Radiotherapy Treatment Planning Research Articles

Prospective risk analysis of the online-adaptive artificial intelligence-driven workflow using the Ethos treatment system

PurposeThe recently introduced Varian Ethos system allows adjusting radiotherapy treatment plans to anatomical changes on a daily basis. The system uses artificial intelligence to speed up the process of creating adapted plans, comes with its own software solutions and requires a substantially different workflow. A detailed analysis of possible risks of the associated workflow is presented. MethodsA prospective risk analysis of the adaptive workflow with the Ethos system was performed using Failure Modes and Effects Analysis (FMEA). An interprofessional team collected possible adverse events and evaluated their severity as well as their chance of occurrence and detectability. Measures to reduce the risks were discussed. ResultsA total of 122 events were identified, and scored. Within the 20 events with the highest-ranked risks, the following were identified: Challenges due to the stand-alone software solution with very limited connectivity to the existing record and verify software and digital patient file, unfamiliarity with the new software and its limitations and the adaption process relying on results obtained by artificial intelligence. The risk analysis led to the implementation of additional quality assurance measures in the workflow. ConclusionsThe thorough analysis of the risks associated with the new treatment technique was the basis for designing details of the workflow. The analysis also revealed challenges to be addressed by both, the vendor and customers. On the vendor side, this includes improving communication between their different software solutions. On the customer side, this especially includes establishing validation strategies to monitor the results of the black box adaption process making use of artificial intelligence.

Evaluation of Inhomogeneity Correction Performed by Radiotherapy Treatment Planning System

Aim of this study is to evaluate the efficacy of inhomogeneity corrections calculated by radiotherapy treatment planning system (TPS) using various densities of materials. Gammex Computed tomography electron density inserts (EDI's; 14 no's) were used to generate the CT to ED curve with high speed GE CT scanner by noting down the respective HU values of each rod. Treatment plans were generated in XiO TPS with three inhomogeneous phantoms (comprising combination of water, lung and bone equivalent slabs) with different field sizes and for EDI (8 no's) inserted in slots of acrylic tray and validation was carried out using 2D array detector with 20cm×20cm field size for 200 MU. Point dose and fluence measurements were carried with inhomogeneous phantoms combinations and EDI's (placed on the locally fabricated box filled with water medium). The mean percentage deviations with standard deviation of calculated point doses against measured ones obtained with 2D array detector at iso-center plane for all three inhomogeneous phantom combinations were found to be -1.13%±0.13%, -3.51%±0.14% and -0.63%±0.27% respectively. On point doses measured under each individual EDI, over all percentage deviation with standard deviation observed is -2.04% ± 1.1%. The described method can be implemented in any newly established radiotherapy department as a routine quality measure of TPS to verify its efficacy in performing of inhomogeneity calculation.

Deep learning-based internal gross target volume definition in 4D CT images of lung cancer patients.

Contouring of internal gross target volume (iGTV) is an essential part of treatment planning in radiotherapy to mitigate the impact of intra-fractional target motion. However, it is usually time-consuming and easily subjected to intra-observer and inter-observer variability. So far, few studies have been explored to directly predict iGTV by deep learning technique, because the iGTV contains not only the gross target volume (GTV) but also the motion information of the GTV. This work was an exploratory study to present a deep learning-based framework to segment iGTV rapidly and accurately in 4D CT images for lung cancers. Five models, including 3D UNet, mmUNet with point-wise add merging approach (mmUNet-add), mmUNet with concatenate fusion strategy (mmUNet-cat), gruUNet with point-wise add fusion approach (gruUNet-add), and gruUNet with concatenate method (gruUNet-cat), were adopted for iGTV segmentation. All the models originated from the 3D UNet network, with multi-channel multi-path and convolutional gated recurrent unit (GRU) added in the mmUNet and gruUNet networks, respectively. Seventy patients with lung cancers were collected and 55 cases were randomly selected as the training set, and 15 cases as the testing set. In addition, the segmentation results of the five models were compared with the ground truths qualitatively and quantitatively. In terms of Dice Similarity Coefficient (DSC), the proposed four networks (mmUNet-add, mmUNet-cat, gruUNet-add, and gruUNet-cat) increased the DSC score of 3D UNet from 0.6945 to 0.7342, 0.7253, 0.7405, and 0.7365, respectively. However, the differences were not statistically significant (p > 0.05). After a simple post-processing to remove the small isolated connected regions, the mean 95th percentile Hausdorff distances (HD_95s) of the 3D UNet, mmUNet-add, mmUNet-cat, gruUNet-add, and gruUNet-cat networks were 19.70, 15.75, 15.84, 15.61, and 15.83 mm, respectively, corresponding to 25.35, 25.96, 25.11, 28.23, and 24.47 mm before the post-processing. With regard to runtime, significant elapsed time growths (about 70s and 230s) were observed both in the mmUNet and gruUNet architectures due to the increasing parameters. But the mmUNet structure showed less growth. Our study demonstrated the ability of the deep learning technique to predict iGTVs directly. With the introduction of multi-channel multi-path and convolutional GRU, the segmentation accuracy was improved under certain conditions with a reduced segmentation efficiency and a further research topic when the 3D UNet network would lead to poor performance is elicited. Less efficiency degradation was observed in the mmUNet structure. Besides, the element-wise add fusing strategy was favorable to increase DSC, whereas HD_95 benefited from the concentrate merging approach. Nevertheless, the segmentation accuracy by deep learning still remains to be improved.

