Dosimetric Comparison and Selection Criteria of Intensity-Modulated Proton Therapy and Intensity-Modulated Radiation Therapy for Adaptive Re-Plan in T3-4 Nasopharynx Cancer Patients.

BACKGROUND: Proton therapy requires caution when treating patients with targets near neural structures. Intuitive and quantitative guidelines are needed to support decision-making concerning the treatment modality. This study compared dosimetric profiles of intensity-modulated proton therapy (IMPT) and intensity-modulated radiation therapy (IMRT) using helical tomotherapy (HT) for adaptive re-planning in cT3-4 nasopharyngeal cancer (NPCa) patients, aiming to establish criteria for selecting appropriate treatment modalities. METHODS: HT and IMPT plans were generated for 28 cT3-4 NPCa patients undergoing definitive radiotherapy. Dosimetric comparisons were performed for target coverage and high-priority organs at risk (OARs). The correlation between dosimetric parameters and RT modality selection was analyzed with the target OAR distances. RESULTS: Target coverages were similar, while IMPT achieved better dose spillage. HT was more favorable for brainstem D1, optic chiasm Dmax, optic nerves Dmax, and p-cord D1. IMPT showed advantages for oral cavity Dmean. Actually, 14 IMPT and 14 HT plans were selected as adaptive plans, with IMPT allocated to most cT3 patients (92.9% vs. 42.9%, p = 0.013). The shortest distances from the target to neural structures were negatively correlated with OAR doses. Receiver operating characteristic curve analyses were carried out to discover the optimal cut-off values of the shortest distances between the target and the OARs (temporal lobes and brainstem), which were 0.75 cm (AUC = 0.908, specificity = 1.00) and 0.85 cm (AUC = 0.857, specificity = 0.929), respectively. CONCLUSIONS: NPCa patients with cT4 tumor or with the shortest distance between the target and critical neural structures < 0.8 cm were suboptimal candidates for IMPT adaptive re-planning. These criteria may improve resource utilization and clinical outcomes.

A marginal structural model for normal tissue complication probability.

The goal of radiation therapy for cancer is to deliver prescribed radiation dose to the tumor while minimizing dose to the surrounding healthy tissues. To evaluate treatment plans, the dose distribution to healthy organs is commonly summarized as dose-volume histograms (DVHs). Normal tissue complication probability (NTCP) modeling has centered around making patient-level risk predictions with features extracted from the DVHs, but few have considered adapting a causal framework to evaluate the safety of alternative treatment plans. We propose causal estimands for NTCP based on deterministic and stochastic interventions, as well as propose estimators based on marginal structural models that impose bivariable monotonicity between dose, volume, and toxicity risk. The properties of these estimators are studied through simulations, and their use is illustrated in the context of radiotherapy treatment of anal canal cancer patients.

A systematic review of 4D magnetic resonance imaging techniques for abdominal radiotherapy treatment planning.

Background and purpose: Four-dimensional magnetic resonance imaging (4DMRI) has gained interest as an alternative to the current standard for motion management four-dimensional tomography (4DCT) in abdominal radiotherapy treatment planning (RTP). This review aims to assess the 4DMRI literature in abdomen, focusing on technical considerations and the validity of using 4DMRI for patients within radiotherapy protocols. Materials and methods: The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A comprehensive search was performed across the Medline, Embase, Scopus, and Web of Science databases, covering all years up to December 31, 2023. The studies were grouped into two categories: 4DMRI reconstructed from 3DMRI acquisition; and 4DMRI reconstructed from multi-slice 2DMRI acquisition. Results: A total of 39 studies met the inclusion criteria and were analysed to provide key findings. Key findings were 4DMRI had the potential to improve abdominal RTP for patients by providing accurate tumour definition and motion assessment compared to 4DCT. 4DMRI reconstructed from 3DMRI acquisition showed promise as a feasible approach for motion management in abdominal RTP regarding spatial resolution. Currently,the slice thickness achieved on 4DMRI reconstructed from multi-slice 2DMRI acquisitions was unsuitable for clinical purposes. Lastly, the current barriers for clinical implementation of 4DMRI were the limited availability of validated commercial solutions and the lack of larger cohort comparative studies to 4DCT for target delineation and plan optimisation. Conclusion: 4DMRI showed potential improvements in abdominal RTP, but standards and guidelines for the use of 4DMRI in radiotherapy were required to demonstrate clinical benefits.

