Development of automated region of interest selection algorithms for surface-guided radiation therapy of breast cancer.

To investigate automation of the preparation of the region of interest (ROI) for surface-guided radiotherapy (SGRT) of the whole breast with two algorithms based on contour anatomies: using the body contour, and using the breast contour. The patient dataset used for modeling consisted of 39 breast cancer patients previously treated with SGRT. The patient's anatomical structures (body and ipsilateral breast) were retrieved from the planning system, and the clinical ROI (cROI) drawn by the planners was retrieved from the SGRT system for comparison. For the body-contour-based algorithm, a convolutional neural network (MobileNet-v2) was utilized to train a synthetic human model dataset to predict body joint locations. With the body joint location knowledge, an automated ROI (aROIbody ) can be created based on: (1) the superior-inferior (S-I) borders defined by the joint locations, (2) the left-right (L-R) borders defined with 3/4 of chest width, and (3) a curation of the ROI to avoid the ipsilateral armpit. For the breast-contour-based algorithm, an aROIbreast was created by first defining the ROI in the S-I direction with the ipsilateral breast boundaries. Other steps are the same as with the body-contour-based algorithm. Among the 39 patients, 24 patients were used to fine-tune the algorithm parameters, and the remaining 15 patients were used to evaluate the quality of the aROIs against the cROIs. A blinded evaluation was performed by three SGRT expert physicists to rate the acceptability and the quality (1-10 scale) of the aROIs and cROIs, and the dice similarity coefficient (DSC) was also calculated to compare the similarity between the aROIs and cROIs. The results showed that the average acceptability was 14/15 (range: 13/15-15/15) for cROIs, 13.3/15 (range: 13/15-14/15) for aROIbody , and 14.6/15 (range: 14/15-15/15) for aROIbreast . The average quality was 7.4 ± 0.8 for cROIs, 8.1 ± 1.2 for aROIbody , and 8.2 ± 0.9 for aROIbreast . The DSC with cROIs was 0.81 ± 0.06 for aROIbody , and 0.83 ± 0.04 for aROIbreast . The ROI creation time was ∼120 s for clinical, 1.3 s for aROIbody , and 1.2 s for aROIbreast . The proposed automated algorithms can improve the ROI compliance with the SGRT protocol, with a shortened preparation time. It is ready to be integrated into the clinical workflow for automated ROI preparation.

Geometric evaluation and quantifying dosimetric impact of diverse deformable image registration algorithms on abdomen images with biomechanically modeled deformations.

PURPOSE: Deformable image registration (DIR) has been increasingly used in radiation therapy (RT). The accuracy of DIR algorithms and how it impacts on the RT plan dosimetrically were examined in our study for abdominal sites using biomechanically modeled deformations. METHODS: Five pancreatic cancer patients were enrolled in this study. Following the guidelines of AAPM TG-132, a patient-specific quality assurance (QA) workflow was developed to evaluate DIR for the abdomen using the TG-132 recommended virtual simulation software ImSimQA (Shrewsbury, UK). First, the planning CT was deformed to simulate respiratory motion using the embedded biomechanical model in ImSimQA. Additionally, 5 mm translational motion was added to the stomach, duodenum, and small bowel. The original planning CT and the deformed CT were then imported into Eclipse and MIM to perform DIR. The output displacement vector fields (DVFs) were compared with the ground truth from ImSimQA. Furthermore, the original treatment plan was recalculated on the ground-truth deformed CT and the deformed CT (with Eclipse and MIM DVF). The dose errors were calculated on a voxel-to-voxel basis. RESULTS: Data analysis comparing DVF from Eclipse versus MIM show the average mean DVF magnitude errors of 2.8 ± 1.0  versus 1.1 ± 0.7 mm for stomach and duodenum, 5.2 ± 4.0  versus 2.5 ± 1.0 mm for small bowel, and 4.8 ± 4.1  versus 2.7 ± 1.1 mm for the gross tumor volume (GTV), respectively, across all patients. The mean dose error on stomach+duodenum and small bowel were 2.3 ± 0.6% for Eclipse, and 1.0 ± 0.3% for MIM. As the DIR magnitude error increases, the dose error range increase, for both Eclipse and MIM. CONCLUSION: In our study, an initial assessment was conducted to evaluate the accuracy of DIR and its dosimetric impact on radiotherapy. A patient-specific DIR QA workflow was developed for pancreatic cancer patients. This workflow exhibits promising potential for future implementation as a clinical workflow.

Standardization of scan protocols for RT CT simulator from different vendors using quantitative image quality technique.

