Characterization of mechanical and radiation isocenter on an MR-guided radiotherapy (MRgRT) Linac.

BACKGROUND AND PURPOSE: In the emerging paradigm of stereotactic radiosurgery being proposed for MR-guided radiotherapy (MRgRT), assessment of mechanical geometric accuracy is critical for the implementation of stereotactic delivery. We benchmarked the mechanical accuracy of an MR Linac system that lacks an onboard detector/array. Our mechanical tests utilize a half beam block (HBB) geometry that takes advantage of the sensitivity of a partially occluded detector. MATERIALS AND METHODS: Mechanical tests benchmarked the couch, MLC, and gantry geometric accuracy for an MR-Linac system. An HBB technique was used to irradiate an ionization chamber profiler (ICP) array with partial occlusion of individual detectors for characterization of MLC skew, beam divergence displacement, and RT isocenter localization. The sensitivity of the partially occluded detector's ICP-X (detector width) and ICP-Y (detector length) was characterized by displacing the detector relative to radiation isocenter by 0.2 mm increments, introduced through couch motion. The accuracy of the HBB ICP technique was verified with a starshot using radiochromic film, and the reproducibility was verified on a conventional C-arm Linac and compared to Winston-Lutz. RESULTS: The sensitivity of the HBB technique as quantified through the dose difference normalized to open field as a function of displacement from RT isocenter was 6.4%/mm and 13.0%/mm for the ICP-X and ICP-Y orientation, respectively, due to the oblong detector orientation. Couch positional accuracy and sag was within ±0.1 mm. Maximum MLC positional displacement was 0.7 mm with mean MLC skew at 0.07°. The maximum beam divergence displacement was 0.03 mm. The gantry angle was within 0.1°. Independent verification of the RT isocenter localization procedure produced repeatable results. CONCLUSION: This work serves for characterizing the mechanical and geometric radiation accuracy for the foundation of an MR-guided stereotactic radiosurgery program, as demonstrated with high sensitivity and independent validation.

Characterization of imaging performance of a novel helical kVCT for use in image-guided and adaptive radiotherapy.

ClearRT helical kVCT imaging for the Radixact helical tomotherapy system recently received FDA approval and is available for clinical use. The system is intended to enhance image fidelity in radiation therapy treatment planning and delivery compared to the prior MV-based onboard imaging approach. The purpose of this work was to characterize the imaging performance of this system and compare this performance with that of clinical systems used in image-guided and/or adaptive radiotherapy (ART) or computed tomography (CT) simulation, including Radixact MVCT, TomoTherapy MVCT, Varian TrueBeam kV OBI CBCT, and the Siemens SOMATOM Definition Edge kVCT. A CT image quality phantom was scanned across clinically relevant acquisition modes for each system to evaluate image quality metrics, including noise, uniformity, contrast, spatial resolution, and CT number linearity. Similar noise levels were observed for ClearRT and Siemens Edge, whereas noise for the other systems was ∼1.5-5 times higher. Uniformity was best for Siemens Edge, whereas most scans for ClearRT exhibited a slight "cupping" or "capping" artifact. The ClearRT and Siemens Edge performed best for contrast metrics, which included low-contrast visibility and contrast-to-noise ratio evaluations. Spatial resolution was best for TrueBeam and Siemens Edge, whereas the three kVCT systems exhibited similar CT number linearity. Overall, these results provide an initial indication that ClearRT image quality is adequate for image guidance in radiotherapy and sufficient for delineating anatomic structures, thus enabling its use for ART. ClearRT also showed significant improvement over MVCT, which was previously the only onboard imaging modality available on Radixact. Although the acquisition of these scans does come at the cost of additional patient dose, reported CTDI values indicate a similar or generally reduced machine output for ClearRT compared to the other systems while maintaining comparable or improved image quality overall.

Using 4D dose accumulation to calculate organ-at-risk dose deviations from motion-synchronized liver and lung tomotherapy treatments.