CycleSeg: Simultaneous synthetic CT generation and unsupervised segmentation for MR-only radiotherapy treatment planning of prostate cancer.

BACKGROUND: MR-only radiotherapy treatment planning is an attractive alternative to conventional workflow, reducing scan time and ionizing radiation. It is crucial to derive the electron density map or synthetic CT (sCT) from MR data to perform dose calculations to enable MR-only treatment planning. Automatic segmentation of relevant organs in MR images can accelerate the process by preventing the time-consuming manual contouring step. However, the segmentation label is available only for CT data in many cases. PURPOSE: We propose CycleSeg, a unified framework that generates sCT and corresponding segmentation from MR images without access to MR segmentation labels METHODS: CycleSeg utilizes the CycleGAN formulation to perform unpaired synthesis of sCT and image alignment. To enable MR (sCT) segmentation, CycleSeg incorporates unsupervised domain adaptation by using a pseudo-labeling approach with feature alignment in semantic segmentation space. In contrast to previous approaches that perform segmentation on MR data, CycleSeg could perform segmentation on both MR and sCT. Experiments were performed with data from prostate cancer patients, with 78/7/10 subjects in the training/validation/test sets, respectively. RESULTS: CycleSeg showed the best sCT generation results, with the lowest mean absolute error of 102.2 and the lowest Fréchet inception distance of 13.0. CycleSeg also performed best on MR segmentation, with the highest average dice score of 81.0 and 81.1 for MR and sCT segmentation, respectively. Ablation experiments confirmed the contribution of the proposed components of CycleSeg. CONCLUSION: CycleSeg effectively synthesized CT and performed segmentation on MR images of prostate cancer patients. Thus, CycleSeg has the potential to expedite MR-only radiotherapy treatment planning, reducing the prescribed scans and manual segmentation effort, and increasing throughput.

Radiotherapy of prostate and head-and-neck tumour: An optimal treatment planning comparison for intensity modulated radiotherapy and volumetric modulated arctherapy techniques.

OBJECTIVE: To evaluate and compare merits between intensity modulated radiotherapy and volumetric modulated arc-therapy radiotherapy techniques to determine which technique can achieve better treatment plan quality and efficient delivery. METHODS: The retrospective study was conducted at the Radiation Oncology Department of SanBorotlo Hospital, Vicenza, Italy, in 2019, and comprised data from Jan 2019 to Dec 2019 related to prostate and head-and-neck patients in whom Pinnacle³ treatment planning system was used for optimisation with different prescribed doses and target geometries for intensity modulated radiotherapy and volumetric modulated arc-therapy techniques. Treatment plans were simulated using 6MV photon beam of SynergyS® Linac (Linear accelerator). The plan quality was evaluated using dose-volume indices for planning target-volume and organs-at-risk. ArcCHECK™ phantom was used for dose agreement verification between planed and delivered doses. RESULTS: Data of 8 patients was analysed. Intensity modulated radiotherapy and volumetric modulated arc-therapy treatment plan quality for prostate was found to be similar, but volumetric modulated arc-therapy had significant results for maximum dose (p=0.005). Intensity modulated radiotherapy and volumetric modulated arc-therapy plans for head-and-neck achieved adequate target coverage and sparing of organs at risk, and produced clinically acceptable treatment plans. The percentage of target coverage (p=0.001), dose maximum (p=0.013) and conformity index (p=0.000) were significant. A significant gain for all planning target volume dose-volume indices was noted (p<0.05). Volumetric modulated arc-therapy obtained better plan with significant values and improved sparing of organs at risk compared to intensity modulated radiotherapy for both prostate and head-and-neck treatments while maintaining doses to the organs at risk (p<0.05). CONCLUSIONS: Dynamic arc mode of beam delivery provided increased degrees of freedom of volumetric modulated arc-therapy beam intensity modulation, depicting superior dose distribution than intensity modulated radiotherapy.

Treatment planning for patients with low rectal cancer in a multicenter prospective organ preservation study.