OBJECTIVE: To investigate the feasibility of standardizing RT simulation CT scanner protocols between vendors using target-based image quality (IQ) metrics. METHOD AND MATERIALS: A systematic assessment process in phantom was developed to standardize clinical scan protocols for scanners from different vendors following these steps: (a) images were acquired by varying CTDIvol and using an iterative reconstruction (IR) method (IR: iDose and model-based iterative reconstruction [IMR] of CTp-Philips Big Bore scanner, SAFIRE of CTs-Siemens biograph PETCT scanner), (b) CT exams were classified into body and brain protocols, (c) the rescaled noise power spectrum (NPS) was calculated, (d) quantified the IQ change due to varied CTDIvol and IR, and (e) matched the IR strength level. IQ metrics included noise and texture from NPS, contrast, and contrast-to-noise ratio (CNR), low contrast detectability (d'). Area under curve (AUC) of the receiver operation characteristic curve of d' was calculated and compared. RESULTS: The level of change in the IQ ratio was significant (>0.6) when using IMR. The IQ ratio change was relatively low to moderate when using either iDose in CTp (0.1-0.5) or SAFIRE in CTs (0.1-0.6). SAFIRE-2 in CTs showed a closer match to the reference body protocol when compared to iDose-3 in CTp. In the brain protocol, iDose-3 in CTp could be matched to the low to moderate level of SAFIRE in CTs. The AUC of d' was highest when using IMR in CTp with lower CTDIvol, and SAFIRE in CTs performed better than iDose in CTp CONCLUSION: It is possible to use target-based IQ metrics to evaluate the performance of the system and operations across various scanners in a phantom. This can serve as an initial reference to convert clinical scanned protocols from one CT simulation scanner to another.

Reproducibility of chestwall and heart position using surface-guided versus RPM-guided DIBH radiotherapy for left breast cancer.

This study compared the reproducibility of chestwall and heart position using surface-guided versus RPM (real-time position management)-guided deep inspiration breath hold (DIBH) radiotherapy for left sided breast cancer. Forty DIBH patients under either surface-guided radiotherapy (SGRT) or RPM guidance were studied. For patients treated with tangential fields, reproducibility was measured as the displacements in central lung distance (CLD) and heart shadow to field edge distance (HFD) between pretreatment MV (megavoltage) images and planning DRRs (digitally reconstructed radiographs). For patients treated with volumetric modulated arc therapy (VMAT), sternum to isocenter (ISO) distance (StID), spine to rib edge distance (SpRD), and heart shadow to central axis (CAX) distance (HCD) between pretreatment kV images and planning DRRs were measured. These displacements were compared between SGRT and RPM-guided DIBH. In tangential patients, the mean absolute displacements of SGRT versus RPM guidance were 0.19 versus 0.23 cm in CLD, and 0.33 versus 0.62 cm in HFD. With respect to planning DRR, heart appeared closer to the field edge by 0.04 cm with surface imaging versus 0.62 cm with RPM. In VMAT patients, the displacements of surface imaging versus RPM guidance were 0.21 versus 0.15 cm in StID, 0.24 versus 0.19 cm in SpRD, and 0.72 versus 0.41 cm in HCD. Heart appeared 0.41 cm further away from CAX with surface imaging, whereas 0.10 cm closer to field CAX with RPM. None of the differences between surface imaging and RPM guidance was statistically significant. In conclusion, the displacements of chestwall were small and were comparable with SGRT- or RPM-guided DIBH. The position deviations of heart were larger than those of chestwall with SGRT or RPM. Although none of the differences between SGRT and RPM guidance were statistically significant, there was a trend that the position deviations of heart were smaller and more favorable with SGRT than with RPM guidance in tangential patients.

Comparison of setup accuracy of optical surface image versus orthogonal x-ray images for VMAT of the left breast using deep-inspiration breath-hold.

To compare the setup accuracy of optical surface image (OSI) versus orthogonal x-ray images (2DkV) using cone beam computed tomography (CBCT) as ground truth for radiotherapy of left breast cancer in deep-inspiration breath-hold (DIBH). Ten left breast DIBH patients treated with volumetric modulated arc therapy (VMAT) were studied retrospectively. OSI, 2DkV, and CBCT were acquired weekly at treatment setup. OSI, 2DkV, and CBCT were registered to planning CT or planning DRR based on a breast surface region of interest (ROI), bony anatomy (chestwall and sternum), and both bony anatomy and breast surface, respectively. These registrations provided couch shifts for each imaging system. The setup errors, or the difference in couch shifts between OSI and CBCT were compared to those between 2DkV and CBCT. A second OSI was acquired during last beam delivery to evaluate intrafraction motion. The median absolute setup errors were (0.21, 0.27, 0.23 cm, 0.6°, 1.3°, 1.0°) for OSI, and (0.26, 0.24, 0.18 cm, 0.9°, 1.0°, 0.6°) for 2DkV in vertical, longitudinal and lateral translations, and in rotation, roll and pitch, respectively. None of the setup errors was significantly different between OSI and 2DkV. For both systems, the systematic and random setup errors were ≤0.6 cm and ≤1.5° in all directions. Nevertheless, larger setup errors were observed in some sessions in both systems. There was no correlation between OSI and CBCT whereas there was modest correlation between 2DkV and CBCT. The intrafraction motion in DIBH detected by OSI was small with median absolute translations <0.2 cm, and rotations ≤0.4°. Though OSI showed comparable and small setup errors as 2DkV, it showed no correlation with CBCT. We concluded that to achieve accurate setup for both bony anatomy and breast surface, daily 2DkV can't be omitted following OSI for left breast patients treated with DIBH VMAT.