Tracking systems such as Radixact Synchrony change the planned delivery of radiation during treatment to follow the target. This is typically achieved without considering the location changes of organs at risk (OARs). The goal of this work was to develop a novel 4D dose accumulation framework to quantify OAR dose deviations due to the motion and tracked treatment. The framework obtains deformation information and the target motion pattern from a four-dimensional computed tomography dataset. The helical tomotherapy treatment plan is split into 10 plans and motion correction is applied separately to the jaw pattern and multi-leaf collimator (MLC) sinogram for each phase based on the location of the target in each phase. Deformable image registration (DIR) is calculated from each phase to the references phase using a commercial algorithm, and doses are accumulated according to the DIR. The effect of motion synchronization on OAR dose was analyzed for five lung and five liver subjects by comparing planned versus synchrony-accumulated dose. The motion was compensated by an average of 1.6 cm of jaw sway and by an average of 5.7% of leaf openings modified, indicating that most of the motion compensation was from jaw sway and not MLC changes. OAR dose deviations as large as 19 Gy were observed, and for all 10 cases, dose deviations greater than 7 Gy were observed. Target dose remained relatively constant (D95% within 3 Gy), confirming that motion-synchronization achieved the goal of maintaining target dose. Dose deviations provided by the framework can be leveraged during the treatment planning process by identifying cases where OAR doses may change significantly from their planned values with respect to the critical constraints. The framework is specific to synchronized helical tomotherapy treatments, but the OAR dose deviations apply to any real-time tracking technique that does not consider location changes of OARs.

Effects of variable-width jaw motion on beam characteristics for Radixact Synchrony®.

PURPOSE: Radixact Synchrony corrects for target motion during treatment by adjusting the jaw and MLC positions in real time. As the jaws move off axis, Synchrony attempts to adjust for a loss in output due to the un-flattened 6 MV beam by increasing the jaw aperture width. The purpose of this work was to assess the impact of the variable-width aperture on delivered dose using measurements and simulations. METHODS: Longitudinal beam profile measurements were acquired using an Edge diode with static gantry. Jaw-offset peak, width, and integral factors were calculated for profiles with the jaws in the extreme positions using both variable-width (Synchrony) and fixed-width apertures. Treatment plans with target motion and compensation were compared to planned doses to study the impact of the variable aperture on volumetric dose. RESULTS: The jaw offset peak factor (JOPF) for the Synchrony jaw settings were 0.964 and 0.983 for the 1.0- and 2.5-cm jaw settings, respectively. These values decreased to 0.925 and 0.982 for the fixed-width settings, indicating that the peak value of the profile would decrease by 7.5% compared to centered if the aperture width was held constant. The IMRT dose distributions reveal similar results, where gamma pass rates are above tolerance for the Synchrony jaw settings but fall significantly for the fixed-width 1-cm jaws. CONCLUSIONS: The variable-width behavior of Synchrony jaws provides a larger output correction for the 1-cm jaw setting. Without the variable-aperture correction, plans with the 1-cm jaw setting would underdose the target if the jaws spend a significant amount of time in the extreme positions. This work investigated the change in delivered dose with jaws in the extreme positions, therefore overall changes in dose due to offset jaws are expected to be less for composite treatment deliveries.

Technical note: On the impact of the kV imaging configuration on doses from planar images during motion-synchronized treatments on Radixact®.

Kilovoltage radiographs are acquired during motion-synchronized treatments on Radixact to localize the tumor during the treatment. Several previous publications have provided estimates of patient dose from these planar radiographs. However, a recent hardware update changed several aspects of the kV imaging system, including a new X-ray tube, an extended source-to-axis distance (SAD), and a larger field size. This is denoted the extended configuration. The purpose of this work was to assess the impact of the configuration change on patient dose from these procedures. Point doses in water were measured using the TG-61 protocol for tube potentials between 100 and 140 kVp for both the standard and extended configurations under the same water tank setup. Comparisons were made for equal mAs since the same protocols (kVp, mAs) will be used for both configurations. In comparison to the standard configuration, doses per mAs from the extended configuration were found to be ~66% less and falloff less steep due to the increased SAD. However, a larger volume of tissue is irradiated due to the larger field size. Beam quality for a given tube potential was the same as determined by half-value layer measurements. Both kV configurations are available from the vendor, therefore, the values in this work can be used to compare values previously published in the literature for the standard configuration or to intercompare doses from these two system configurations.

The quantification and potential impact of dark current on treatments with an MR-guided radiotherapy (MRgRT) system.