Customizable landmark-based field aperture design for automated whole-brain radiotherapy treatment planning.

To develop and evaluate an automated whole-brain radiotherapy (WBRT) treatment planning pipeline with a deep learning-based auto-contouring and customizable landmark-based field aperture design. The pipeline consisted of the following steps: (1) Auto-contour normal structures on computed tomography scans and digitally reconstructed radiographs using deep learning techniques, (2) locate the landmark structures using the beam's-eye-view, (3) generate field apertures based on eight different landmark rules addressing different clinical purposes and physician preferences. Two parallel approaches for generating field apertures were developed for quality control. The performance of the generated field shapes and dose distributions were compared with the original clinical plans. The clinical acceptability of the plans was assessed by five radiation oncologists from four hospitals. The performance of the generated field apertures was evaluated by the Hausdorff distance (HD) and mean surface distance (MSD) from 182 patients' field apertures used in the clinic. The average HD and MSD for the generated field apertures were 16±7 and 7±3mm for the first approach, respectively, and 17±7 and 7±3mm, respectively, for the second approach. The differences regarding HD and MSD between the first and the second approaches were 1±2 and 1±3mm, respectively. A clinical review of the field aperture design, conducted using 30 patients, achieved a 100% acceptance rate for both the first and second approaches, and the plan review achieved a 100% acceptance rate for the first approach and a 93% acceptance rate for the second approach. The average acceptance rate for meeting lens dosimetric recommendations was 80% (left lens) and 77% (right lens) for the first approach, and 70% (both left and right lenses) for the second approach, compared with 50% (left lens) and 53% (right lens) for the clinical plans. This study provided an automated pipeline with two field aperture generation approaches to automatically generate WBRT treatment plans. Both quantitative and qualitative evaluations demonstrated that our novel pipeline was comparable with the original clinical plans.

A predictive model of the effect of therapeutic radiation on the human ovary.

Radiation to the female pelvis as part of treatment for cancer predisposes young women to develop Premature Ovarian Insufficiency (POI). As the human female is born with their full complement of non-growing follicles which decline in an exponential fashion until the menopause, the age at which POI occurs is dependent on the age of the patient at treatment and the dose received by the ovary. A model that predicts the age at which POI occurs for a known dose at a known age will aid counselling patients on their fertility risk. Patients deemed to be at high risk of POI may be considered to be good candidates for established fertility preservation techniques. An updated and externally validated model of the age-related decline in human ovarian reserve was combined with the best available estimate of the median lethal dose LD50 for the human ovary. Using known age at diagnosis and posited radiotherapy treatment plan to estimate the dose to the least-affected ovary, we use an age-related model of the decline in ovarian reserve to generate a personalized age prediction of premature ovarian insufficiency. Our algorithm is available as an online calculator which graphs model outputs to inform discussions around survivor fertility. We report four example cases across different ages and diagnoses, each with two carefully designed photon and proton treatment plans. The treatment options are compared in terms of remaining fertile lifespan for the survivor. International oncology guidelines now mandate the consideration of later fertility when reviewing treatment options for children diagnosed with cancer. Our calculator (https://sites.cs.st-andrews.ac.uk/radiosensitivity), and the underlying algorithm and models, allow detailed predictions of the impact of various radiotherapy plans on fertility. These patient-specific data enhance pre-treatment discussions around post-treatment fertility and fertility preservation.

Structural network alterations in patients with nasopharyngeal carcinoma after radiotherapy: A 1-year longitudinal study.

This longitudinal study explored the changed patterns of structural brain network after radiotherapy (RT) in patients with nasopharyngeal carcinoma (NPC). Diffusion tensor imaging (DTI) data were gathered from 35 patients with NPC at four time points: before RT (baseline), 0∼3 (acute), 6 (early delayed), and 12 months (late-delayed) after RT. The graph theory was used to characterize the dynamic topological properties after RT and the significant changes were detected over time at the global, regional and modular levels. Significantly altered regional metrics (nodal efficiency and degree centrality) were distributed in the prefrontal, temporal, parietal, frontal, and subcortical regions. The module, that exhibited a significantly altered within-module connectivity, had a high overlap with the default mode network (DMN). In addition, the global, regional and modular metrics showed a tendency of progressive decrease at the acute and early delayed stages, and a partial/full recovery at the late-delayed stage. This changed pattern illustrated that the radiation-induced brain damage began at the acute reaction stage and were aggravated at the early-delayed stage, and then partially recovered at the late-delayed stage. Furthermore, the spearman's correlations between the abnormal nodal metrics and temporal dose were calculated and high correlations were found at the temporal (MTG.R and HES.L), subcortical (INS.R), prefrontal (ORBinf.L and ACG.L), and parietal (IPL.R) indicating that these regions were more sensitive to dose and should be mainly considered in radiotherapy treatment plan.