BACKGROUND: Non-surgical management of rectal cancer relies on (chemo)radiotherapy as the definitive treatment modality. This study reports and evaluates the clinical high dose radiotherapy treatment plans delivered to patients with low resectable rectal cancer in a Danish multicenter trial. METHODS: The Danish prospective multicenter phase II Watchful Waiting 2 trial (NCT02438839) investigated definitive chemoradiation for non-surgical management of low rectal cancer. Three Danish centers participated in the trial and committed to protocol-specified treatment planning and delivery requirements. The protocol specified a dose of 50.4 Gy in 28 fractions to the elective volume (CTV-/PTV-E) and a concomitant boost of 62 Gy in 28 fractions to the primary target volume (CTV-/PTV-T). RESULTS: The trial included 108 patients, of which 106 treatment plans were available for retrospective analysis. Dose coverage planning goals for the main target structures were fulfilled for 94% of the treatment plans. However, large intercenter differences in doses to organs-at-risk (OARs) were seen, especially for the intestines. Five patients had a V60Gy>10 cm3 for the intestines and two patients for the bladder. CONCLUSION: Prescribed planning goals for target coverage were fulfilled for 94% of the treatment plans, however analysis of OAR doses and volumes indicated intercenter variations. Dose escalation to 62 Gy (as a concomitant boost to the primary tumor) introduced no substantial high dose volumes (>60 Gy) to the bladder and intestines. The treatment planning goals may be used for future prospective evaluation of highdose radiotherapy for organ preservation for low rectal cancer.

Current status of remote radiotherapy treatment planning in Japan: findings from a national survey†.

The purpose of this study was to investigate the status of remote-radiotherapy treatment planning (RRTP) in Japan through a nationwide questionnaire survey. The survey was conducted between 29 June and 4 August 2022, at 834 facilities in Japan that were equipped with linear accelerators. The survey utilized a Google form that comprised 96 questions on facility information, information about the respondent, utilization of RRTP between facilities, usage for telework and the inclination to implement RRTPs in the respondent's facility. The survey analyzed the utilization of the RRTP system in four distinct implementation types: (i) utilization as a supportive facility, (ii) utilization as a treatment facility, (iii) utilization as a teleworker outside of the facility and (iv) utilization as a teleworker within the facility. The survey response rate was 58.4% (487 facilities responded). Among the facilities that responded, 10% (51 facilities) were implementing RRTP. 13 served as supportive facilities, 23 as treatment facilities, 17 as teleworkers outside of the facility and 5 as teleworkers within the facility. In terms of system usage between supportive and treatment facilities, 70-80% of the participants utilized the system for emergencies or as overtime work for external workers. A substantial number of facilities (38.8%) reported that they were unfamiliar with RRTP implementation. The survey showed that RRTP utilization in Japan is still limited, with a significant number of facilities unfamiliar with the technology. The study highlights the need for greater understanding and education about RRTP and financial funds of economical compensation.

Computed tomography synthesis from magnetic resonance imaging using cycle Generative Adversarial Networks with multicenter learning.

Background and Purpose: Addressing the need for accurate dose calculation in MRI-only radiotherapy, the generation of synthetic Computed Tomography (sCT) from MRI has emerged. Deep learning (DL) techniques, have shown promising results in achieving high sCT accuracies. However, existing sCT synthesis methods are often center-specific, posing a challenge to their generalizability. To overcome this limitation, recent studies have proposed approaches, such as multicenter training . Material and methods: The purpose of this work was to propose a multicenter sCT synthesis by DL, using a 2D cycle-GAN on 128 prostate cancer patients, from four different centers. Four cases were compared: monocenter cases, monocenter training and test on another center, multicenter trainings and a test on a center not included in the training and multicenter trainings with an included center in the test. Trainings were performed using 20 patients. sCT accuracy evaluation was performed using Mean Absolute Error, Mean Error and Peak-Signal-to-Noise-Ratio. Dose accuracy was assessed with gamma index and Dose Volume Histogram comparison. Results: Qualitative, quantitative and dose results show that the accuracy of sCTs for monocenter trainings and multicenter trainings using a seen center in the test did not differ significantly. However, when the test involved an unseen center, the sCT quality was inferior. Conclusions: The aim of this work was to propose generalizable multicenter training for MR-to-CT synthesis. It was shown that only a few data from one center included in the training cohort allows sCT accuracy equivalent to a monocenter study.

A Radiotherapy Dose Map-Guided Deep Learning Method for Predicting Pathological Complete Response in Esophageal Cancer Patients after Neoadjuvant Chemoradiotherapy Followed by Surgery.