PURPOSE: Dark current radiation produced during linac beam-hold has the potential to lead to unplanned dose delivered to the patient. With the increased usage of motion management and step-and-shoot IMRT deliveries for MR-guided systems leading to increased beam-hold time, it is necessary to consider the impact of dark current radiation on patient treatments. METHODS: The relative dose rate due to dark current for the ViewRay MRIdian linac was measured longitudinally over 15 months (June 2018-August 2019). Ion chamber measurements were acquired with the linac in the beam-hold state and the beam-on state, with the ratio representing the relative dark current dose rate. The potential contribution of the dark current dose to the overall prescription was retrospectively analyzed for 972 fractions from 83 patients over the same time period. The amount of time spent in the beam-hold state was combined with the monthly measured relative dark current dose rate to estimate the dark current dose contribution. RESULTS: The relative dark current dose rate compared to the beam-on dose rate was 0.12% ± 0.027%. In a near worst-case estimation, the dark current dose contribution accounted for 0.90% ± 0.67% of the prescription dose across all fractions (3.61% maximum). Gantry and MLC motion between segments accounted for 87% of the dark current contribution, with the remaining 13% attributable to gating during segment delivery. The largest dark current contributions were associated with plans delivering a small dose per treatment segment. CONCLUSIONS: The dark current associated with new clinical treatment units should be considered prior to treatment delivery to ensure it will not lead to dosimetric inaccuracies. For the MRIdian linac system investigated in this work, the contribution from dark current remained relatively low, though users should be cognizant of the larger potential dosimetric contribution for plans with small doses per segment.

Evaluation of radixact motion synchrony for 3D respiratory motion: Modeling accuracy and dosimetric fidelity.

The Radixact® linear accelerator contains the motion Synchrony system, which tracks and compensates for intrafraction patient motion. For respiratory motion, the system models the motion of the target and synchronizes the delivery of radiation with this motion using the jaws and multi-leaf collimators (MLCs). It was the purpose of this work to determine the ability of the Synchrony system to track and compensate for different phantom motions using a delivery quality assurance (DQA) workflow. Thirteen helical plans were created on static datasets from liver, lung, and pancreas subjects. Dose distributions were measured using a Delta4® Phantom+ mounted on a Hexamotion® stage for the following three case scenarios for each plan: (a) no phantom motion and no Synchrony (M0S0), (b) phantom motion and no Synchrony (M1S0), and (c) phantom motion with Synchrony (M1S1). The LEDs were placed on the Phantom+ for the 13 patient cases and were placed on a separate one-dimensional surrogate stage for additional studies to investigate the effect of separate target and surrogate motion. The root-mean-square (RMS) error between the Synchrony-modeled positions and the programmed phantom positions was <1.5 mm for all Synchrony deliveries with the LEDs on the Phantom+. The tracking errors increased slightly when the LEDs were placed on the surrogate stage but were similar to tracking errors observed for other motion tracking systems such as CyberKnife Synchrony. One-dimensional profiles indicate the effects of motion interplay and dose blurring present in several of the M1S0 plans that are not present in the M1S1 plans. All 13 of the M1S1 measured doses had gamma pass rates (3%/2 mm/10%T) compared to the planned dose > 90%. Only two of the M1S0 measured doses had gamma pass rates > 90%. Motion Synchrony offers a potential alternative to the current, ITV-based motion management strategy for helical tomotherapy deliveries.

Characterization and longitudinal assessment of daily quality assurance for an MR-guided radiotherapy (MRgRT) linac.

PURPOSE: To describe and characterize daily machine quality assurance (QA) for an MR-guided radiotherapy (MRgRT) linac system, in addition to reporting a longitudinal assessment of the dosimetric and mechanical stability over a 7-month period of clinical operation. METHODS: Quality assurance procedures were developed to evaluate MR imaging/radiation isocenter, imaging and patient handling system, and linear accelerator stability. A longitudinal assessment was characterized for safety interlocks, laser and imaging isocenter coincidence, imaging and radiation (RT) isocentricity, radiation dose rate and output, couch motion, and MLC positioning. A cylindrical water phantom and an MR-compatible A1SL detector were utilized. MR and RT isocentricity and MLC positional accuracy was quantified through dose measured with a 0.40 cm2  x 0.83 cm2 field at each cardinal angle. The relationship between detector response to MR/RT isocentricity and MLC positioning was established through introducing known errors in phantom position. RESULTS: Correlation was found between detector response and introduced positional error (N = 27) with coefficients of determination of 0.9996 (IEC-X), 0.9967 (IEC-Y), 0.9968 (IEC-Z) in each respective shift direction. The relationship between dose (DoseMR/RT+MLC ) and the vector magnitude of MLC and MR/RT positional error (Errormag ) was calculated to be a nonlinear response and resembled a quadratic function: DoseMR/RT+MLC [%] = -0.0253 Errormag [mm]2  - 0.0195 Errormag [mm]. For the temporal assessment (N = 7 months), safety interlocks were functional. Laser coincidence to MR was within ±2.0 mm (99.6%) and ±1.0 mm (86.8%) over the 7-month assessment. IGRT position-reposition shifts were within ±2.0 mm (99.4%) and ±1.0 mm (92.4%). Output was within ±3% (99.4%). Mean MLC and MR/RT isocenter accuracy was 1.6 mm, averaged across cardinal angles for the 7-month period. CONCLUSIONS: The linac and IGRT accuracy of an MR-guided radiotherapy system has been validated and monitored over seven months for daily QA. Longitudinal assessment demonstrated a drift in dose rate, but temporal assessment of output, MLC position, and isocentricity has been stable.