Risk of infections following spleen irradiation–FCCSS study

<h3>Objectives</h3> Infection-related outcomes associated with asplenia or impaired splenic function in childhood cancer survivors (CCS) remain understudied. Splenectomy and splenic radiation have been previously found to be risk factors regarding late infection-related mortality but few data exist concerning overall infections. <h3>Methods</h3> Late infection-related morbidity and mortality was evaluated in the FCCSS cohort including 7670 5-year CCS diagnosed before the age of 21 treated before 2001 using data registered in the national medical database. Using a radiotherapy treatment planning system the dose distribution on different organs had been calculated on CT-based phantom. <h3>Results</h3> 4259 CCS had received radiotherapy, 5762 chemotherapy, and 87 splenectomy. Between 2006-2018: 580 patients were hospitalized at least once for bacterial infection, and 110 patients had severe sepsis or septic shock. 1493 late death are recorded, with an infectious cause in 3.8%(n=57). In a multivariate analysis, spleen irradiation was found to be a risk factor for infection (RR1.8-2.0[CI95%1.2-2.9]), non-significantly modified by radiotherapy, if considered as a binary variable, (RR1.7[CI95%1.4-2.0]) nor lung irradiation (RR1.7-2.1[CI95%1.3-3.2]) nor pituitary irradiation (RR1.5-1.9[CI95%1.1-3.2]).The risk was not related to the dose received to the spleen nor to the volume of spleen involved. Gender, age at diagnosis and chemotherapy were not found to play a significant role in the risk of infection. These results were not altered when considering any regular antibiotic prophylaxis nor vaccine therapy. When analyzing the risk of death by infection, the average dose to the spleen was a risk factor, from doses < 5 Gy, with an increased risk depending on the dose received, as well as a dose > 30 Gy on the pituitary gland. <h3>Conclusion</h3> Splenic radiation was found to significantly increase the risk of late infection-related mortality but not of late severe infection. Nevertheless these data should be taken with caution as 45.6% of the deaths with an infectious cause were patients with a second progressive cancer. The dose to the pituitary gland was also a risk factor that may suggest a link with a hormonal deficit, possibly adrenal, that may increase the risk of decompensation during an infection. This hypothesis must be explored secondarily.

RADT-31. 68GA-DOTATATE PET VERSUS MRI-BASED TREATMENT PLANNING FOR POST-OPERATIVE AND INTACT MENINGIOMA: A DOSIMETRIC ANALYSIS

Abstract Background Nearly all meningiomas express somatostatin receptor 1/2 (SSTR1/SSTR2); therefore, SSTR ligands such as 68Ga-DOTATATE can be utilized for meningioma radiotherapy treatment planning. Incorporation of 68Ga-DOTATATE PET assists with target delineation and may reduce the dose to organs-at-risk (OARs). We hypothesize that 68Ga-DOTATATE PET-based treatment plans will reduce dose to OARs when compared to MRI-based plans. METHODS All treatment plans were rendered on computed tomography (CT) planning datasets, using RapidArc and 6MV photon energy. Two arcs were positioned on the coplanar axis with alternated collimator settings of 0 and 90 degrees to promote MLC sparing with a third unique vertex arc chosen independently to spare normal brain tissue and associated organs at risk (OAR). For MRI structures, a 1 cm expansion from the gross tumor volume (GTV) created a clinical treatment volume (CTV)60. A 2 cm expansion created a CTV54. For PET structures, a 1 cm expansion from the GTV created a CTV60. A 3 mm isotropic expansion created planning treatment volumes (PTVs). Plans were optimized such that Dmax to brainstem remained limited to 60Gy or less and Dmax to optic structures was limited to 54Gy or less. RESULTS 18 meningioma patients were included (9 post-operative, 9 intact). The mean PTV volume using MRI was 258 ccs (306 ccs for post-operative, 210 ccs for intact), and the mean PTV volume using PET was 60 ccs (97 ccs for post-operative, 91 ccs for intact). The mean radiation dose to both optic nerves, the optic chiasm, both hippocampi, and the brainstem was reduced favoring PET-based treatment plans for both post-operative and intact patients. CONCLUSION Our study shows that incorporation of 68Ga-DOTATATE PET can reduce dose to OARs for post-operative and intact patients, and this difference may translate to reduced toxicity. 68Ga-DOTATATE PET guided radiation treatments are worthy of future prospective investigation.