Automated tracking of morphologic changes in weekly magnetic resonance imaging during head and neck radiotherapy.

BACKGROUND AND PURPOSE: Anatomic changes during head and neck radiotherapy can impact dose delivery, necessitate adaptive replanning, and indicate patient-specific response to treatment. We have developed an automated system to track these changes through longitudinal MRI scans to aid identification and clinical intervention. The purpose of this article is to describe this tracking system and present results from an initial cohort of patients. MATERIALS AND METHODS: The Automated Watchdog in Adaptive Radiotherapy Environment (AWARE) was developed to process longitudinal MRI data for radiotherapy patients. AWARE automatically identifies and collects weekly scans, propagates radiotherapy planning structures, computes structure changes over time, and reports important trends to the clinical team. AWARE also incorporates manual structure review and revision from clinical experts and dynamically updates tracking statistics when necessary. AWARE was applied to patients receiving weekly T2-weighted MRI scans during head and neck radiotherapy. Changes in nodal gross tumor volume (GTV) and parotid gland delineations were tracked over time to assess changes during treatment and identify early indicators of treatment response. RESULTS: N = 91 patients were tracked and analyzed in this study. Nodal GTVs and parotids both shrunk considerably throughout treatment (-9.7 ± 7.7% and -3.7 ± 3.3% per week, respectively). Ipsilateral parotids shrunk significantly faster than contralateral (-4.3 ± 3.1% vs. -2.9 ± 3.3% per week, p = 0.005) and increased in distance from GTVs over time (+2.7 ± 7.2% per week, p < 1 × 10-5 ). Automatic structure propagations agreed well with manual revisions (Dice = 0.88 ± 0.09 for parotids and 0.80 ± 0.15 for GTVs), but for GTVs the agreement degraded 4-5 weeks after the start of treatment. Changes in GTV volume observed by AWARE as early as one week into treatment were predictive of large changes later in the course (AUC = 0.79). CONCLUSION: AWARE automatically identified longitudinal changes in GTV and parotid volumes during radiotherapy. Results suggest that this system may be useful for identifying rapidly responding patients as early as one week into treatment.

Intrafractional accuracy and efficiency of a surface imaging system for deep inspiration breath hold during ablative gastrointestinal cancer treatment.

PURPOSE: Beam gating with deep inspiration breath hold (DIBH) usually depends on some external surrogate to infer internal target movement, and the exact internal movement is unknown. In this study, we tracked internal targets and characterized residual motion during DIBH treatment, guided by a surface imaging system, for gastrointestinal cancer. We also report statistics on treatment time. METHODS AND MATERIALS: We included 14 gastrointestinal cancer patients treated with surface imaging-guided DIBH volumetrically modulated arc therapy, each with at least one radiopaque marker implanted near or within the target. They were treated in 25, 15, or 10 fractions. Thirteen patients received treatment for pancreatic cancer, and one underwent separate treatments for two liver metastases. The surface imaging system monitored a three-dimensional surface with ± 3 mm translation and ± 3° rotation threshold. During delivery, a kilovolt image was automatically taken every 20° or 40° gantry rotation, and the internal marker was identified from the image. The displacement and residual motion of the markers were calculated. To analyze the treatment efficiency, the treatment time of each fraction was obtained from the imaging and treatment timestamps in the record and verify system. RESULTS: Although the external surface was monitored and limited to ± 3 mm and ± 3°, significant residual internal target movement was observed in some patients. The range of residual motion was 3-21 mm. The average displacement for this cohort was 0-3 mm. In 19% of the analyzed images, the magnitude of the instantaneous displacement was > 5 mm. The mean treatment time was 17 min with a standard deviation of 4 min. CONCLUSIONS: Precaution is needed when applying surface image guidance for gastrointestinal cancer treatment. Using it as a solo DIBH technique is discouraged when the correlation between internal anatomy and patient surface is limited. Real-time radiographic verification is critical for safe treatments.

