The results of the challenge showed that the lungs and heart can be segmented fairly accurately by various algorithms, while deep-learning methods performed better on the esophagus. Our dataset together with the manual contours for all training cases continues to be available publicly as an ongoing benchmarking resource.

Purpose Intensity‐modulated radiation therapy (IMRT) quality assurance (QA) measurements are routinely performed prior to treatment delivery to verify dose calculation and delivery accuracy. In this work, we applied a machine learning‐based approach to predict portal dosimetry based IMRT QA gamma passing rates. Methods 182 IMRT plans for various treatment sites were planned and delivered with portal dosimetry on two TrueBeam and two Trilogy LINACs. A total of 1497 beams were collected and analyzed using gamma criteria of 2%/2 mm with a 5% threshold. The datasets for building the machine learning models consisted of 1269 beams. Ten‐fold cross‐validation was utilized to tune the model and prevent “overfitting.” A separate test set with the remaining 228 beams was used to evaluate model performance. Each beam was characterized by a set of 31 features including both plan complexity metrics and machine characteristics. Three tree‐based machine learning algorithms (AdaBoost, Random Forest, and XGBoost) were used to train the models and predict gamma passing rates. Results Both AdaBoost and Random Forest had 98% of predictions within 3% of the measured 2%/2 mm gamma passing rates with a maximum error less than 4% and a mean absolute error < 1%. XGBoost showed a slightly worse prediction accuracy with 95% of the predictions within 3% of the measured gamma passing rates and a maximum error of 4.5%. The three models identified the same nine features in the top 10 most important ones that are related to plan complexity and maximum aperture displacement from the central axis or the maximum jaw size in a beam. Conclusion We have demonstrated that portal dosimetry IMRT QA gamma passing rates can be accurately predicted using tree‐based ensemble learning models. The machine learning based approach allows physicists to better identify the failures of IMRT QA measurements and to develop proactive QA approaches.

Machine learning methods can be used to predict OF for double-scatter proton machines with greater prediction accuracy than the most popular semi-empirical prediction model. By incorporating the gantry angle dependence and field size dependence, the machine learning-based methods can be used for a sanity check of OF measurements and bears the potential to eliminate the time-consuming patient-specific OF measurements.

No abstract

Clinical implementation of online adaptive radiation therapy requires initial and ongoing performance assessment of the underlying auto-segmentation and adaptive planning algorithms, although a straightforward and efficient process for this in phantom is lacking. The purpose of this work was to investigate robustness and repeatability of the artificial intelligence-assisted online segmentation and adaptive planning process on the Varian Ethos adaptive platform, and to develop an end-to-end test strategy for online adaptive radiation therapy. Five synthetic deformations were generated and applied to a computed tomography image of an anthropomorphic pelvis phantom, and reference treatment plans were generated from each of the resulting deformed images. The undeformed phantom was repeatedly imaged, and the online adaptive process was performed including auto-segmentation, review and manual correction of contours, and adaptive plan creation. One adaptive fractions in five different deformation scenarios were performed. The manually corrected contours had a high degree of consistency (> 93% Dice similarity coefficient and < 1.0 mm mean surface distance) across repeated fractions, with no significant variation across the synthetic deformation instance except for bowel (p = 0.026, one-way ANOVA). Adaptive treatment plans also resulted in highly consistent dose-volume values for targets and organs at risk. A straightforward and efficient process was developed and used to quantify a set of organ specific contouring and dosimetric action levels to help establish uncertainty bounds for an end-to-end test on the Varian Ethos system.