Exploration of dose-toxicity for ophthalmological adverse events in pediatric head and neck rhabdomyosarcoma survivors

<h3>Objectives</h3> Ophthalmological adverse events (OAEs) following local treatment for pediatric head and neck rhabdomyosarcoma (HNRMS) are prevalent. More knowledge on specific radiation dose-toxicity relationships for ophthalmological organs at risk (OARs) in children can help to avoid these OAEs. The aim of the present study is to explore the dose-toxicity probability for multiple OAEs seen in long term HNRMS survivors. <h3>Methods</h3> Clinical data on OAEs were systematically assessed in an international cross-sectional cohort that included survivors of pediatric HNRMS with ≥2 years follow-up. Dosimetric data were collected from the original radiotherapy treatment plans. OARs were delineated on the 3D radiotherapy planning CTs. Dose-toxicity probability curves were constructed using binary logistic regression. <h3>Results</h3> 74 eyes in 38 survivors with a median follow-up time of 9 years (range 2 – 27) were analyzed. With the exception of epiphora, dose-toxicity probability curves with good model-fit (AUC under ROC ≥0.8) were constructed for all OAEs. A >20% probability of globe displacement, dry eyes and/or cataract (any grade) at follow-up was seen for mean doses of >40, 10 and 6 GyEQD2 to the orbital bones, lacrimal gland and lens respectively. Retinopathy and optic neuropathy only occurred in eyes receiving maximum doses > 40 and 53 GyEQD2 to the retina and optic nerve respectively. No multivariable analyses were conducted because of low numbers of events per OAE. <h3>Conclusion</h3> We offer the first detailed description of specific dose-toxicity probability for ophthalmological OARs in children derived from clinically assessed OAEs in long-term survivors. These results are critical to: 1) quantify expected variation in clinical toxicity between different treatment techniques, 2) understand the additional OAEs children will incur if prescription dose for orbital RMS is increased from 45 Gy to 50.4 Gy on future clinical trials, and 3) inform physicians and future patients about expected long-term toxicity and necessary surveillance.

4DCT ventilation function image-based functional lung protection for esophageal cancer radiotherapy.

4DCT (four-dimensional computed tomography) can effectively obtain functional lung ventilation images for patients and integrate them into radiotherapy treatment planning. Studies have not been performed on esophageal cancer, and there is no clear consensus on the optimal functional lung threshold for functional lung. Functional lung images were generated for 11patients with esophageal cancer. The correlation between the dose-volume parameters of functional lung (FL) as defined by different thresholds and the change of PFT/PDFT (pulmonary [diffusion] function test) metrics before and after radiotherapy were evaluated. FL-sparing planning was generated for each patient to preserve the functional lung and compared to conventional anatomical CT (non-sparing) planning. There was asignificant positive correlation between the FL0.8 (defined Jacobian value ≤ 0.8), FL0.84, and FL0.9 dose-volume parameters and ΔFEV1/FVC (reduction before and after radiotherapy), and the FL0.8‑V30 correlation was the strongest (r = 0.819, P < 0.01). The FL-sparing planning had atarget area conformity index and homogeneity index comparable to the non-sparing planning (P > 0.05). For FL, the FL-sparing planning achieved lower FL-MLD (6.30 ± 2.14 Gy vs. 7.83 ± 2.70 Gy), V10 (17.13 ± 7.70% vs. 27.40 ± 9.48%), and V20 (6.96 ± 3.85% vs. 11.63 ± 7.19%) compared to the non-sparing planning (P < 0.05), while heart and spinal cord doses were not significantly different between the two planning groups. The 4DCT-based FL irradiation dose for esophageal cancer was significantly associated with adecrease in FEV1/FVC. The optimal FL defined as a Jacobian value ≤ 0.8 or about 21% of the whole lung volume may be a good choice. FL-sparing planning significantly reduced the FL dose without compromising target area coverage.

Deep learning empowered volume delineation of whole-body organs-at-risk for accelerated radiotherapy

In radiotherapy for cancer patients, an indispensable process is to delineate organs-at-risk (OARs) and tumors. However, it is the most time-consuming step as manual delineation is always required from radiation oncologists. Herein, we propose a lightweight deep learning framework for radiotherapy treatment planning (RTP), named RTP-Net, to promote an automatic, rapid, and precise initialization of whole-body OARs and tumors. Briefly, the framework implements a cascade coarse-to-fine segmentation, with adaptive module for both small and large organs, and attention mechanisms for organs and boundaries. Our experiments show three merits: 1) Extensively evaluates on 67 delineation tasks on a large-scale dataset of 28,581 cases; 2) Demonstrates comparable or superior accuracy with an average Dice of 0.95; 3) Achieves near real-time delineation in most tasks with <2 s. This framework could be utilized to accelerate the contouring process in the All-in-One radiotherapy scheme, and thus greatly shorten the turnaround time of patients.

Evaluation of AI vs. Clinical Experts SBRT-Thorax Computed Tomography OARs Delineation