An analysis of the ArcCHECK-MR diode array's performance for ViewRay quality assurance.

The ArcCHECK-MR diode array utilizes a correction system with a virtual inclinometer to correct the angular response dependencies of the diodes. However, this correction system cannot be applied to measurements on the ViewRay MR-IGRT system due to the virtual inclinometer's incompatibility with the ViewRay's multiple simultaneous beams. Additionally, the ArcCHECK's current correction factors were determined without magnetic field effects taken into account. In the course of performing ViewRay IMRT quality assurance with the ArcCHECK, measurements were observed to be consistently higher than the ViewRay TPS predictions. The goals of this study were to quantify the observed discrepancies and test whether applying the current factors improves the ArcCHECK's accuracy for measurements on the ViewRay. Gamma and frequency analysis were performed on 19 ViewRay patient plans. Ion chamber measurements were performed at a subset of diode locations using a PMMA phantom with the same dimensions as the ArcCHECK. A new method for applying directionally dependent factors utilizing beam information from the ViewRay TPS was developed in order to analyze the current ArcCHECK correction factors. To test the current factors, nine ViewRay plans were altered to be delivered with only a single simultaneous beam and were measured with the ArcCHECK. The current correction factors were applied using both the new and current methods. The new method was also used to apply corrections to the original 19 ViewRay plans. It was found the ArcCHECK systematically reports doses higher than those actually delivered by the ViewRay. Application of the current correction factors by either method did not consistently improve measurement accuracy. As dose deposition and diode response have both been shown to change under the influence of a magnetic field, it can be concluded the current ArcCHECK correction factors are invalid and/or inadequate to correct measurements on the ViewRay system.

Characterization of a 0.35T MR system for phantom image quality stability and in vivo assessment of motion quantification.

ViewRay is a novel MR-guided radiotherapy system capable of imaging in near real-time at four frames per second during treatment using 0.35T field strength. It allows for improved gating techniques and adaptive radiotherapy. Three cobalt-60 sources (~ 15,000 Curies) permit multiple-beam, intensity-modulated radiation therapy. The primary aim of this study is to assess the imaging stability, accuracy, and automatic segmentation algorithm capability to track motion in simulated and in vivo targets. Magnetic resonance imaging (MRI) characteristics of the system were assessed using the American College of Radiology (ACR)-recommended phantom and accreditation protocol. Images of the ACR phantom were acquired using a head coil following the ACR scanning instructions. ACR recommended T1- and T2-weighted sequences were evaluated. Nine measurements were performed over a period of seven months, on just over a monthly basis, to establish consistency. A silicon dielectric gel target was attached to the motor via a rod. 40 mm total amplitude was used with cycles of 3 to 9 s in length in a sinusoidal trajectory. Trajectories of six moving clinical targets in four canine patients were quantified and tracked. ACR phantom images were analyzed, and the results were compared with the ACR acceptance levels. Measured slice thickness accuracies were within the acceptance limits. In the 0.35 T system, the image intensity uniformity was also within the ACR acceptance limit. Over the range of cycle lengths, representing a wide range of breathing rates in patients imaged at four frames/s, excellent agreement was observed between the expected and measured target trajectories. In vivo canine targets, including the gross target volume (GTV), as well as other abdominal soft tissue structures, were visualized with inherent MR contrast, allowing for preliminary results of target tracking.

Spatial mapping of functional pelvic bone marrow using FLT PET.

