, TRE/FRE) and assessing the marker spatial distribution. Comparison has been performed with randomly selected marker configuration and with the GETS algorithm (Genetic Evolutionary Taboo Search), also taking into account the presence of organs at risk. RESULTS: The results obtained with SAPS highlight improvements with respect to the other approaches: (i) TRE/FRE ratio decreases; (ii) marker distribution satisfies both marker visibility and spatial constraints. We have also investigated how the TRE/FRE ratio is influenced by the number of markers, obtaining significant TRE/FRE reduction with respect to the random configurations, when a high number of markers is used. CONCLUSIONS: The SAPS algorithm is a valuable strategy for fiducial configuration optimization in IR optical tracking applied for patient set-up error detection and correction in radiation therapy, showing that taking into account prior knowledge is valuable in this optimization process. Further work will be focused on the computational optimization of the SAPS algorithm toward fast point-of-care applications.

An image-based method to synchronize cone-beam CT and optical surface tracking.

The integration of in-room X-ray imaging and optical surface tracking has gained increasing importance in the field of image guided radiotherapy (IGRT). An essential step for this integration consists of temporally synchronizing the acquisition of X-ray projections and surface data. We present an image-based method for the synchronization of cone-beam computed tomography (CBCT) and optical surface systems, which does not require the use of additional hardware. The method is based on optically tracking the motion of a component of the CBCT/gantry unit, which rotates during the acquisition of the CBCT scan. A calibration procedure was implemented to relate the position of the rotating component identified by the optical system with the time elapsed since the beginning of the CBCT scan, thus obtaining the temporal correspondence between the acquisition of X-ray projections and surface data. The accuracy of the proposed synchronization method was evaluated on a motorized moving phantom, performing eight simultaneous acquisitions with an Elekta Synergy CBCT machine and the AlignRT optical device. The median time difference between the sinusoidal peaks of phantom motion signals extracted from the synchronized CBCT and AlignRT systems ranged between -3.1 and 12.9 msec, with a maximum interquartile range of 14.4 msec. The method was also applied to clinical data acquired from seven lung cancer patients, demonstrating the potential of the proposed approach in estimating the individual and daily variations in respiratory parameters and motion correlation of internal and external structures. The presented synchronization method can be particularly useful for tumor tracking applications in extracranial radiation treatments, especially in the field of patient-specific breathing models, based on the correlation between internal tumor motion and external surface surrogates.

Reproducibility of the external surface position in left-breast DIBH radiotherapy with spirometer-based monitoring.

Deep inspiration breath hold (DIBH) in left-sided breast cancer radiotherapy treatments allows for a reduction in cardiac and pulmonary doses without compromising target coverage. The selection of the most appropriate technology for DIBH monitoring is a crucial issue. We evaluated the stability and reproducibility of DIBHs controlled by a spirometric device, by assessing the variability of the external surface position within a single DIBH (intra-DIBH) and between DIBHs performed in the same treatment session (intrafraction) or in different sessions (interfraction). The study included seven left-breast cancer patients treated with spirometer-based DIBH radiotherapy. Infrared optical tracking was used to record the 3D coordinates of seven to eleven passive markers placed on the patient's thoraco-abdominal surface during 29-43 DIBHs performed in six to eight treatment sessions. The obtained results showed displacements of the external surface between different sessions up to 6.3mm along a single direction, even at constant inspired volumes. The median value of the interfraction variability in the position of breast passive markers was 2.9 mm (range 1.9-4.8 mm) in the latero-lateral direction, 3.6 mm (range 2.2-4.6mm) in the antero-posterior direction, and 4.3mm (range 2.8-6.2 mm) in the cranio-caudal direction. There were no significant dose distribution variations for target and organs at risk with respect to the treatment plan, confirming the adequacy of the applied clinical margins (15 mm) to compensate for the measured setup uncertainties. This study demonstrates that spirometer-based control does not guarantee a stable and reproducible position of the external surface in left-breast DIBH radiotherapy, suggesting the need for more robust DIBH monitoring techniques when reduced margins and setup uncertainties are required for improving normal tissue sparing and decreasing cardiac and pulmonary toxicity.