<h3>Purpose/Objective(s)</h3> Organ-at-risk (OAR) delineation is an essential step for radiotherapy treatment (RT) planning. Standard practice during the last decade referred to manual delineation of OARs from medical experts. Manual delineation is the current standard of practice and can be a tedious, time-consuming process, prone to errors due to intra and inter-observer variations. This is the case for lung stereotactic body radiation therapy (SBRT) due to the variety of structures to be contoured. The latest advances of artificial intelligence methods offer new perspectives towards a fully automatic delineation. This study aims at evaluating an AI-based auto-contouring solution (AC) and compare its clinical acceptability against contours delineated by experts (EC). <h3>Materials/Methods</h3> In this study, a CE/FDA cleared anatomically preserving ensemble deep-learning architecture contouring solution was used. The AI-solution was trained using more than 300+ fully annotated multi-centric SBRT cases according to the ESTRO guidelines (male and female) cases. A fully external cohort of 30 additional patients was considered to assess performance quantitatively and qualitatively. In terms of quantitative metrics, the Dice coefficient was used while qualitative assessment was done through a scoring mechanism: A/acceptable, B/ acceptable after minor corrections, and C/ not acceptable for clinical use. In terms of reference/comparisons with clinical experts, two independent annotations were used while treatment experts' contours were blended blindly with the AI-solution at 50/50 ratio. Random blending at the patient level was performed guaranteeing that, among contours being evaluated per patient and OAR, the 50/50 split was satisfied. <h3>Results</h3> The mean Dice coefficient between expert and AI was 88 % which dropped to 87% among experts. Overall clinical acceptability after aggregating blinded evaluations for the combined categories (A+B) was 95% for the AC which dropped to 81% for EC. When looking at the overall acceptability of contours that do not require any modifications (A), the AC (67%) outperformed the EC (48%) by significant margin. AC outperformed the EC on 14 structures. <h3>Conclusion</h3> This work reports clinical evaluation of AC solution on CT scans for Thorax-SBRT. Our results suggest that the deep learning model could be a viable clinical alternative to the human expert offering treatment standardization and potentially better outcomes.

68Ga-DOTATATE PET vs. MRI-Based Treatment Planning for Post-Operative and Intact Meningioma: A Dosimetric Analysis

<h3>Purpose/Objective(s)</h3> The current standard for radiation treatment planning involves MRI-based image guidance to define intact disease and residual tumor. Because nearly all meningiomas express somatostatin receptor 1/2 (SSTR1/SSTR2), SSTR ligands such as ⁶⁸Ga-DOTATATE are being explored for meningioma radiotherapy treatment planning. Incorporation of ⁶⁸Ga-DOTATATE PET can assist with radiation target volume delineation and reduce the dose to organs-at-risk (OARs). We hypothesize that ⁶⁸Ga-DOTATATE PET-based treatment plans will reduce dose to OARs when compared to MRI-based plans. <h3>Materials/Methods</h3> For each patient, 7 blinded physician contours were retrospectively performed on MRI and PET. A representative MRI and PET physician contour was chosen. All treatment plans were rendered on computed tomography (CT) planning datasets, using VMAT and 6MV photon energy. Each arrangement consisted of 3 independent partial arcs. Two arcs were positioned on the coplanar axis with alternated collimator settings of 0 and 90 degrees to promote MLC sparing with a third unique vertex arc chosen independently to spare normal brain tissue and associated organs at risk (OAR). For MRI structures, a 1 cm expansion from the gross tumor volume (GTV) was used to create a clinical treatment volume (CTV) 60. A 2 cm expansion was used to create a CTV 54. For PET structures, a 1 cm expansion from the GTV was used to create a CTV 60. All CTVs were cropped from outside the brain and a 3 mm isotropic expansion was used to create planning treatment volumes (PTVs). Each plan was prescribed to 95% coverage of prescription dose to the high risk PTV, and dosimetric maximum points to 0.03cc (Dmax) were evaluated for brainstem, optic chiasm, and both optic nerves. Plans were optimized such that Dmax to brainstem remained limited to 60 Gy or less and Dmax to optic structures was limited to 54 Gy or less. <h3>Results</h3> 18 meningioma patients were included (9 post-operative, 9 intact). The mean PTV volume on MRI was 258 ccs (306 ccs for post-operative, 210 ccs for intact), and the mean PTV volume on PET was 60 ccs (97 ccs for post-operative, 91 ccs for intact). The mean radiation dose was reduced to critical structures (Table 1). The mean brain dose was reduced by an average of 42% in PET-based plans. <h3>Conclusion</h3> ⁶⁸Ga-DOTATATE PET has potential to reduce treatment volume and dose to OARs for patients diagnosed with a meningioma. Our study shows that incorporation of ⁶⁸Ga-DOTATATE PET is especially useful in both the post-operative and intact setting and may reduce treatment toxicity. PET guided radiation treatments are worthy of future prospective investigation.

Study on the Effect of Different Bladder Filling Volume on Target Area and Organs at Risk during Three-Dimensional Brachytherapy for Postoperative Early Cervical Cancer