Assessment of the use of different imaging and delivery techniques for cranial treatments on the Halcyon linac.

PURPOSE: In this work, we investigated the effect on the workflow and setup accuracy of using surface guided radiation therapy (SGRT) for patient setup, megavoltage cone beam CT (MVCBCT) or kilovoltage cone beam CT (kVCBCT) for imaging and fixed IMRT or volumetric-modulated arc therapy (VMAT) for treatment delivery with the Halcyon linac. METHODS: We performed a retrospective investigation of 272 treatment fractions, using three different workflows. The first and second workflows used MVCBCT and fixed IMRT for imaging and treatment delivery, and the second one also used SGRT for patient setup. The third workflow used SGRT for setup, kVCBCT for imaging and VMAT for delivery. Workflows were evaluated by comparing the number of fractions requiring repeated imaging acquisitions and the time required for setup, imaging and treatment delivery. Setup position accuracy was assessed by comparing the daily kV- or MV- CBCT with the planning CT and measuring the residual rotational errors for pitch, yaw and roll angles. RESULTS: Without the use of SGRT, the imaging fields were delivered more than once on 11.1% of the fractions, while re-imaging was necessary in 5.5% of the fractions using SGRT. The total treatment time, including setup, imaging, and delivery, for the three workflows was 531 ± 157 s, 503 ± 130 s and 457 ± 91 s, respectively. A statistically significant difference was observed when comparing the third workflow with the first two. The total residual rotational errors were 1.96 ± 1.29°, 1.28 ± 0.67° and 1.22 ± 0.76° and statistically significant differences were observed when comparing workflows with and without SGRT. CONCLUSIONS: The use of SGRT allowed for a reduction of re-imaging during patient setup and improved patient position accuracy by reducing residual rotational errors. A reduction in treatment time using kVCBCT with SGRT was observed. The most efficient workflow was the one including kVCBCT and SGRT for setup and VMAT for delivery.

Intrafractional relationship changes between an external breathing signal and fiducial marker positions in pancreatic cancer patients.

BACKGROUND AND PURPOSE: The purpose of this study of pancreatic cancer patients treated with respiratory-guided stereotactic body radiotherapy (SBRT) on a standard linac was to investigate (a) the intrafractional relationship change (IRC) between a breathing signal and the tumor position, (b) the impact of IRC on the delivered dose, and (c) potential IRC predictors. MATERIALS AND METHODS: We retrospectively investigated 10 pancreatic cancer patients with 2-4 implanted fiducial markers in the tumor treated with SBRT. Fluoroscopic images were acquired before and after treatment delivery simultaneously with the abdominal breathing motion. We quantified the IRC as the change in fiducial location for a given breathing amplitude in the left-right (LR), anterior-posterior (AP), and superior-inferior (SI) directions from before to after treatment delivery. The treatment plans were re-calculated after changing the isocenter coordinates according to the IRCs. Four treatment- or patient-related factors were investigated as potential predictors for IRC using linear models. RESULTS: The average (±1 SD) absolute IRCs in the LR, AP, and SI directions were 1.2 ± 1.2 mm, 0.7 ± 0.7 mm, and 1.1 ± 0.8 mm, respectively. The average 3D IRC was 2.0 ± 1.3 mm (range: 0.4-5.3 mm) for a median treatment delivery time of 8.5 min (range: 5.7-19.9 min; n = 31 fractions). The dose coverage of the internal target volume (ITV) decreased by more than 3% points in three of 31 fractions. In those cases, the 3D IRC had been larger than 4.3 mm. The 3D IRC was found to correlate with changes in the minimum breathing amplitude during treatment delivery. CONCLUSION: On average, 2 mm of treatment delivery accuracy was lost due to IRC. Periodical intrafractional imaging is needed to safely deliver respiratory-guided SBRT.

A survey of surface imaging use in radiation oncology in the United States.

Surface imaging (SI) has been rapidly integrated into radiotherapy clinics across the country without specific guidelines and recommendations on its commissioning and use aside from vendor-provided information. A survey was created under the auspices of AAPM TG-302 to assess the current status of SI to identify if there is need for formal guidance. The survey was designed to determine the institutional setting of responders, availability and length of its use, commissioning procedures, and clinical applications. This survey was created in REDCap, and approved as IRB exempt to collect anonymized data. Questions were reviewed by multiple physicists to ensure concept validity and piloted by a small group of independent physicists to ensure process validity. All full members of AAPM self-identified as "therapy" or "other" were sent the survey link by email. The survey was active from February to March 2018. Of 3677 members successfully contacted, 439 completed responses; the summary of these responses provides insight on current surface imaging clinical practices, though they should not be assumed to be representative of radiation oncology as a whole. Results showed that 53.3% of respondents have SI in their clinics, mostly in treatment rooms, rarely in simulation rooms. Half of those without SI plan on purchasing it within 3 years. Over 10% have SI but do not use it clinically, 36.8% classify themselves as "expert" users, and 85.5% agreed/strongly agreed that SI guidelines are needed. Initial positioning with SI is most common for breast/chestwall and SRS/SBRT treatments, least common for pediatrics. Use of SI for intra-fraction monitoring follows a similar distribution. Gating with SI is most prevalent for breast/chestwall (66.0%) but also used in SBRT (33.0%), and non-SBRT lung/abdomen (<30%) treatments. SI is a rapidly growing technology in the field with widespread use for several anatomic sites. Guidelines and recommendations on commissioning and clinical use are warranted.