Proton beam radiotherapy: report of the first ten patients treated at the "Centro Nazionale di Adroterapia Oncologica (CNAO)" for skull base and spine tumours.

PURPOSE: The Italian National Centre for Oncological Hadrontherapy (Centro Nazionale di Adroterapia Oncologica, CNAO), equipped with a proton and ion synchrotron, started clinical activity in September 2011. The clinical and technical characteristics of the first ten proton beam radiotherapy treatments are reported. MATERIALS AND METHODS: Ten patients, six males and four females (age range 27-73 years, median 55.5), were treated with proton beam radiotherapy. After one to two surgical procedures, seven patients received a histological diagnosis of chordoma (of the skull base in three cases, the cervical spine in one case and the sacrum in three cases) and three of low-grade chondrosarcoma (skull base). Prescribed doses were 74 GyE for chordoma and 70 GyE for chondrosarcoma at 2 GyE/fraction delivered 5 days per week. RESULTS: Treatment was well tolerated without toxicity-related interruptions. The maximal acute toxicity was grade 2, with oropharyngeal mucositis, nausea and vomiting for the skull base tumours, and grade 2 dermatitis for the sacral tumours. After 6-12 months of follow-up, no patient developed tumour progression. CONCLUSIONS: The analysis of the first ten patients treated with proton therapy at CNAO showed that this treatment was feasible and safe. Currently, patient accrual into these as well as other approved protocols is continuing, and a longer follow-up period is needed to assess tumour control and late toxicity.

Imaging error reduction in radial cine-MRI with deep learning-based intra-frame motion compensation.

OBJECTIVE: Radial cine-MRI allows for sliding window reconstruction at nearly arbitrary frame rate, promising high-speed imaging for intra-fractional motion monitoring in MRgRT. However, motion within the reconstruction window may determine the location of the reconstructed target to deviate from the true real-time position (target positioning errors), particularly in cases of fast breathing or for anatomical structures affected by the heartbeat. In this work, we present a proof-of-concept study aiming to enhance radial cine-MR imaging by implementing deep-learning-based intra-frame motion compensation techniques. Approach: A novel network (TransSin-UNet) was proposed to continuously estimate the final-position image of the target, corresponding to end of the frame acquisition. Within the radial k-space reconstruction window, the spatial-temporal dependencies among the sinogram representation of the spokes were modeled by a transformer encoder subnetwork, followed by a UNet subnetwork operating in the spatial domain for pixel-level fine-tuning. By simulating motion-dependent radial sampling with (tiny) golden angles, we generated datasets from 25 4D digital anthropomorphic lung cancer phantoms. The network was then trained and extensively evaluated across datasets characterized by varying azimuthal radial profile increments. Main Results: The method required additional 4.8ms per frame over the conventional approach involving direct image reconstruction with motion-corrupted spokes. TransSin-UNet outperformed architectures relying solely on transformer encoders or UNets across all the comparative evaluations, leading to a noticeable enhancement in image quality and target positioning accuracy. The normalized root mean-squared error (NRMSE) decreased by 50% from the initial value of 0.188 on average, whereas the mean Dice similarity coefficient (DSC) of the gross tumor volume (GTV) increased from 85.1% to 96.2% in the investigated cases. Furthermore, the final-positions of anatomical structures undergoing substantial intra-frame deformations were precisely derived. Significance: The proposed approach enables an effective intra-frame motion compensation, offering an opportunity to reduce errors in radial cine-MR imaging for real-time motion management. .

Intra-frame motion deterioration effects and deep-learning-based compensation in MR-guided radiotherapy.