The purpose of this study was to determine the ability of regions identified with bony landmarks on CT imaging to accurately represent active bone marrow when compared to FLT PET imaging. These surrogate regions could then be used to create a bone marrow sparing radiation therapy plan when FLT PET imaging is not available. Whole body (WB) FLT PET images were obtained of 18 subjects prior to chemoradiation therapy. The FLT image of each subject was registered to a CT image acquired for that subject to obtain anatomic information of the pelvis. Seventeen regions were identified based on features of the pelvic bones, sacrum, and femoral heads. The probability of FLT uptake being located in each of 17 different CT-based regions of the bony pelvis was calculated using Tukey's multiple comparison test. Statistical analysis of FLT uptake in the pelvis indicated four distinct groups within the 17 regions that had similar levels of activity. Regions located in the central part of the pelvis, including the superior part of the sacrum, the inner halves of the iliac crests, and the L5 vertebral body, had greater FLT uptake than those in the peripheral regions (p-value < 0.05). We have developed a method to use CT-defined pelvic bone regions to represent FLT PET-identified functional bone marrow. Individual regions that have a statistically significant probability of containing functional bone marrow can be used as avoidance regions to reduce radiation dose to functional bone marrow in radiation therapy planning. However, because likely active bone marrow regions and pelvic targets typically overlap, patient-specific spatial detail may be advantageous in IMRT planning scenarios and may best be provided using FLT PET imaging.

A Phase 2 Randomized Clinical Trial Evaluating 4-Dimensional Computed Tomography Ventilation-Based Functional Lung Avoidance Radiation Therapy for Non-Small Cell Lung Cancer.

PURPOSE: To determine whether 4-dimensional computed tomography (4DCT) ventilation-based functional lung avoidance radiation therapy preserves pulmonary function compared with standard radiation therapy for non-small cell lung cancer (NSCLC). METHODS AND MATERIALS: This single center, randomized, phase 2 trial enrolled patients with NSCLC receiving curative intent radiation therapy with either stereotactic body radiation therapy or conventionally fractionated radiation therapy between 2016 and 2022. Patients were randomized 1:1 to standard of care radiation therapy or functional lung avoidance radiation therapy. The primary endpoint was the change in Jacobian-based ventilation as measured on 4DCT from baseline to 3 months postradiation. Secondary endpoints included changes in volume of high- and low-ventilating lung, pulmonary toxicity, and changes in pulmonary function tests (PFTs). RESULTS: A total of 122 patients were randomized and 116 were available for analysis. Median follow up was 29.9 months. Functional avoidance plans significantly (P < .05) reduced dose to high-functioning lung without compromising target coverage or organs at risk constraints. When analyzing all patients, there was no difference in the amount of lung showing a reduction in ventilation from baseline to 3 months between the 2 arms (1.91% vs 1.87%; P = .90). Overall grade ≥2 and grade ≥3 pulmonary toxicities for all patients were 24.1% and 8.6%, respectively. There was no significant difference in pulmonary toxicity or changes in PFTs between the 2 study arms. In the conventionally fractionated cohort, there was a lower rate of grade ≥2 pneumonitis (8.2% vs 32.3%; P = .049) and less of a decline in change in forced expiratory volume in 1 second (-3 vs -5; P = .042) and forced vital capacity (1.5 vs -6; P = .005) at 3 months, favoring the functional avoidance arm. CONCLUSIONS: There was no difference in posttreatment ventilation as measured by 4DCT between the arms. In the cohort of patients treated with conventionally fractionated radiation therapy with functional lung avoidance, there was reduced pulmonary toxicity, and less decline in PFTs suggesting a clinical benefit in patients with locally advanced NSCLC.

Equivalent-quality unflattened photon beam modeling, planning, and delivery.

The clinical application of the flattening filter-free photon beam (FFF) has enjoyed greater use due to its advantage of reduced treatment time because of the increased dose rate. Its unique beam characteristics, along with the very high-dose rate, require a thorough knowledge of the capability and accuracy in FFF beam modeling, planning, and delivery. This work verifies the feasibility of modeling an equivalent quality unflattened photon beam (eqUF), and the dosimetric accuracy in eqUF beam planning and delivery. An eqUF beam with a beam quality equivalent to a conventional 6 MV photon beam with the filter in place (WF) was modeled for the Pinnacle3 TPS and the beam model quality was evaluated by gamma index test. Results showed that the eqUF beam modeling was similar to that of the WF beam, as the overall passing rate of the 2%/2 mm gamma index test was 99.5% in the eqUF beam model and 96% in the WF beam model. Hypofractionated IMRT plans were then generated with the same constraints using both WF and eqUF beams, and the similarity was evaluated by DVH comparison and generalized 3D gamma index test. The WF and eqUF plans showed no clinically significant differences in DVH comparison and, on average > 98% voxels passed the 3%/3 mm 3D gamma index test. Dosimetric accuracy in gated phantom delivery was verified by ion chamber and film measurements. All ion chamber measurements at the isocenter were within 1% of calculated values and film measurements passed the 3 mm/3% gamma index test with an overall passing rate > 95% in the high-dose and low-gradient region in both WF and eqUF cases. Treatment plan quality assurance (QA), using either measurement-based or independent calculation-based methods of ten clinically treated eqUF IMRT plans were analyzed. In both methods, the point dose differences were all within 2% difference. In the relative 2D dose distribution comparison, >95% points were within 3% dose difference or 3 mm DTA.