Dosimetric study of the plan quality and dose to organs at risk on tangential breast treatments using the Halcyon linac.

PURPOSE: To investigate the plan quality and doses to the heart, contralateral breast (CB), ipsilateral lung (IL), and contralateral lung (CL) in tangential breast treatments using the Halcyon linac with megavoltage setup fields. METHODS: Radiotherapy treatment plans with tangential beams from 25 breast cancer patients previously treated on a C-arm linac were replanned for Halcyon. Thirteen corresponded to right-sided breasts and 12 to left-sided breasts, all with a dose prescription of 50 Gy in 25 fractions. Plans were created with the following setup imaging techniques: low-dose (LD) MVCBCT, high-quality (HQ) MVCBCT, LD-MV and HQ-MV pairs and the imaging dose was included in the plans. Plan quality metric values for the lumpectomy cavity, whole-breast and doses to the organs at risk (OARs) were measured and compared with those from the original plans. RESULTS: No significant differences in plan quality were observed between the original and Halcyon plans. An increase in the mean dose (Mean) for all the organs was observed for the Halcyon plans. For right-sided plans, the accumulated Mean over the 25 fractions in the C-arm plans was 0.4 ± 0.3, 0.2 ± 0.2, 5.4 ± 1.3, and 0.1 ± 0.1 Gy for the heart, CB, IL, and CL, respectively, while values in the MVCBCT-LD Halcyon plans were 1.2 ± 0.2, 0.6 ± 0.1, 6.5 ± 1.4, and 0.4 ± 0.1 Gy, respectively. For left-sided treatments, Mean in the original plans was 0.9 ± 0.2, 0.1 ± 0.0, 4.2 ± 1.2, and 0.0 ± 0.0 Gy, while for the MVCBCT-LD Halcyon plans values were 1.9 ± 0.2, 0.6 ± 0.2, 5.1 ± 1.2, and 0.5 ± 0.2 Gy, respectively. CONCLUSIONS: Plan quality for breast treatments using Halcyon is similar to the quality for a 6 MV, C-arm plan. For treatments using megavoltage setup fields, the dose contribution to OARs from the imaging fields can be equal or higher than the dose from treatment fields.

Automatic patient positioning and gating window settings in respiratory-gated stereotactic body radiation therapy for pancreatic cancer using fluoroscopic imaging.

Before treatment delivery of respiratory-gated radiation therapy (RT) in patients with implanted fiducials, both the patient position and the gating window thresholds must be set. In linac-based RT, this is currently done manually and setup accuracy will therefore be dependent on the skill of the user. In this study, we present an automatic method for finding the patient position and the gating window thresholds. Our method uses sequentially acquired anterior-posterior (AP) and lateral fluoroscopic imaging with simultaneous breathing amplitude monitoring and intends to reach 100% gating accuracy while keeping the duty cycle as high as possible. We retrospectively compared clinically used setups to the automatic setups by our method in five pancreatic cancer patients treated with hypofractionated RT. In 15 investigated fractions, the average (±standard deviation) differences between the clinical and automatic setups were -0.4 ± 0.8 mm, -1.0 ± 1.1 mm, and 1.8 ± 1.3 mm in the left-right (LR), the AP, and the superior-inferior (SI) direction, respectively. For the clinical setups, typical interfractional setup variations were 1-2 mm in the LR and AP directions, and 2-3 mm in the SI direction. Using the automatic method, the duty cycle could be improved in six fractions, in four fractions the duty cycle had to be lowered to improve gating accuracy, and in five fractions both duty cycle and gating accuracy could be improved. Our automatic method has the potential to increase accuracy and decrease user dependence of setup for patients with implanted fiducials treated with respiratory-gated RT. After fluoroscopic image acquisition, the calculated patient shifts and gating window thresholds are calculated in 1-2 s. The method gives the user the possibility to evaluate the effect of different patient positions and gating window thresholds on gating accuracy and duty cycle. If deemed necessary, it can be used at any time during treatment delivery.