BACKGROUND: Current commercially available hybrid magnetic resonance linear accelerators (MR-Linac) use 2D+t cine MR imaging to provide intra-fractional motion monitoring. However, given the limited temporal resolution of cine MR imaging, target intra-frame motion deterioration effects, resulting in effective time latency and motion artifacts in the image domain, can be appreciable, especially in the case of fast breathing. PURPOSE: The aim of this work is to investigate intra-frame motion deterioration effects in MR-guided radiotherapy (MRgRT) by simulating the motion-corrupted image acquisition, and to explore the feasibility of deep-learning-based compensation approaches, relying on the intra-frame motion information which is spatially and temporally encoded in the raw data (k-space). METHODS: An intra-frame motion model was defined to simulate motion-corrupted MR images, with 4D anthropomorphic digital phantoms being exploited to provide ground truth 2D+t cine MR sequences. A total number of 10 digital phantoms were generated for lung cancer patients, with randomly selected eight patients for training or validation and the remaining two for testing. The simulation code served as the data generator, and a dedicated motion pattern perturbation scheme was proposed to build the intra-frame motion database, where three degrees of freedom were designed to guarantee the diversity of intra-frame motion trajectories, enabling a thorough exploration in the domain of the potential anatomical structure positions. U-Nets with three types of loss functions: L1 or L2 loss defined in image or Fourier domain, referred to as NNImgLoss-L1 , NNFloss-L1 and NNL2-Loss were trained to extract information from the motion-corrupted image and used to estimate the ground truth final-position image, corresponding to the end of the acquisition. Images before and after compensation were evaluated in terms of (i) image mean-squared error (MSE) and mean absolute error (MAE), and (ii) accuracy of gross tumor volume (GTV) contouring, based on optical-flow image registration. RESULTS: Image degradation caused by intra-frame motion was observed: for a linearly and fully acquired Cartesian readout k-space trajectory, intra-frame motion resulted in an imaging latency of approximately 50% of the acquisition time; in comparison, the motion artifacts exhibited only a negligible contribution to the overall geometric errors. All three compensation models led to a decrease in image MSE/MAE and GTV position offset compared to the motion-corrupted image. In the investigated testing dataset for GTV contouring, the average dice similarity coefficients (DSC) improved from 88% to 96%, and the 95th percentile Hausdorff distance (HD95 ) dropped from 4.8 mm to 2.1 mm. Different models showed slight performance variations across different intra-frame motion amplitude categories: NNImgLoss-L1 excelled for small/medium amplitudes, whereas NNFloss-L1 demonstrated higher DSC median values at larger amplitudes. The saliency maps of the motion-corrupted image highlighted the major contribution of the later acquired k-space data, as well as the edges of the moving anatomical structures at their final positions, during the model inference stage. CONCLUSIONS: Our results demonstrate the deep-learning-based approaches have the potential to compensate for intra-frame motion by utilizing the later acquired data to drive the convergence of the earlier acquired k-space components.

Simultaneous object detection and segmentation for patient-specific markerless lung tumor tracking in simulated radiographs with deep learning.