<h3>Purpose/Objective(s)</h3> To investigate the optimal bladder filling volume in the 3D brachytherapy of postoperative cervical cancer. <h3>Materials/Methods</h3> Patients of early cervical cancer with a positive incisal margin or insufficient safety boundary were included. 50ml, 60ml, 70ml, 80ml, 90ml, and 100ml of normal saline were randomly filled into their bladders. There was a total of 6 groups. The CT simulation positioning was performed. According to ICRU89 report, high-risk clinical target volume, bladder, rectum, and other organs at risk were delineated. The radiotherapy treatment planning system was used to make the treatment plan. The volume of HR-CTV, D90, and D2cc and D1cc of organs at risk under different volumes were recorded respectively. <h3>Results</h3> A total of 111 patients were enrolled and there were 398 person-times of 3D brachytherapy treatment program were performed. Among them, there were 66 person-times in the 50ml group, 69 person-times in the 60ml group, 66 person-times in the 70ml group, 69 person-times in the 80ml group, 72 person-times in the 90ml group, and 56 person-times in the 100ml group. Compared with the 60ml group, the volume and dosage of HR-CTV in the groups of 50ml, 70ml, 80ml, 90ml, and 100ml had no significant difference (P>0.05), D2cc and D1cc of the bladder and rectum significantly decreased, the difference was statistically significant (P<0.05). Pairwise comparison between other 5 groups, the volume and dosage of HR-CTV, D2cc and D1cc of bladder and rectum had no significant difference (P>0.05). <h3>Conclusion</h3> In the 3D brachytherapy of postoperative early cervical cancer, compared with other volumes, the bladder perfusion of 60ml can ensure the volume and dose of HR-CTV and reduce better the dose of the bladder and rectum.

Application of 3D Printed Multi-Channel Vaginal Cylinder for Vaginal Brachytherapy in the Cervical Cancer Invading the Middle and Lower Thirds of Vagina

<h3>Purpose/Objective(s)</h3> Patients with cervical cancer invading the middle and lower thirds of vagina (CCIV) had high recurrence rates and worse prognosis. Single-channel vaginal cylinder (SCVC) could make the dose curve wrap completely around the high-risk clinical target volume (HRCTV) for CCIV patients. However, the dose of HRCTV and organs at risk (OARs) would increase at the same time for symmetrical dose distribution of SCVC dose curve, especially in patients with eccentric distribution of the tumor. It was still difficult to achieve a conformal target dose distribution when the SCVC's lead block was used for protecting the OARs. The 3D printed multi-channel vaginal cylinder (3D-MCVC) could arrange the radioactive source according to the tumor location. But there was very little foundation in the literature on dose contrasting in SCVC and MCVC for brachytherapy in CCIV. Therefore, this study proposed to confirm that 3D-MCVC brachytherapy (3D-MCVCB) was better than SCVCB to reduce the OARs dose and achieved conformal target dose distribution. <h3>Materials/Methods</h3> All patients were given the dose of 28Gy/4fraction/4week for brachytherapy. According to ICRU89 document, HRCTV and OARs (the bladder, rectum, and sigmoid colon) were contoured on the CT images in the Nuclear 3D brachytherapy radiotherapy treatment planning system. By self-control study, a simulation plan for the SCVCB was matched while making a 3D-MCVCB plan. The two plans of OARs doses (D_2cc, D_1cc, D_0.1cc) and HRCTVD₉₀ were divided into 3D-MCVCB groups and virtual SCVCB groups, separately. The statistical analysis was the paired sample t-test. <h3>Results</h3> From October 2018 to May 2021, 16 patients of CCIV were included. 128 plans were made for two groups. The prescription dose for HRCTV was 7Gy/1f. When HRCTVD₉₀ was almost 7Gy in the virtual SCVCB group, the doses of OARs were roughly at the threshold of dose limiting. However, when HRCTVD₉₀ of 3D-MCVC group was 7.45±0.47Gy, all OARs doses of 3D-MCVCB group were lower than them of SCVCB group, (P<0.05) (Table1). <h3>Conclusion</h3> The empirical findings in this study provided a new understanding that 3D-MCVCB achieved more conformal target dose distribution for CCIV patient. 3D-MCVCB could better maximize the dose to target tissues and minimize the dose to organs at risk (OARs) than SCVCB. Furthermore, the clinical efficacy and adverse reactions need to be further studied. Table 1. Doses given for HRCTVD₉₀ and OARs for the whole group of plans (n = 128)

Machine Learning to Predict the Quality of a Left-Sided Whole Breast Radiotherapy Treatment Plan

<h3>Purpose/Objective(s)</h3> Develop a Support Vector Machine (SVM) model that uses dosimetric data to predict the quality of a breast radiotherapy treatment plan. <h3>Materials/Methods</h3> One hundred twenty left-sided breast cancer patients from a single institute were included in the study. They were treated with 15-fractions whole breast radiotherapy up to 40.05 Gy with a concomitant boost to deliver 48 Gy to the tumor bed. There was no treatment of lymph nodes. Dose was delivered with a fixed-beam treatment mode and delivered plans were manually created. For each patient, a second treatment plan was generated, using an in-house system for automated plan generation. The treating clinician evaluated the two available plans for each patient and assigned a sentence regarding the quality of the plan in the format of yes/no variable (well optimized plan i.e., "yes" or plan that can be further optimized i.e., "no"). A quantitative scoring tool (ST) already used by Orecchia et al was used to assign 0, 0.5 or 1 points for each involved OAR and PTVs depending on both the fulfilment of multiple DVH constraints and on achieved coverage level for PTVs. Four SVM models were trained using 216 plans (arbitrarily selected from the available 240) using: i) dose volume parameters for target and OARs of each plan, ii) ST and iii) the clinician's plan sentence. Training was performed through iterative optimization using ten Cross-validation folds with four models: a) Model 1 with the ST only, b) Model 2 with the difference between dose volume parameters and the DVH constraints, c) Model 3 with both information from models 1 and 2 and d) Model 4 with ST and dose volume parameter. All models had clinician's sentence as response variable. Finally, the 24/240 patients not used for training were used for final validation. Models' performances were assessed through confusion matrix parameters and multiclass ROC analyses on the training and independent validation set. <h3>Results</h3> The involved clinician refereed as "yes" 92 plans (77%) out of 120 and "no" the remaining 28 in the manual set; 79 plans (66%) were judged "yes" and 41 "no" in the automatic set. Results in terms of Accuracy (ACC), Positive Predicted Values (PPV), True Positive Rates (TPR) and Area under Curve (AUC) for the four models are presented in table 1. <h3>Conclusion</h3> Clinicians' judgements are crucial in daily plan evaluations. We have proposed supervised learning to predict if a plan will be considered by clinician well optimized or not. Results are encouraging and the tool could potentially become useful for aiding treatment planners and reducing clinicians' workload.