Clinical efficacy and safety of surface imaging guided radiosurgery (SIG-RS) in the treatment of benign skull base tumors.

Frameless, surface imaging guided radiosurgery (SIG-RS) is a novel platform for stereotactic radiosurgery (SRS) wherein patient positioning is monitored in real-time through infra-red camera tracking of facial topography. Here we describe our initial clinical experience with SIG-RS for the treatment of benign neoplasms of the skull base. We identified 48 patients with benign skull base tumors consecutively treated with SIG-RS at a single institution between 2009 and 2011. Patients were diagnosed with meningioma (n = 22), vestibular schwannoma (n = 20), or nonfunctional pituitary adenoma (n = 6). Local control and treatment-related toxicity were retrospectively assessed. Median follow-up was 65 months (range 61-72 months). Prescription doses were 12-13 Gy in a single fraction (n = 18), 8 Gy × 3 fractions (n = 6), and 5 Gy × 5 fractions (n = 24). Actuarial tumor control rate at 5 years was 98%. No grade ≥3 treatment-related toxicity was observed. Grade ≤2 toxicity was associated with symptomatic lesions (p = 0.049) and single fraction treatment (p = 0.005). SIG-RS for benign skull base tumors produces clinical outcomes comparable to conventional frame-based SRS techniques while enhancing patient comfort.

Dosimetric assessment of brass mesh bolus for postmastectomy photon radiotherapy.

Brass mesh bolus has been shown to be an acceptable substitute for tissue-equivalent bolus to increase superficial dose for chest wall tangent photon radiotherapy. This work investigated the increase in surface dose, the change in the dose at depth, and the safety implications of higher energy photon beams when using brass mesh bolus for postmastectomy chest wall radiotherapy. A photon tangent plan was delivered to a thorax phantom, and the superficial dose ranged from 40%-72% of prescription dose with no bolus. The surface dose increased to 75%-110% of prescription dose with brass mesh bolus and 85%-109% of prescription dose with tissue-equivalent bolus. It was also found that the dose at depth when using brass mesh bolus is comparable to that measured with no bolus for en face and oblique incidence. Monte Carlo calculations were used to assess the photoneutron production from brass mesh bolus used with 15 MV and 24 MV photon beams. The effective dose from photoneutrons was approximated and found to be relatively small, yet not negligible. Activation products generated by these photoneutrons, the surface dose rate due to the activation products, and the half-life of the activation products were also considered in this work. The authors conclude that brass mesh bolus is a reasonable alternative to tissue-equivalent bolus, and it may be used with high-energy beam; but one should be aware of the potential increased effective dose to staff and patients due to the activation products produced by photoneutrons.

SPARSE: Seed Point Auto-Generation for Random Walks Segmentation Enhancement in medical inhomogeneous targets delineation of morphological MR and CT images.

In medical image processing, robust segmentation of inhomogeneous targets is a challenging problem. Because of the complexity and diversity in medical images, the commonly used semiautomatic segmentation algorithms usually fail in the segmentation of inhomogeneous objects. In this study, we propose a novel algorithm imbedded with a seed point autogeneration for random walks segmentation enhancement, namely SPARSE, for better segmentation of inhomogeneous objects. With a few user-labeled points, SPARSE is able to generate extended seed points by estimating the probability of each voxel with respect to the labels. The random walks algorithm is then applied upon the extended seed points to achieve improved segmentation result. SPARSE is implemented under the compute unified device architecture (CUDA) programming environment on graphic processing unit (GPU) hardware platform. Quantitative evaluations are performed using clinical homogeneous and inhomogeneous cases. It is found that the SPARSE can greatly decrease the sensitiveness to initial seed points in terms of location and quantity, as well as the freedom of selecting parameters in edge weighting function. The evaluation results of SPARSE also demonstrate substantial improvements in accuracy and robustness to inhomogeneous target segmentation over the original random walks algorithm.

A method for generating large datasets of organ geometries for radiotherapy treatment planning studies.

BACKGROUND: With the rapidly increasing application of adaptive radiotherapy, large datasets of organ geometries based on the patient's anatomy are desired to support clinical application or research work, such as image segmentation, re-planning, and organ deformation analysis. Sometimes only limited datasets are available in clinical practice. In this study, we propose a new method to generate large datasets of organ geometries to be utilized in adaptive radiotherapy. METHODS: Given a training dataset of organ shapes derived from daily cone-beam CT, we align them into a common coordinate frame and select one of the training surfaces as reference surface. A statistical shape model of organs was constructed, based on the establishment of point correspondence between surfaces and non-uniform rational B-spline (NURBS) representation. A principal component analysis is performed on the sampled surface points to capture the major variation modes of each organ. RESULTS: A set of principal components and their respective coefficients, which represent organ surface deformation, were obtained, and a statistical analysis of the coefficients was performed. New sets of statistically equivalent coefficients can be constructed and assigned to the principal components, resulting in a larger geometry dataset for the patient's organs. CONCLUSIONS: These generated organ geometries are realistic and statistically representative.