BACKGROUND: Real-time tumor tracking is one motion management method to address motion-induced uncertainty. To date, fiducial markers are often required to reliably track lung tumors with X-ray imaging, which carries risks of complications and leads to prolonged treatment time. A markerless tracking approach is thus desirable. Deep learning-based approaches have shown promise for markerless tracking, but systematic evaluation and procedures to investigate applicability in individual cases are missing. Moreover, few efforts have been made to provide bounding box prediction and mask segmentation simultaneously, which could allow either rigid or deformable multi-leaf collimator tracking. PURPOSE: The purpose of this study was to implement a deep learning-based markerless lung tumor tracking model exploiting patient-specific training which outputs both a bounding box and a mask segmentation simultaneously. We also aimed to compare the two kinds of predictions and to implement a specific procedure to understand the feasibility of markerless tracking on individual cases. METHODS: We first trained a Retina U-Net baseline model on digitally reconstructed radiographs (DRRs) generated from a public dataset containing 875 CT scans and corresponding lung nodule annotations. Afterwards, we used an independent cohort of 97 lung patients to develop a patient-specific refinement procedure. In order to determine the optimal hyperparameters for automatic patient-specific training, we selected 13 patients for validation where the baseline model predicted a bounding box on planning CT (PCT)-DRR with intersection over union (IoU) with the ground-truth higher than 0.7. The final test set contained the remaining 84 patients with varying PCT-DRR IoU. For each testing patient, the baseline model was refined on the PCT-DRR to generate a patient-specific model, which was then tested on a separate 10-phase 4DCT-DRR to mimic the intrafraction motion during treatment. A template matching algorithm served as benchmark model. The testing results were evaluated by four metrics: the center of mass (COM) error and the Dice similarity coefficient (DSC) for segmentation masks, and the center of box (COB) error and the DSC for bounding box detections. Performance was compared to the benchmark model including statistical testing for significance. RESULTS: A PCT-DRR IoU value of 0.2 was shown to be the threshold dividing inconsistent (68%) and consistent (100%) success (defined as mean bounding box DSC > 0.6) of PS models on 4DCT-DRRs. Thirty-seven out of the eighty-four testing cases had a PCT-DRR IoU above 0.2. For these 37 cases, the mean COM error was 2.6 mm, the mean segmentation DSC was 0.78, the mean COB error was 2.7 mm, and the mean box DSC was 0.83. Including the validation cases, the model was applicable to 50 out of 97 patients when using the PCT-DRR IoU threshold of 0.2. The inference time per frame was 170 ms. The model outperformed the benchmark model on all metrics, and the comparison was significant (p < 0.001) over the 37 PCT-DRR IoU > 0.2 cases, but not over the undifferentiated 84 testing cases. CONCLUSIONS: The implemented patient-specific refinement approach based on a pre-trained baseline model was shown to be applicable to markerless tumor tracking in simulated radiographs for lung cases.

Comparison of deep learning networks for fully automated head and neck tumor delineation on multi-centric PET/CT images.

OBJECTIVES: Deep learning-based auto-segmentation of head and neck cancer (HNC) tumors is expected to have better reproducibility than manual delineation. Positron emission tomography (PET) and computed tomography (CT) are commonly used in tumor segmentation. However, current methods still face challenges in handling whole-body scans where a manual selection of a bounding box may be required. Moreover, different institutions might still apply different guidelines for tumor delineation. This study aimed at exploring the auto-localization and segmentation of HNC tumors from entire PET/CT scans and investigating the transferability of trained baseline models to external real world cohorts. METHODS: We employed 2D Retina Unet to find HNC tumors from whole-body PET/CT and utilized a regular Unet to segment the union of the tumor and involved lymph nodes. In comparison, 2D/3D Retina Unets were also implemented to localize and segment the same target in an end-to-end manner. The segmentation performance was evaluated via Dice similarity coefficient (DSC) and Hausdorff distance 95th percentile (HD95). Delineated PET/CT scans from the HECKTOR challenge were used to train the baseline models by 5-fold cross-validation. Another 271 delineated PET/CTs from three different institutions (MAASTRO, CRO, BERLIN) were used for external testing. Finally, facility-specific transfer learning was applied to investigate the improvement of segmentation performance against baseline models. RESULTS: Encouraging localization results were observed, achieving a maximum omnidirectional tumor center difference lower than 6.8 cm for external testing. The three baseline models yielded similar averaged cross-validation (CV) results with a DSC in a range of 0.71-0.75, while the averaged CV HD95 was 8.6, 10.7 and 9.8 mm for the regular Unet, 2D and 3D Retina Unets, respectively. More than a 10% drop in DSC and a 40% increase in HD95 were observed if the baseline models were tested on the three external cohorts directly. After the facility-specific training, an improvement in external testing was observed for all models. The regular Unet had the best DSC (0.70) for the MAASTRO cohort, and the best HD95 (7.8 and 7.9 mm) in the MAASTRO and CRO cohorts. The 2D Retina Unet had the best DSC (0.76 and 0.67) for the CRO and BERLIN cohorts, and the best HD95 (12.4 mm) for the BERLIN cohort. CONCLUSION: The regular Unet outperformed the other two baseline models in CV and most external testing cohorts. Facility-specific transfer learning can potentially improve HNC segmentation performance for individual institutions, where the 2D Retina Unets could achieve comparable or even better results than the regular Unet.