Pneumonitis Prediction Modeling of a Prospective 4DCT-Ventilation Functional Avoidance Clinical Trial

<h3>Purpose/Objective(s)</h3> Functional avoidance radiotherapy proposes to use functional imaging to reduce pulmonary toxicity by designing radiotherapy treatment plans that reduce doses to functional regions of the lung. A novel form of lung functional imaging has been proposed that uses 4DCT imaging to calculate 4DCT-based lung ventilation (4DCT-ventilation) maps. A phase II, multi-center, prospective study was completed that showed that 4DCT-ventilation functional avoidance radiotherapy results in decreased ≥ grade 2 pneumonitis when compared to historical controls. Studies are needed that apply machine learning methods to evaluate the large prospective trial dataset to elucidate which factors are most critical in predicting toxicity. The purpose of this work is to use machine learning to predict ≥ grade 2 radiation pneumonitis for patients treated with 4DCT-ventilation functional avoidance radiotherapy. <h3>Materials/Methods</h3> Lung cancer patients receiving curative chemo-radiotherapy (doses of 45-75 Gy) were accrued from 2 institutions. Patient 4DCTs along with image processing techniques were used to generate 4DCT-ventilation images. The 4DCT-ventilation images were used to generate functional avoidance plans that reduced doses to functional portions of the lung while delivering the prescribed tumor dose and respecting tolerances of organs-at-risk. 49 factors were initially used in a machine learning feature selection algorithm, and 36 factors with the highest participation were selected to train 26 classifiers to predict ≥ grade 2 pneumonitis. The assessed factors included patients (age, gender, performance status, etc.), clinical (immunotherapy, surgery, chemotherapy), dose metrics (lung, esophagus, heart, and PTV), and metrics that combine 4DCT-ventilation based function and dose (e.g., V20 to lung portions with ≥50% function). The classifiers were trained using a 5-fold cross-validation scheme. The performance of the classifiers was evaluated using accuracy and receiver operating characteristic area under the curve (AUC). <h3>Results</h3> Of the 67 accrued patients, 10 (14.9%) developed ≥ grade 2 radiation pneumonitis. The best performing classifier was a decision tree classifier with an accuracy of 0.92 and an AUC of 0.75. Dose-function and heart dosimetric indices were the metrics most predictive of ≥ grade 2 pneumonitis. <h3>Conclusion</h3> This is the first study to comprehensively evaluate factors predictive of pneumonitis for patients treated on a prospective 4DCT-ventilation functional avoidance clinical trial. Machine learning methods revealed that lung dose-function metrics and heart doses were significant in predicting toxicity. The results validate the clinical significance in reducing doses to functional portions of the lung and provide seminal clinical guidance for functional avoidance thoracic radiotherapy.

Impact of random outliers in auto-segmented targets on radiotherapy treatment plans for glioblastoma

AimsTo save time and have more consistent contours, fully automatic segmentation of targets and organs at risk (OAR) is a valuable asset in radiotherapy. Though current deep learning (DL) based models are on par with manual contouring, they are not perfect and typical errors, as false positives, occur frequently and unpredictably. While it is possible to solve this for OARs, it is far from straightforward for target structures. In order to tackle this problem, in this study, we analyzed the occurrence and the possible dose effects of automated delineation outliers.MethodsFirst, a set of controlled experiments on synthetically generated outliers on the CT of a glioblastoma (GBM) patient was performed. We analyzed the dosimetric impact on outliers with different location, shape, absolute size and relative size to the main target, resulting in 61 simulated scenarios. Second, multiple segmentation models where trained on a U-Net network based on 80 training sets consisting of GBM cases with annotated gross tumor volume (GTV) and edema structures. On 20 test cases, 5 different trained models and a majority voting method were used to predict the GTV and edema. The amount of outliers on the predictions were determined, as well as their size and distance from the actual target.ResultsWe found that plans containing outliers result in an increased dose to healthy brain tissue. The extent of the dose effect is dependent on the relative size, location and the distance to the main targets and involved OARs. Generally, the larger the absolute outlier volume and the distance to the target the higher the potential dose effect. For 120 predicted GTV and edema structures, we found 1887 outliers. After construction of the planning treatment volume (PTV), 137 outliers remained with a mean distance to the target of 38.5 ± 5.0 mm and a mean size of 1010.8 ± 95.6 mm3. We also found that majority voting of DL results is capable to reduce outliers.ConclusionsThis study shows that there is a severe risk of false positive outliers in current DL predictions of target structures. Additionally, these errors will have an evident detrimental impact on the dose and therefore could affect treatment outcome.