Esophageal cancer is a deadly disease, and neoadjuvant chemoradiotherapy can improve patient survival, particularly for patients achieving a pathological complete response (ypCR). However, existing imaging methods struggle to accurately predict ypCR. This study explores computer-aided detection methods, considering both imaging data and radiotherapy dose variations to enhance prediction accuracy. It involved patients with node-positive esophageal squamous cell carcinoma undergoing neoadjuvant chemoradiotherapy and surgery, with data collected from 2014 to 2017, randomly split into five subsets for 5-fold cross-validation. The algorithm DCRNet, an advanced version of OCRNet, integrates RT dose distribution into dose contextual representations (DCR), combining dose and pixel representation with ten soft regions. Among the 80 enrolled patients (mean age 55.68 years, primarily male, with stage III disease and middle-part lesions), the ypCR rate was 28.75%, showing no significant demographic or disease differences between the ypCR and non-ypCR groups. Among the three summarization methods, the maximum value across the CTV method produced the best results with an AUC of 0.928. The HRNetV2p model with DCR performed the best among the four backbone models tested, with an AUC of 0.928 (95% CI, 0.884-0.972) based on 5-fold cross-validation, showing significant improvement compared to other models. This underscores DCR-equipped models' superior AUC outcomes. The study highlights the potential of dose-guided deep learning in ypCR prediction, necessitating larger, multicenter studies to validate the results.

Shifting-field-of-view technique enhancing the inflow effect for identifying tumor/vessel boundaries in MRI for radiotherapy treatment planning.

This study presents two cases of tumors in contact with the inferior vena cava during radiotherapy, and introduces a clinically useful technique for identifying tumor boundaries adjacent to blood vessels by adjusting the position of the field-of-view (FOV) to enhance the inflow effect in magnetic resonance imaging. We named this technique "Shifting-FOV." This method consists of three steps: (1) remove the upper and lower saturation pulses outside the FOV, (2) align the FOV to position the lower edge of the imaging slab as close to the tumor as possible, and (3) manually adjust the table position to locate the tumor at the center of the magnetic field. The proposed method allowed for accurate identification of the tumor/vessel boundaries in both cases. This is a useful technique that can be readily applied to other facilities. Furthermore, images obtained using this technique may enable accurate tumor contouring in radiotherapy treatment planning.

Radiotherapy planning of spine and pelvis using single-energy metal artifact reduction corrected computed tomography sets.

Metal artifacts produce incorrect Hounsfield units and impact treatment planning accuracy. This work evaluates the use of single-energy metal artifact reduction (SEMAR) algorithm for treatment planning by comparison to manual artifact overriding. CT datasets of in-house 3D-printed spine and pelvic phantoms with and without metal insert(s) and two treated patients with metal implants were analysed. CT number accuracy improved with the use of SEMAR filter: root mean square deviation (RMSD) from reference (without metal) reduced by 35.4 in spine and 98.8 in hip. The plan dose volume histograms (DVHs) and dosimetric measurements showed comparable results. SEMAR reconstruction improved planning efficiency.

Automated detection, delineation and quantification of whole-body bone metastasis using FDG-PET/CT images.

Non-small cell lung cancer (NSCLC) patients with the metastatic spread of disease to the bone have high morbidity and mortality. Stereotactic ablative body radiotherapy increases the progression free survival and overall survival of these patients with oligometastases. FDG-PET/CT, a functional imaging technique combining positron emission tomography (PET) with 18 F-fluorodeoxyglucose (FDG) and computer tomography (CT) provides improved staging and identification of treatment response. It is also associated with reduction in size of the radiotherapy tumour volume delineation compared with CT based contouring in radiotherapy, thus allowing for dose escalation to the target volume with lower doses to the surrounding organs at risk. FDG-PET/CT is increasingly being used for the clinical management of NSCLC patients undergoing radiotherapy and has shown high sensitivity and specificity for the detection of bone metastases in these patients. Here, we present a software tool for detection, delineation and quantification of bone metastases using FDG-PET/CT images. The tool extracts standardised uptake values (SUV) from FDG-PET images for auto-segmentation of bone lesions and calculates volume of each lesion and associated mean and maximum SUV. The tool also allows automatic statistical validation of the auto-segmented bone lesions against the manual contours of a radiation oncologist. A retrospective review of FDG-PET/CT scans of more than 30 candidate NSCLC patients was performed and nine patients with one or more metastatic bone lesions were selected for the present study. The SUV threshold prediction model was designed by splitting the cohort of patients into a subset of 'development' and 'validation' cohorts. The development cohort yielded an optimum SUV threshold of 3.0 for automatic detection of bone metastases using FDG-PET/CT images. The validity of the derived optimum SUV threshold on the validation cohort demonstrated that auto-segmented and manually contoured bone lesions showed strong concordance for volume of bone lesion (r = 0.993) and number of detected lesions (r = 0.996). The tool has various applications in radiotherapy, including but not limited to studies determining optimum SUV threshold for accurate and standardised delineation of bone lesions and in scientific studies utilising large patient populations for instance for investigation of the number of metastatic lesions that can be treated safety with an ablative dose of radiotherapy without exceeding the normal tissue toxicity.