Reduction of cone-beam CT artifacts in a robotic CBCT device using saddle trajectories with integrated infrared tracking.

BACKGROUND: Cone beam computed tomography (CBCT) is widely used in many medical fields. However, conventional CBCT circular scans suffer from cone beam (CB) artifacts that limit the quality and reliability of the reconstructed images due to incomplete data. PURPOSE: Saddle trajectories in theory might be able to improve the CBCT image quality by providing a larger region with complete data. Therefore, we investigated the feasibility and performance of saddle trajectory CBCT scans and compared them to circular trajectory scans. METHODS: We performed circular and saddle trajectory scans using a novel robotic CBCT scanner (Mobile ImagingRing (IRm); medPhoton, Salzburg, Austria). For the saddle trajectory, the gantry executed yaw motion up to ± 10 ∘ $\pm 10^{\circ }$ using motorized wheels driving on the floor. An infrared (IR) tracking device with reflective markers was used for online geometric calibration correction (mainly floor unevenness). All images were reconstructed using penalized least-squares minimization with the conjugate gradient algorithm from RTK with 0.5 × 0.5 × 0.5 mm 3 $0.5 \times 0.5\times 0.5 \text{ mm}^3$ voxel size. A disk phantom and an Alderson phantom were scanned to assess the image quality. Results were correlated with the local incompleteness value represented by tan ( ψ ) $\tan (\psi)$ , which was calculated at each voxel as a function of the source trajectory and the voxel's 3D coordinates. We assessed the magnitude of CB artifacts using the full width half maximum (FWHM) of each disk profile in the axial center of the reconstructed images. Spatial resolution was also quantified by the modulation transfer function at 10% (MTF10). RESULTS: When using the saddle trajectory, the region without CB artifacts was increased from 43 to 190 mm in the SI direction compared to the circular trajectory. This region coincided with low values for tan ( ψ ) $\tan (\psi)$ . When tan ( ψ ) $\tan (\psi)$ was larger than 0.02, we found there was a linear relationship between the FWHM and tan ( ψ ) $\tan (\psi)$ . For the saddle, IR tracking allowed the increase of MTF10 from 0.37 to 0.98 lp/mm. CONCLUSIONS: We achieved saddle trajectory CBCT scans with a novel CBCT system combined with IR tracking. The results show that the saddle trajectory provides a larger region with reliable reconstruction compared to the circular trajectory. The proposed method can be used to evaluate other non-circular trajectories.

Real-time motion management in MRI-guided radiotherapy: Current status and AI-enabled prospects.

MRI-guided radiotherapy (MRIgRT) is a highly complex treatment modality, allowing adaptation to anatomical changes occurring from one treatment day to the other (inter-fractional), but also to motion occurring during a treatment fraction (intra-fractional). In this vision paper, we describe the different steps of intra-fractional motion management during MRIgRT, from imaging to beam adaptation, and the solutions currently available both clinically and at a research level. Furthermore, considering the latest developments in the literature, a workflow is foreseen in which motion-induced over- and/or under-dosage is compensated in 3D, with minimal impact to the radiotherapy treatment time. Considering the time constraints of real-time adaptation, a particular focus is put on artificial intelligence (AI) solutions as a fast and accurate alternative to conventional algorithms.

Patient-specific deep learning tracking framework for real-time 2D target localization in MRI-guided radiotherapy.