We develop a fully automated QA process to compare the image quality of all kV CBCT protocols on a Halcyon linac with ring gantry design, and evaluate image quality stability over a 10-month period. A total of 19 imaging scan and reconstruction protocols were characterized with measurement on a newly released QUART phantom. A set of image analysis algorithms were developed and integrated into an automated analysis suite to derive key image quality metrics, including HU value accuracy on density inserts, HU uniformity using the background plate, high contrast resolution with the modulation transfer function (MTF) from the edge profiles, low contrast resolution using the signalto-noise ratio (SNR) and contrast-to-noise ratio (CNR), slice thickness with the air gap modules, and geometric accuracy with the diameter of the phantom. Image quality data over 10 months was tracked and analyzed to evaluate the stability of the Halcyon kV imaging system. The HU accuracy over all 19 protocols is within tolerance (±50HU). The maximum uniformity deviation is 12.2 HU. The SNR and CNR, depending on the protocol selected, range from 18.5-911.9 and 1.9-102.8, respectively. A much-improved SNR and CNR were observed for iterative reconstruction (iCBCT) modes and protocols designed for large subjects over low dose and fast scanning modes. The Head and Image Gently protocols have the greatest high contrast resolution with MTF 10% over 1 lp/mm and MTF 50% over 0.6 lp/mm. The iCBCT mode slightly improved the MTF 10% and MTF 50% compared to the Feldkamp-Davis-Kress approach. The slice thickness (maximum error of 0.31 mm) and geometry metrics (maximum error of 0.7 mm) are all within tolerance (±0.5 mm for slice thickness and ±1 mm for geometry metrics). The long-term study over 10-month showed no significant drift for all key image quality metrics, which indicated the kV CBCT image quality is stable over time.

Purpose: To evaluate the targetability of late-stage cervical cancer by magnetic resonance-guided high-intensity focused ultrasound (MRgHIFU)-induced hyperthermia (HT) as an adjuvant to radiation therapy (RT). Methods: Seventy-nine cervical cancer patients (stage IIIB-IVA) who received RT with lesions visible on positron emission tomography-computed tomography (PET-CT) were retrospectively analyzed for targetability using a commercially-available HT-capable MRgHIFU system. Targetability was assessed for both primary targets and/or any metastatic lymph nodes using both posterior (supine) and anterior (prone) patient setups relative to the transducer. Thirty-four different angles of rotation along subjects' longitudinal axis were analyzed. Targetability was categorized as: (1) Targetable with/without minimal intervention; (2) Not targetable. To determine if any factors could be used for prospective screening of patients, potential associations between demographic/anatomical factors and targetability were analyzed. Results: 72.15% primary tumors and 33.96% metastatic lymph nodes were targetable from at least one angle. 49.37% and 39.24% of primary tumors could be targeted with patient laying in supine and prone positions, respectively. 25 -30 rotation and 0 rotation had the highest rate of the posterior and anterior targetability, respectively. The ventral depth of the tumor and its distance to the coccyx were statistically correlated with the anterior and posterior targetability, respectively. Conclusion: Most late-stage cervical cancer primaries were targetable by MRgHIFU HT requiring either no/minimal intervention. A rotation of 0 or 25 -30 relative to the transducer might benefit anterior and posterior targetability, respectively. Certain demographic/anatomic parameters might be useful in screening patients for treatability.

Purpose: To characterize temperature fields and tissue damage profiles of large-volume hyperthermia (HT) induced by magnetic resonance-guided high-intensity focused ultrasound (MRgHIFU) in deep and superficial targets in vivo in a porcine model. Methods: Nineteen HT sessions were performed in vivo with a commercial MRgHIFU system (Sonalleve V R V2, Profound Medical Inc., Mississauga, ON, Canada) in hind leg muscles of eight pigs with temperature fields of cross-sectional diameter of 58-mm. Temperature statistics evaluated in the target region-of-interest (tROI) included accuracy, temporal variation, and uniformity. The impact of the number and location of imaging planes for feedback-based temperature control were investigated. Temperature fields were characterized by time-in-range (TIR, the duration each voxel stays within 40-45 C) maps. Tissue damage was characterized by contrast-enhanced MRI, and macroscopic and histopathological analysis. The performance of the Sonalleve V R system was benchmarked against a commercial phantom. Results: Across all HT sessions, the mean difference between the average temperature (T avg) and the desired temperature was À0.4 ± 0.5 C; the standard deviation of temperature 1.2 ± 0.2 C; the temporal variation of T avg for 30-min HT was 0.6 ± 0.2 C, and the temperature uniformity was 1.5 ± 0.2 C. A difference of 2.2-cm (in pig) and 1.5-cm (in phantom) in TIR dimensions was observed when applying feedback-based plane(s) at different locations. Histopathology showed 62.5% of examined HT sessions presenting myofiber degeneration/necrosis within the target volume. Conclusion: Large-volume MRgHIFU-mediated HT was successfully implemented and characterized in a porcine model in deep and superficial targets in vivo with heating distributions modifiable by userdefinable parameters.