An Almost Linear Time Algorithm for Field Splitting in Radiation Therapy.

In this paper, we study an interesting geometric partition problem, called optimal field splitting, which arises in Intensity-Modulated Radiation Therapy (IMRT). In current clinical practice, a multi-leaf collimator (MLC) with a maximum leaf spread constraint is used to deliver the prescribed intensity maps (IMs). However, the maximum leaf spread of an MLC may require to split a large intensity map into several overlapping sub-IMs with each being delivered separately. We develop a close-to-linear time algorithm for solving the field splitting problem while minimizing the total complexity of the resulting sub-IMs, thus improving the treatment delivery efficiency. Meanwhile, our algorithm strives to minimize the maximum beam-on time of those sub-IMs. Our basic idea is to formulate the field splitting problem as computing a shortest path in a directed acyclic graph, which expresses a special "layered" structure. The edge weights of the graph satisfy the Monge property, which enables us to solve this shortest path problem by examining only a small portion of the graph, yielding a close-to-linear time algorithm. To minimize the maximum beam-on time of the resulting sub-IMs, we consider an interesting min-max slope path problem in a monotone polygon which is solvable in linear time. The min-max slope path problem may be of interest in its own right. Experimental results based on real medical data and computer generated IMs showed that our new algorithm runs fast and produces high quality field splitting results.

Evaluating on-board kVCT- and MVCT-based dose calculation accuracy using a thorax phantom for helical tomotherapy treatments.

Purpose.To evaluate the impact of CT number calibration and imaging parameter selection on dose calculation accuracy relative to the CT planning process in thoracic treatments for on-board helical CT imaging systems used in helical tomotherapy.Methods and Materials.Direct CT number calibrations were performed with appropriate protocols for each imaging system using an electron density phantom. Large volume and SBRT treatment plans were simulated and optimized for planning CT scans of an anthropomorphic thorax phantom and transferred to registered kVCT and MVCT scans of the phantom as appropriate. Relevant DVH metrics and dose-difference maps were used to evaluate and compare dose calculation accuracy relative to the planning CT based on a variation in imaging parameters applied for the on-board systems.Results.For helical kVCT scans of the thorax phantom, median differences in DVH parameters for the large volume treatment plan were less than ±1% with dose to the target volume either over- or underestimated depending on the imaging parameters utilized for CT number calibration and thorax phantom acquisition. For the lung SBRT plan calculated on helical kVCT scans, median dose differences were up to -2.7% with a more noticeable dependence on parameter selection. For MVCT scans, median dose differences for the large volume plan were within +2% with dose to the target overestimated regardless of the imaging protocol.Conclusion.Accurate dose calculations (median errors of <±1%) using a thorax phantom simulating realistic patient geometry and scatter conditions can be achieved with images acquired with a helical kVCT system on a helical tomotherapy unit. This accuracy is considerably improved relative to that achieved with the MV-based approach. In a clinical setting, careful consideration should be made when selecting appropriate kVCT imaging parameters for this process as dose calculation accuracy was observed to vary with both parameter selection and treatment type.

Robust quantification of CT-ventilation biomarker techniques and repeatability in a porcine model.