Evaluating the effects of variation in CT scanning parameters on the image quality and Hounsfield units for optimization of dose in radiotherapy treatment planning: A semi-anthropomorphic thorax phantom study.

Aim: The diagnosis accuracy of computed tomography (CT) systems and the reliability of calculated Hounsfield Units (HUs) are critical in tumor detection and cancer patients' treatment planning. This study evaluated the effects of scan parameters (Kilovoltage peak or kVp, milli-Ampere-second or mAS reconstruction kernels and algorithms, reconstruction field of view, and slice thickness) on image quality, HUs, and the calculated dose in the treatment planning system (TPS). Materials and Methods: A quality dose verification phantom was scanned several times by a 16-slice Siemens CT scanner. The DOSIsoft ISO gray TPS was applied for dose calculations. The SPSS.24 software was used to analyze the results and the P-value <0.05 was considered significant. Results: Reconstruction kernels and algorithms significantly affected noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR). The noise increased and CNR decreased by raising the sharpness of reconstruction kernels. SNR and CNR had considerable increments at iterative reconstruction compared with the filtered back-projection algorithm. The noise decreased by raising mAS in soft tissues. Also, KVp had a significant effect on HUs. TPS--calculated dose variations were less than 2% for mediastinum and backbone and less than 8% for rib. Conclusions: Although HU variation depends on image acquisition parameters across a clinically feasible range, its dosimetric impact on the calculated dose in TPS can be neglected. Hence, it can be concluded that the optimized values of scan parameters can be applied to obtain the maximum diagnostic accuracy and calculate HUs more precisely without affecting the calculated dose in the treatment planning of cancer patients.

Five-Year Sustained Complete Remission With Minimal Adverse Effects Following Radiosurgery for 2-cm Brain Metastasis With Deep Eloquent Location From Lung Adenocarcinoma Despite Low Marginal Dose and High 12 Gy Volume.

In single-fraction (sf) stereotactic radiosurgery (SRS) for brain metastases (BM) from lung adenocarcinoma (LAC), a marginal dose of ≥22-24 Gy is generally deemed desirable for achieving long-term local tumor control, whereas symptomatic brain radionecrosis significantly increases when the surrounding brain volume receiving ≥12 Gy (V12 Gy) exceeds >5-10 cm3, especially in a deep location. Here, we describe a 75-year-old male with a single LAC-BM of 20 mm in diameter, with a deep eloquent location, which was treated with sfSRS followed by erlotinib, resulting in sustained local complete remission (CR) with minimal adverse radiation effect at nearly five years after sfSRS. The LAC harbored epidermal growth factor receptor (EGFR) mutation. The gross tumor volume (GTV) was defined based on contrast-enhanced computed tomography (CECT) alone. sfSRS was implemented 11 days after planning CECT acquisition. The original GTV had some under- and over-coverage of the enhancing lesion. The D98% values of corrected GTV (cGTV) (3.08 cm3) and 2-mm outside the cGTV were 18.0 Gy with 55% isodose and 14.8 Gy, respectively. The irradiated isodose volumes, including the GTV, receiving ≥22 Gy and ≥12 Gy were 2.18 cm3 and 14.32 cm3, respectively. Erlotinib was administered 13 days after sfSRS with subsequent dose adjustments over 22 months. There was a remarkable tumor response and subsequent nearly CR of the BM were observed at 2.7 and 6.3 months, respectively, with the tumor remnant being visible as a tiny cavitary lesion located in the cortex of the post-central gyrus at 56.4 months. The present case suggests the existence of: (i) extremely radio- and tyrosine kinase inhibitor (TKI)-sensitive LAC-BM for which sfSRS of ≤18 Gy combined with EGFR-TKI is sufficient for attaining long-term CR; and (ii) long-term brain tolerance following sfSRS despite high 12 Gy volume and deep eloquent location in the late 70s The moderate marginal dose of the GTV, the main location of the BM in the cerebral cortex, and the excellent tumor responses with sufficient extrication from the mass effect may render the BM immune to late adverse radiation effect.