PURPOSE: We propose a tumor tracking framework for 2D cine MRI based on a pair of deep learning (DL) models relying on patient-specific (PS) training. METHODS AND MATERIALS: The chosen DL models are: 1) an image registration transformer and 2) an auto-segmentation convolutional neural network (CNN). We collected over 1,400,000 cine MRI frames from 219 patients treated on a 0.35 T MRI-linac plus 7,500 frames from additional 35 patients which were manually labelled and subdivided into fine-tuning, validation, and testing sets. The transformer was first trained on the unlabeled data (without segmentations). We then continued training (with segmentations) either on the fine-tuning set or for PS models based on eight randomly selected frames from the first 5 s of each patient's cine MRI. The PS auto-segmentation CNN was trained from scratch with the same eight frames for each patient, without pre-training. Furthermore, we implemented B-spline image registration as a conventional model, as well as different baselines. Output segmentations of all models were compared on the testing set using the Dice similarity coefficient (DSC), the 50% and 95% Hausdorff distance (HD50%/HD95%), and the root-mean-square-error of the target centroid in superior-inferior direction (RMSESI). RESULTS: The PS transformer and CNN significantly outperformed all other models, achieving a median (inter-quartile range) DSC of 0.92 (0.03)/0.90 (0.04), HD50% of 1.0 (0.1)/1.0 (0.4) mm, HD95% of 3.1 (1.9)/3.8 (2.0) mm and RMSESI of 0.7 (0.4)/0.9 (1.0) mm on the testing set. Their inference time was about 36/8 ms per frame and PS fine-tuning required 3 min for labelling and 8/4 min for training. The transformer was better than the CNN in 9/12 patients, the CNN better in 1/12 patients and the two PS models achieved the same performance on the remaining 2/12 testing patients. CONCLUSION: For targets in the thorax, abdomen and pelvis, we found two PS DL models to provide accurate real-time target localization during MRI-guided radiotherapy.

An adaptive fuzzy prediction model for real time tumor tracking in radiotherapy via external surrogates.

In the radiation treatment of moving targets with external surrogates, information on tumor position in real time can be extracted by using accurate correlation models. A fuzzy environment is proposed here to correlate input surrogate data with tumor motion estimates in real time. In this study, two different data clustering approaches were analyzed due to their substantial effects on the fuzzy modeler performance. Moreover, a comparative investigation was performed on two fuzzy-based and one neuro-fuzzy-based inference systems with respect to state-of-the-art models. Finally, due to the intrinsic interpatient variability in fuzzy models' performance, a model selectivity algorithm was proposed to select an adaptive fuzzy modeler on a case-by-case basis. The performance of multiple and adaptive fuzzy logic models were retrospectively tested in 20 patients treated with CyberKnife real-time tumor tracking. Final results show that activating adequate model selection of our fuzzy-based modeler can significantly reduce tumor tracking errors.

Deep inspiration breath-hold technique guided by an opto- electronic system for extracranial stereotactic treatments.

The purpose of this work was to evaluate the intrapatient tumor position reproducibility in a deep inspiration breath-hold (DIBH) technique based on two infrared optical tracking systems, ExacTrac and ELITETM, in stereotactic treatment of lung and liver lesions. After a feasibility study, the technique was applied to 15 patients. Each patient, provided with a real-time visual feedback of external optical marker displacements, underwent a full DIBH, a free-breathing (FB), and three consecutive DIBH CT-scans centered on the lesion to evaluate the tumor position reproducibility. The mean reproducibility of tumor position during repeated DIBH was 0.5 ± 0.3 mm in laterolateral (LL), 1.0 ± 0.9 mm in anteroposterior (AP), and 1.4 ± 0.9 mm in craniocaudal (CC) direction for lung lesions, and 1.0 ± 0.6 mm in LL, 1.1 ± 0.5 mm in AP, and 1.2 ± 0.4 mm in CC direction for liver lesions. Intra- and interbreath-hold reproducibility during treatment, as determined by optical markers displacements, was below 1 mm and 3 mm, respectively, in all directions for all patients. Optically-guided DIBH technique provides a simple noninvasive method to minimize breathing motion for collaborative patients. For each patient, it is important to ensure that the tumor position is reproducible with respect to the external markers configuration.

Real-time tumour tracking in particle therapy: technological developments and future perspectives.