On-board synthetic 4D MRI generation from 4D CBCT for radiotherapy of abdominal tumors: A feasibility study.

BACKGROUND: Magnetic resonance-guided radiotherapy with an MR-guided LINAC represents potential clinical benefits in abdominal treatments due to the superior soft-tissue contrast compared to kV-based images in conventional treatment units. However, due to the high cost associated with this technology, only a few centers have access to it. As an alternative, synthetic 4D MRI generation based on artificial intelligence methods could be implemented. Nevertheless, appropriate MRI texture generation from CT images might be challenging and prone to hallucinations, compromising motion accuracy. PURPOSE: To evaluate the feasibility of on-board synthetic motion-resolved 4D MRI generation from prior 4D MRI, on-board 4D cone beam CT (CBCT) images, motion modeling information, and deep learning models using the digital anthropomorphic phantom XCAT. METHODS: The synthetic 4D MRI corresponds to phases from on-board 4D CBCT. Each synthetic MRI volume in the 4D MRI was generated by warping a reference 3D MRI (MRIref, end of expiration phase from a prior 4D MRI) with a deformation field map (DFM) determined by (I) the eigenvectors from the principal component analysis (PCA) motion-modeling of the prior 4D MRI, and (II) the corresponding eigenvalues predicted by a convolutional neural network (CNN) model using the on-board 4D CBCT images as input. The CNN was trained with 1000 deformations of one reference CT (CTref, same conditions as MRIref) generated by applying 1000 DFMs computed by randomly sampling the original eigenvalues from the prior 4D MRI PCA model. The evaluation metrics for the CNN model were root-mean-square error (RMSE) and mean absolute error (MAE). Finally, different on-board 4D-MRI generation scenarios were assessed by changing the respiratory period, the amplitude of the diaphragm, and the chest wall motion of the 4D CBCT using normalized root-mean-square error (nRMSE) and structural similarity index measure (SSIM) for image-based evaluation, and volume dice coefficient (VDC), volume percent difference (VPD), and center-of-mass shift (COMS) for contour-based evaluation of liver and target volumes. RESULTS: The RMSE and MAE values of the CNN model reported 0.012 ± 0.001 and 0.010 ± 0.001, respectively for the first eigenvalue predictions. SSIM and nRMSE were 0.96 ± 0.06 and 0.22 ± 0.08, respectively. VDC, VPD, and COMS were 0.92 ± 0.06, 3.08 ± 3.73 %, and 2.3 ± 2.1 mm, respectively, for the target volume. The more challenging synthetic 4D-MRI generation scenario was for one 4D-CBCT with increased chest wall motion amplitude, reporting SSIM and nRMSE of 0.82 and 0.51, respectively. CONCLUSIONS: On-board synthetic 4D-MRI generation based on predicting actual treatment deformation from on-board 4D-CBCT represents a method that can potentially improve the treatment-setup localization in abdominal radiotherapy treatments with a conventional kV-based LINAC.

Creating patient-specific digital phantoms with a longitudinal atlas for evaluating deformable CT-CBCT registration in adaptive lung radiotherapy.

BACKGROUND: Quality assurance of deformable image registration (DIR) is challenging because the ground truth is often unavailable. In addition, current approaches that rely on artificial transformations do not adequately resemble clinical scenarios encountered in adaptive radiotherapy. PURPOSE: We developed an atlas-based method to create a variety of patient-specific serial digital phantoms with CBCT-like image quality to assess the DIR performance for longitudinal CBCT imaging data in adaptive lung radiotherapy. METHODS: A library of deformations was created by extracting the longitudinal changes observed between a planning CT and weekly CBCT from an atlas of lung radiotherapy patients. The planning CT of an inquiry patient was first deformed by mapping the deformation pattern from a matched atlas patient, and subsequently appended with CBCT artifacts to imitate a weekly CBCT. Finally, a group of digital phantoms around an inquiry patient was produced to simulate a series of possible evolutions of tumor and adjacent normal structures. We validated the generated deformation vector fields (DVFs) to ensure numerically and physiologically realistic transformations. The proposed framework was applied to evaluate the performance of the DIR algorithm implemented in the commercial Eclipse treatment planning system in a retrospective study of eight inquiry patients. RESULTS: The generated DVFs were inverse consistent within less than 3 mm and did not exhibit unrealistic folding. The deformation patterns adequately mimicked the observed longitudinal anatomical changes of the matched atlas patients. Worse Eclipse DVF accuracy was observed in regions of low image contrast or artifacts. The structure volumes exhibiting a DVF error magnitude of equal or more than 2 mm ranged from 24.5% (spinal cord) to 69.2% (heart) and the maximum DVF error exceeded 5 mm for all structures except the spinal cord. Contour-based evaluations showed a high degree of alignment with dice similarity coefficients above 0.8 in all cases, which underestimated the overall DVF accuracy within the structures. CONCLUSIONS: It is feasible to create and augment digital phantoms based on a particular patient of interest using multiple series of deformation patterns from matched patients in an atlas. This can provide a semi-automated procedure to complement the quality assurance of CT-CBCT DIR and facilitate the clinical implementation of image-guided and adaptive radiotherapy that involve longitudinal CBCT imaging studies.