Combining distance and anatomical information for deep-learning based dose distribution predictions for nasopharyngeal cancer radiotherapy planning.

Purpose: Deep-learning effectively predicts dose distributions in knowledge-based radiotherapy planning. Using anatomical information that includes a structure map and computed tomography (CT) data as input has been proven to work well. The minimum distance from each voxel in normal structures to planning target volume (DPTV) closely affects each voxel's dose. In this study, we combined DPTV and anatomical information as input for a deep-learning-based dose-prediction network to improve performance. Materials and methods: One hundred patients who underwent volumetric-modulated arc therapy for nasopharyngeal cancer were selected in this study. The prediction model based on a residual network had DPTV maps, structure maps, and CT as inputs and the corresponding dose distribution maps as outputs. The performances of the combined distance and anatomical information (COM) model and the traditional anatomical (ANAT) model with two-channel inputs (structure maps and CT) were compared. A 10-fold cross validation was performed to separately train and test the COM and ANAT models. The voxel-based mean error (ME), mean absolute error (MAE), dosimetric parameters, and dice similarity coefficient (DSC) of isodose volumes were used for modeling evaluation. Results: The mean MAE of the body volume of the COM model were 4.89 ± 1.35%, highly significantly lower than those for the ANAT model of 5.07 ± 1.37% (p<0.001). The ME values of the body for the 2-type models were similar (p >0.05). The mean DSC values of the isodose volumes in the range of 60 Gy were all better in the COM model (p<0.05), and there were highly significant differences between 10 Gy and 55 Gy (p<0.001). For most organs at risk, the ME, MAE, and dosimetric parameters predicted by both models were concurrent with the ground truth values except the MAE values of the pituitary and optic chiasm in the ANAT model and the average mean dose of the right parotid in the ANAT model. Conclusions: The COM model outperformed the ANAT model and could improve automated planning with statistically highly significant differences.

Automated Contouring and Planning in Radiation Therapy: What Is 'Clinically Acceptable'?

Developers and users of artificial-intelligence-based tools for automatic contouring and treatment planning in radiotherapy are expected to assess clinical acceptability of these tools. However, what is 'clinical acceptability'? Quantitative and qualitative approaches have been used to assess this ill-defined concept, all of which have advantages and disadvantages or limitations. The approach chosen may depend on the goal of the study as well as on available resources. In this paper, we discuss various aspects of 'clinical acceptability' and how they can move us toward a standard for defining clinical acceptability of new autocontouring and planning tools.

Input feature design and its impact on the performance of deep learning models for predicting fluence maps in intensity-modulated radiation therapy.

Objective. Deep learning (DL) models for fluence map prediction (FMP) have great potential to reduce treatment planning time in intensity-modulated radiation therapy (IMRT) by avoiding the lengthy inverse optimization process. This study aims to improve the rigor of input feature design in a DL-FMP model by examining how different designs of input features influence model prediction performance.Approach. This study included 231 head-and-neck intensity-modulated radiation therapy patients. Three input feature designs were investigated. The first design (D1) assumed that information of all critical structures from all beam angles should be combined to predict fluence maps. The second design (D2) assumed that local anatomical information was sufficient for predicting radiation intensity of a beamlet at a respective beam angle. The third design (D3) assumed the need for both local anatomical information and inter-beam modulation to predict radiation intensity values of the beamlets that intersect at a voxel. For each input design, we tailored the DL model accordingly. All models were trained using the same set of ground truth plans (GT plans). The plans generated by DL models (DL plans) were analyzed using key dose-volume metrics. One-way ANOVA with multiple comparisons correction (Bonferroni method) was performed (significance level = 0.05).Main results. For PTV-related metrics, all DL plans had significantly higher maximum dose (p < 0.001), conformity index (p < 0.001), and heterogeneity index (p < 0.001) compared to GT plans, with D2 being the worst performer. Meanwhile, except for cord+5 mm (p < 0.001), DL plans of all designs resulted in OAR dose metrics that are comparable to those of GT plans.Significance. Local anatomical information contains most of the information that DL models need to predict fluence maps for clinically acceptable OAR sparing. Input features from beam angles are needed to achieve the best PTV coverage. These results provide valuable insights for further improvement of DL-FMP models and DL models in general.