A key challenge in radiation oncology is accurate delivery of the prescribed dose to tumours that move because of respiration. Tumour tracking involves real-time target localisation and correction of radiation beam geometry to compensate for motion. Uncertainties in tumour localisation are important in particle therapy (proton therapy, carbon-ion therapy) because charged particle beams are highly sensitive to geometrical and associated density and radiological variations in path length, which will affect the treatment plan. Target localisation and motion compensation methods applied in x-ray photon radiotherapy require careful performance assessment for clinical applications in particle therapy. In this Review, we summarise the efforts required for an application of real-time tumour tracking in particle therapy, by comparing and assessing competing strategies for time-resolved target localisation and related clinical outcomes in x-ray radiation oncology.

On the robustness of multilateration of ionoacoustic signals for localization of the Bragg peak at pre-clinical proton beam energies in water.

Objectives.The energy deposited in a medium by a pulsed proton beam results in the emission of thermoacoustic waves, also called ionoacoustics (IA). The proton beam stopping position (Bragg peak) can be retrieved from a time-of-flight analysis (ToF) of IA signals acquired at different sensor locations (multilateration). This work aimed to assess the robustness of multilateration methods in proton beams at pre-clinical energies for the development of a small animal irradiator.Approach.The accuracy of multilateration performed using different algorithms; namely, time of arrival and time difference of arrival, was investigatedin-silicofor ideal point sources in the presence of realistic uncertainties on the ToF estimation and ionoacoustic signals generated by a 20 MeV pulsed proton beam stopped in a homogeneous water phantom. The localisation accuracy was further investigated experimentally based on two different measurements with pulsed monoenergetic proton beams at energies of 20 and 22 MeV.Main results.It was found that the localisation accuracy mainly depends on the position of the acoustic detectors relative to the proton beam due to spatial variation of the error on the ToF estimation. By optimally positioning the sensors to reduce the ToF error, the Bragg peak could be locatedin-silicowith an accuracy better than 90µm (2% error). Localisation errors going up to 1 mm were observed experimentally due to inaccurate knowledge of the sensor positions and noisy ionoacoustic signals.Significance.This study gives a first overview of the implementation of different multilateration methods for ionoacoustics-based Bragg peak localisation in two- and three-dimensions at pre-clinical energies. Different sources of uncertainty were investigated, and their impact on the localisation accuracy was quantifiedin-silicoand experimentally.

Deep learning based automatic segmentation of organs-at-risk for 0.35 T MRgRT of lung tumors.

BACKGROUND AND PURPOSE: Magnetic resonance imaging guided radiotherapy (MRgRT) offers treatment plan adaptation to the anatomy of the day. In the current MRgRT workflow, this requires the time consuming and repetitive task of manual delineation of organs-at-risk (OARs), which is also prone to inter- and intra-observer variability. Therefore, deep learning autosegmentation (DLAS) is becoming increasingly attractive. No investigation of its application to OARs in thoracic magnetic resonance images (MRIs) from MRgRT has been done so far. This study aimed to fill this gap. MATERIALS AND METHODS: 122 planning MRIs from patients treated at a 0.35 T MR-Linac were retrospectively collected. Using an 80/19/23 (training/validation/test) split, individual 3D U-Nets for segmentation of the left lung, right lung, heart, aorta, spinal canal and esophagus were trained. These were compared to the clinically used contours based on Dice similarity coefficient (DSC) and Hausdorff distance (HD). They were also graded on their clinical usability by a radiation oncologist. RESULTS: Median DSC was 0.96, 0.96, 0.94, 0.90, 0.88 and 0.78 for left lung, right lung, heart, aorta, spinal canal and esophagus, respectively. Median 95th percentile values of the HD were 3.9, 5.3, 5.8, 3.0, 2.6 and 3.5 mm, respectively. The physician preferred the network generated contours over the clinical contours, deeming 85 out of 129 to not require any correction, 25 immediately usable for treatment planning, 15 requiring minor and 4 requiring major corrections. CONCLUSIONS: We trained 3D U-Nets on clinical MRI planning data which produced accurate delineations in the thoracic region. DLAS contours were preferred over the clinical contours.