High Risk CTV Asymmetry Predicts Dosimetric Benefit with Use of Combined Intracavitary/Interstitial Brachytherapy Applicators for Cervical Cancer

<h3>Purpose/Objective(s)</h3> Compared to intracavitary (IC) brachytherapy applicators, combined intracavitary/interstitial (IC/IS) applicators provide greater dose to the High-risk CTV (CTV_HR) with similar doses to organ at risk (OAR) structures. Despite increasing use, there is little data to guide clinicians as to which patients are most likely to benefit from use of IC/IS applicators. Recent data from retroEMBRACE suggests CTV_HR volume greater than 30 cm³(cc) is associated with increased likelihood of benefit. In previous work, we provided initial evidence that asymmetry of the CTV_HR may correlate with benefit to use of IC/IS applicators. Here, we provide an expanded analysis, hypothesizing that the degree of asymmetry of the CTV_HR with respect to location of the tandem rod may better correlate with benefit than CTV_HR volume. To test this hypothesis, we performed a dosimetric study of image guided adaptive brachytherapy treatment plans modeled with IC versus IC/IS applicators. <h3>Materials/Methods</h3> Treatment planning MRI or CT datasets for thirty patients treated with IC or IC/IS brachytherapy applicators as part of definitive therapy for FIGO 2018 stage IB3 to IIIC cervical cancer were acquired. IC and IC/IS HDR brachytherapy treatment plans were generated for each dataset using a radiotherapy treatment planning system software from a precision radiation medicine company. For patients treated with IC applicators, needles were virtually digitized and dwell positions modeled based on ovoid location to create IC/IS plans. For patients treated with IC/IS applicators, needle dwell positions were removed and plans re-optimized based on tandem and ovoid dwell positions only. All plans were optimized to minimize dose to the D_2cc volume of OARs while reaching an equivalent dose to the D₉₀ of the CTV_HR. OAR (bladder, rectum, and sigmoid colon), and CTV_HR volumes were not modified from the original administered treatments. The maximum distance (d_max) between the tandem rod and distal edge of the CTV_HR in the axial plan at the level of greatest CTV_HR asymmetry, as determined visually by the author (DS), was obtained for each dataset. Conformity index (CI) was defined as the ratio of the prescription isodose volume to the CTV_HR, and was computed for each plan. Spearman correlation coefficients were used to compare the CI improvement (%CI) in IC/IS versus IC plans, OAR dose reductions, and to measure the correlation between the %CI improvement and the asymmetry surrogate d_max. <h3>Results</h3> Mean %CI improvement in IC/IS plans compared to IC plans was 20.6%. The %CI improvement in IC/IS plans demonstrated a significant correlation with D_2cc dose reduction for bladder (p<0.01), rectum (p<0.01), and sigmoid (p<0.01). The %CI improvement in IC/IS plans demonstrated a significant correlation with d_max (p<0.01). We found no correlation between %CI improvement and volume of the CTV_HR. <h3>Conclusion</h3> During cervical cancer brachytherapy, greater CTV_HR asymmetry, as measured by d_max, correlates with likelihood of benefit from use of IC/IS versus IC brachytherapy applicators.

Development of a computational pregnant female phantom and calculation of fetal dose during a photon breast radiotherapy.

The incidence of carcinoma during pregnancy is reported to be 1:1000-1:1500 pregnancies with the breast carcinoma being the most commonly diagnosed. Since the fetus is most sensitive to ionizing radiation during the first two trimesters, there are mixed clinical opinions and no uniform guidelines on the use of radiotherapy during pregnancy. Within this study the pregnant female phantom in the second trimester, that can be used for radiotherapy treatment planning (as DICOM data), Monte Carlo simulations (as voxelized geometry) and experimental dosimetry utilizing 3D printing of the molds (as .STL files), was developed. The developed phantom is based on MRI images of a female patient in her 18th week of pregnancy and CT images after childbirth. Phantom was developed in such a manner that a pregnant female was scanned "in vivo" using MRI during pregnancy and CT after childbirth. For the treatment of left breast carcinoma, 3D conformal radiotherapy was used. The voxelized geometry of the phantom was used for Monte Carlo (MC) simulations using Monte Carlo N-Particle transport codeTM 6.2 (MCNP). The modeled photon breast radiotherapy plan, applied to the phantom, indicated that the fetus dose is 59 mGy for 50 Gy prescribed to the breast. The results clearly indicate that only 9.5% of the fetal dose is caused by photons that are generated in the accelerator head through scattering and leakage, but the dominant component is scattered radiation from the patient's body.