Image quality comparisons of coil setups in 3T MRI for brain and head and neck radiotherapy simulations.

PURPOSE: MRI is increasingly used for brain and head and neck radiotherapy treatment planning due to its superior soft tissue contrast. Flexible array coils can be arranged to encompass treatment immobilization devices, which do not fit in diagnostic head/neck coils. Selecting a flexible coil arrangement to replace a diagnostic coil should rely on image quality characteristics and patient comfort. We compared image quality obtained with a custom UltraFlexLarge18 (UFL18) coil setup against a commercial FlexLarge4 (FL4) coil arrangement, relative to a diagnostic Head/Neck20 (HN20) coil at 3T. METHODS: The large American College of Radiology (ACR) MRI phantom was scanned monthly in the UFL18, FL4, and HN20 coil setup over 2 years, using the ACR series and three clinical sequences. High-contrast spatial resolution (HCSR), image intensity uniformity (IIU), percent-signal ghosting (PSG), low-contrast object detectability (LCOD), signal-to-noise ratio (SNR), and geometric accuracy were calculated according to ACR recommendations for each series and coil arrangement. Five healthy volunteers were scanned with the clinical sequences in all three coil setups. SNR, contrast-to-noise ratio (CNR) and artifact size were extracted from regions-of-interest along the head for each sequence and coil setup. For both experiments, ratios of image quality parameters obtained with UFL18 or FL4 over those from HN20 were formed for each coil setup, grouping the ACR and clinical sequences. RESULTS: Wilcoxon rank-sum tests revealed significantly higher (p < 0.001) LCOD, IIU and SNR, and lower PSG ratios with UFL18 than FL4 on the phantom for the clinical sequences, with opposite PSG and SNR trends for the ACR series. Similar statistical tests on volunteer data corroborated that SNR ratios with UFL18 (0.58 ± 0.19) were significantly higher (p < 0.001) than with FL4 (0.51 ± 0.18) relative to HN20. CONCLUSIONS: The custom UFL18 coil setup was selected for clinical application in MR simulations due to the superior image quality demonstrated on a phantom and volunteers for clinical sequences and increased volunteer comfort.

Domain knowledge driven 3D dose prediction using moment-based loss function.

Objective.To propose a novel moment-based loss function for predicting 3D dose distribution for the challenging conventional lung intensity modulated radiation therapy plans. The moment-based loss function is convex and differentiable and can easily incorporate clinical dose volume histogram (DVH) domain knowledge in any deep learning (DL) framework without computational overhead.Approach.We used a large dataset of 360 (240 for training, 50 for validation and 70 for testing) conventional lung patients with 2 Gy × 30 fractions to train the DL model using clinically treated plans at our institution. We trained a UNet like convolutional neural network architecture using computed tomography, planning target volume and organ-at-risk contours as input to infer corresponding voxel-wise 3D dose distribution. We evaluated three different loss functions: (1) the popular mean absolute error (MAE) loss, (2) the recently developed MAE + DVH loss, and (3) the proposed MAE + moments loss. The quality of the predictions was compared using different DVH metrics as well as dose-score and DVH-score, recently introduced by theAAPM knowledge-based planning grand challenge. Main results.Model with (MAE + moment) loss function outperformed the model with MAE loss by significantly improving the DVH-score (11%,p< 0.01) while having similar computational cost. It also outperformed the model trained with (MAE + DVH) by significantly improving the computational cost (48%) and the DVH-score (8%,p< 0.01).Significance.DVH metrics are widely accepted evaluation criteria in the clinic. However, incorporating them into the 3D dose prediction model is challenging due to their non-convexity and non-differentiability. Moments provide a mathematically rigorous and computationally efficient way to incorporate DVH information in any DL architecture. The code, pretrained models, docker container, and Google Colab project along with a sample dataset are available on our DoseRTX GitHub (https://github.com/nadeemlab/DoseRTX).

The Impact of System-Related Magnetic Resonance Imaging Geometric Distortion in Stereotactic Radiotherapy: A Case Report.

Magnetic resonance imaging (MRI) is now essential in stereotactic radiotherapy (SRT) planning for brain tumors because of its excellence in soft-tissue contrast and high spatial resolution. However, MRI distortion is sometimes difficult to recognize, and it may cause large misalignments in radiotherapy planning. In this case report, we will show how much difference in the dose distribution of SRT can be made by using MRI without distortion correction.