BACKGROUND: Biomarkers estimating local lung ventilation have been derived from computed tomography (CT) imaging using various image acquisition and post-processing techniques. CT-ventilation biomarkers have potential clinical use in functional avoidance radiation therapy (RT), in which RT treatment plans are optimized to reduce dose delivered to highly ventilated lung. Widespread clinical implementation of CT-ventilation biomarkers necessitates understanding of biomarker repeatability. Performing imaging within a highly controlled experimental design enables quantification of error associated with remaining variables. PURPOSE: To characterize CT-ventilation biomarker repeatability and dependence on image acquisition and post-processing methodology in anesthetized and mechanically ventilated pigs. METHODS: Five mechanically ventilated Wisconsin Miniature Swine (WMS) received multiple consecutive four-dimensional CT (4DCT) and maximum inhale and exhale breath-hold CT (BH-CT) scans on five dates to generate CT-ventilation biomarkers. Breathing maneuvers were controlled with an average tidal volume difference <200 cc. As surrogates for ventilation, multiple local expansion ratios (LERs) were calculated from the acquired CT scans using Jacobian-based post-processing techniques. L E R 2 $LER_2$ measured local expansion between an image pair using either inhale and exhale BH-CT images or two 4DCT breathing phase images. L E R N $LER_N$ measured the maximum local expansion across the 4DCT breathing phase images. Breathing maneuver consistency, intra- and interday biomarker repeatability, image acquisition and post-processing technique dependence were quantitatively analyzed. RESULTS: Biomarkers showed strong agreement with voxel-wise Spearman correlation ρ > 0.9 $\rho > 0.9$ for intraday repeatability and ρ > 0.8 $\rho > 0.8$ for all other comparisons, including between image acquisition techniques. Intra- and interday repeatability were significantly different (p < 0.01). LER2 and LERN post-processing did not significantly affect intraday repeatability. CONCLUSIONS: 4DCT and BH-CT ventilation biomarkers derived from consecutive scans show strong agreement in controlled experiments with nonhuman subjects.

Predicting pulmonary ventilation damage after radiation therapy for nonsmall cell lung cancer using a ResNet generative adversarial network.

BACKGROUND: Functional lung avoidance radiation therapy (RT) is a technique being investigated to preferentially avoid specific regions of the lung that are predicted to be more susceptible to radiation-induced damage. Reducing the dose delivered to high functioning regions may reduce the occurrence radiation-induced lung injuries (RILIs) and toxicities. However, in order to develop effective lung function-sparing plans, accurate predictions of post-RT ventilation change are needed to determine which regions of the lung should be spared. PURPOSE: To predict pulmonary ventilation change following RT for nonsmall cell lung cancer using machine learning. METHODS: A conditional generative adversarial network (cGAN) was developed with data from 82 human subjects enrolled in a randomized clinical trial approved by the institution's IRB to predict post-RT pulmonary ventilation change. The inputs to the network were the pre-RT pulmonary ventilation map and radiation dose distribution. The loss function was a combination of the binary cross-entropy loss and an asymmetrical structural similarity index measure (aSSIM) function designed to increase penalization of under-prediction of ventilation damage. Network performance was evaluated against a previously developed polynomial regression model using a paired sample t-test for comparison. Evaluation was performed using eight-fold cross-validation. RESULTS: From the eight-fold cross-validation, we found that relative to the polynomial model, the cGAN model significantly improved predicting regions of ventilation damage following radiotherapy based on true positive rate (TPR), 0.14±0.15 to 0.72±0.21, and Dice similarity coefficient (DSC), 0.19±0.16 to 0.46±0.14, but significantly declined in true negative rate, 0.97±0.05 to 0.62±0.21, and accuracy, 0.79±0.08 to 0.65±0.14. Additionally, the average true positive volume increased from 104±119 cc in the POLY model to 565±332 cc in the cGAN model, and the average false negative volume decreased from 654±361 cc in the POLY model to 193±163 cc in the cGAN model. CONCLUSIONS: The proposed cGAN model demonstrated significant improvement in TPR and DSC. The higher sensitivity of the cGAN model can improve the clinical utility of functional lung avoidance RT by identifying larger volumes of functional lung that can be spared and thus decrease the probability of the patient developing RILIs.

Quantifying robustness of CT-ventilation biomarkers to image noise.