Radiotherapy Dose Accumulation Routine (RADAR)- A novel dose accumulation script with built-in uncertainty.

PURPOSE: To incorporate uncertainty into dose accumulation for reirradiation. METHODS: The RADAR script for the Eclipse treatment planning system (Varian Medical Systems, Palo Alto, CA) is described and the voxel-wise ellipsoid search algorithm is introduced as a means of incorporating uncertainty. RADAR is first demonstrated on a test patient reirradiated to the spine illustrating the effect of the uncertainty algorithm. A summary of initial evaluation testing by 11 users, each of whom ran a separate spine reirradiation case, follows. Finally, RADAR run times are reported for different conditions. RESULTS: In the demonstration case in which a 3mm ellipsoid search was used, maximum RADAR 2 Gy equivalent (EQD2) accumulated spinal cord dose increased from 7244 cGy to 12689 cGy because the ellipsoid search pulled dose from closer to the adjacent target structure. When the ellipsoid search was restricted to voxels within the spinal cord, the maximum accumulated cord dose was reduced to 6523 cGy and did not exceed the sum of the maximum EQD2 spinal cord doses of the individual plans (6730 cGy). In the evaluation cases, the RADAR EQD2 maximum dose for the spinal cord increased an average of 31.6% with uncertainty applied compared to a conventional dose accumulation and decreased an average of 16.7% compared to a conventional dose accumulation when the uncertainty calculation was restricted to voxels within the spinal cord. RADAR run times depend on the number of plans being added and the type of uncertainty being used. CONCLUSION: RADAR offers a novel way to directly account for uncertainty in dose accumulation by means of a voxel-wise ellipsoid search algorithm. EQD2 dose accumulation with and without dose discounts is also available.

Deep learning-based target decomposition for markerless lung tumor tracking in radiotherapy.

BACKGROUND: In radiotherapy, real-time tumor tracking can verify tumor position during beam delivery, guide the radiation beam to target the tumor, and reduce the chance of a geometric miss. Markerless kV x-ray image-based tumor tracking is challenging due to the low tumor visibility caused by tumor-obscuring structures. Developing a new method to enhance tumor visibility for real-time tumor tracking is essential. PURPOSE: To introduce a novel method for markerless kV image-based tracking of lung tumors via deep learning-based target decomposition. METHODS: We utilized a conditional Generative Adversarial Network (cGAN), known as Pix2Pix, to build a patient-specific model and generate the synthetic decomposed target image (sDTI) to enhance tumor visibility on the real-time kV projection images acquired by the onboard kV imager equipped on modern linear accelerators. We used 4DCT simulation images to generate the digitally reconstructed radiograph (DRR) and DTI image pairs for model training. We augmented the training dataset by randomly shifting the 4DCT in the superior-inferior, anterior-posterior, and left-right directions during the DRR and DTI generation process. We performed real-time 2D tumor tracking via template matching between the DTI generated from the CT simulation and the sDTI generated from the real-time kV projection images. We validated the proposed method using nine patients' datasets with implanted beacons near the tumor. RESULTS: The sDTI can effectively improve the image contrast around the lung tumors on the kV projection images for the nine patients. With the beacon motion as ground truth, the tracking errors were on average 0.8 ± 0.7 mm in the superior-inferior (SI) direction and 0.9 ± 0.8 mm in the in-plane left-right (IPLR) direction. The percentage of successful tracking, defined as a tracking error less than 2 mm in the SI direction, is 92.2% on the 4312 tested images. The patient-specific model took approximately 12 h to train. During testing, it took approximately 35 ms to generate one sDTI, and 13 ms to perform the tumor tracking using template matching. CONCLUSIONS: Our method offers the potential solution for nearly real-time markerless lung tumor tracking. It achieved a high level of accuracy and an impressive tracking rate. Further development of 3D lung tumor tracking is warranted.