Purpose: To quantify the impact of image noise on CT-based lung ventilation biomarkers calculated using Jacobian determinant techniques. Methods: Five mechanically ventilated swine were imaged on a multi-row CT scanner with acquisition parameters of 120 kVp and 0.6 mm slice thickness in static and 4-dimensional CT (4DCT) modes with respective pitches of 1 and 0.09. A range of tube current time product (mAs) values were used to vary image dose. On two dates, subjects received two 4DCTs: one with 10 mAs/rotation (low-dose, high-noise) and one with CT simulation standard of care 100 mAs/rotation (high-dose, low-noise). Additionally, 10 intermediate noise level breath-hold (BHCT) scans were acquired with inspiratory and expiratory lung volumes. Images were reconstructed with and without iterative reconstruction (IR) using 1 mm slice thickness. The Jacobian determinant of an estimated transformation from a B-spline deformable image registration was used to create CT-ventilation biomarkers estimating lung tissue expansion. 24 CT-ventilation maps were generated per subject per scan date: four 4DCT ventilation maps (two noise levels each with and without IR) and 20 BHCT ventilation maps (10 noise levels each with and without IR). Biomarkers derived from reduced dose scans were registered to the reference full dose scan for comparison. Evaluation metrics were gamma pass rate (Γ) with 2 mm distance-to-agreement and 6% intensity criterion, voxel-wise Spearman correlation (ρ) and Jacobian ratio coefficient of variation (CoV JR ). Results: Comparing biomarkers derived from low (CTDI vol = 6.07 mGy) and high (CTDI vol = 60.7 mGy) dose 4DCT scans, mean Γ, ρ and CoV JR values were 93% ± 3%, 0.88 ± 0.03 and 0.04 ± 0.009, respectively. With IR applied, those values were 93% ± 4%, 0.90 ± 0.04 and 0.03 ± 0.003. Similarly, comparisons between BHCT-based biomarkers with variable dose (CTDI vol = 1.35-7.95 mGy) had mean Γ, ρ and CoV JR of 93% ± 4%, 0.97 ± 0.02 and 0.03 ± 0.006 without IR and 93% ± 4%, 0.97 ± 0.03 and 0.03 ± 0.007 with IR. Applying IR did not significantly change any metrics (p > 0.05). Discussion: This work demonstrated that CT-ventilation, calculated using the Jacobian determinant of an estimated transformation from a B-spline deformable image registration, is invariant to Hounsfield Unit (HU) variation caused by image noise. This advantageous finding may be leveraged clinically with potential applications including dose reduction and/or acquiring repeated low-dose acquisitions for improved ventilation characterization.

Validation of CT-based ventilation and perfusion biomarkers with histopathology confirms radiation-induced pulmonary changes in a porcine model.

Imaging biomarkers can assess disease progression or prognoses and are valuable tools to help guide interventions. Particularly in lung imaging, biomarkers present an opportunity to extract regional information that is more robust to the patient's condition prior to intervention than current gold standard pulmonary function tests (PFTs). This regional aspect has particular use in functional avoidance radiation therapy (RT) in which treatment planning is optimized to avoid regions of high function with the goal of sparing functional lung and improving patient quality of life post-RT. To execute functional avoidance, detailed dose-response models need to be developed to identify regions which should be protected. Previous studies have begun to do this, but for these models to be clinically translated, they need to be validated. This work validates two metrics that encompass the main components of lung function (ventilation and perfusion) through post-mortem histopathology performed in a novel porcine model. With these methods validated, we can use them to study the nuanced radiation-induced changes in lung function and develop more advanced models.

Metrics of dose to highly ventilated lung are predictive of radiation-induced pneumonitis in lung cancer patients.

PURPOSE: To identify metrics of radiation dose delivered to highly ventilated lung that are predictive of radiation-induced pneumonitis. METHODS AND MATERIALS: A cohort of 90 patients with locally advanced non-small cell lung cancer treated with standard fractionated radiation therapy (RT) (60-66 Gy in 30-33 fractions) were evaluated. Regional lung ventilation was determined from pre-RT 4-dimensional computed tomography (4DCT) using the Jacobian determinant of a B-spline deformable image registration to estimate lung tissue expansion during respiration. Multiple voxel-wise population- and individual-based thresholds for defining high functioning lung were considered. Mean dose and volumes receiving dose ≥ 5-60 Gy were analyzed for both total lung-ITV (MLD,V5-V60) and highly ventilated functional lung-ITV (fMLD,fV5-fV60). The primary endpoint was symptomatic grade 2+ (G2+) pneumonitis. Receiver operator curve (ROC) analyses were used to identify predictors of pneumonitis. RESULTS: G2+ pneumonitis occurred in 22.2% of patients, with no differences between stage, smoking status, COPD, or chemo/immunotherapy use between G<2 and G2+ patients (P≥ 0.18). Highly ventilated lung was defined as voxels exceeding the population-wide median of 18% voxel-level expansion. All total and functional metrics were significantly different between patients with and without pneumonitis (P≤ 0.039). Optimal ROC points predicting pneumonitis from functional lung dose were fMLD ≤ 12.3 Gy, fV5 ≤ 54% and fV20 ≤ 19 %. Patients with fMLD ≤ 12.3 Gy had a 14% risk of developing G2+ pneumonitis whereas risk significantly increased to 35% for those with fMLD > 12.3 Gy (P = 0.035). CONCLUSIONS: Dose to highly ventilated lung is associated with symptomatic pneumonitis and treatment planning strategies should focus on limiting dose to functional regions. These findings provide important metrics to be used in functional lung avoidance RT planning and designing clinical trials.