A review of deep learning based methods for medical image multi-organ segmentation

Deep learning has revolutionized image processing and achieved the-state-of-art performance in many medical image segmentation tasks. Many deep learning-based methods have been published to segment different parts of the body for different medical applications. It is necessary to summarize the current state of development for deep learning in the field of medical image segmentation. In this paper, we aim to provide a comprehensive review with a focus on multi-organ image segmentation, which is crucial for radiotherapy where the tumor and organs-at-risk need to be contoured for treatment planning. We grouped the surveyed methods into two broad categories which are 'pixel-wise classification' and 'end-to-end segmentation'. Each category was divided into subgroups according to their network design. For each type, we listed the surveyed works, highlighted important contributions and identified specific challenges. Following the detailed review, we discussed the achievements, shortcomings and future potentials of each category. To enable direct comparison, we listed the performance of the surveyed works that used thoracic and head-and-neck benchmark datasets.

Artificial intelligence in radiation oncology

Artificial intelligence (AI) is a computer science that tries to mimic human-like intelligence in machines that use computer software and algorithms to perform specific tasks without direct human input. Machine learning (ML) is a subunit of AI that uses data-driven algorithms that learn to imitate human behavior based on a previous example or experience. Deep learning is an ML technique that uses deep neural networks to create a model. The growth and sharing of data, increasing computing power, and developments in AI have initiated a transformation in healthcare. Advances in radiation oncology have produced a significant amount of data that must be integrated with computed tomography imaging, dosimetry, and imaging performed before each fraction. Of the many algorithms used in radiation oncology, has advantages and limitations with different computational power requirements. The aim of this review is to summarize the radiotherapy (RT) process in workflow order by identifying specific areas in which quality and efficiency can be improved by ML. The RT stage is divided into seven stages: patient evaluation, simulation, contouring, planning, quality control, treatment application, and patient follow-up. A systematic evaluation of the applicability, limitations, and advantages of AI algorithms has been done for each stage.

Standard radiotherapy for patients with nasopharyngeal carcinoma results in progressive tract-specific brain white matter alterations: A one-year follow-up via diffusion tensor imaging

BACKGROUND AND PURPOSE

Radiation therapy (RT)-induced neurocognitive disability may be mediated by brain tissue damage. The aim of the present study was to investigate the effects of standard RT on normal brain tissue via in vivo neuroimaging in patients with nasopharyngeal carcinoma (NPC).

MATERIALS AND METHODS

A total of 146 newly diagnosed NPC patients who were treated with standard RT were longitudinally followed up at multiple time points during the first year post-RT, with 19 comparable healthy controls followed up in parallel serving as normal age-related benchmarks. Magnetic resonance diffusion tensor imaging was used to evaluate longitudinal brain white matter tract changes in NPC patients. The relationships between RT-related white matter changes, hippocampal atrophy, and cognitive impairment were also assessed.

RESULTS

Bilateral cingulate angular bundle (CAB) fibers had progressive diffusion reduction [radial diffusivity (RD) and mean diffusivity] over time (P < 0.05, corrected for multiple comparisons) in NPC patients during the first year after RT. RT-associated progressive RD reduction in the left CAB correlated with longitudinal atrophy of the ipsilateral hippocampus (P = 0.033). Additionally, RT-associated progressive RD reduction in the left CAB correlated with progressive cognitive impairment in NPC patients post-RT (P = 0.048).

CONCLUSION

We present evidence of progressive RT-associated changes in the bilateral CAB in NPC patients, which may underlie RT-related cognitive impairment. These findings illustrate that the use of white matter tract alterations as potential biomarkers to detect RT-related brain injury in NPC patients may be useful for better understanding the pathogenesis of RT-induced cognitive decline.

First experience of autonomous, un-supervised treatment planning integrated in adaptive MR-guided radiotherapy and delivered to a patient with prostate cancer

BACKGROUND AND PURPOSE

Currently clinical radiotherapy (RT) planning consists of a multi-step routine procedure requiring human interaction which often results in a time-consuming and fragmented process with limited robustness. Here we present an autonomous un-supervised treatment planning approach, integrated as basis for online adaptive magnetic resonance guided RT (MRgRT), which was delivered to a prostate cancer patient as a first-in-human experience.

MATERIALS AND METHODS

For an intermediate risk prostate cancer patient OARs and targets were automatically segmented using a deep learning-based software and logical volume operators. A baseline plan for the 1.5 T MR-Linac (20x3 Gy) was automatically generated using particle swarm optimization (PSO) without any human interaction. Plan quality was evaluated by predefined dosimetric criteria including appropriate tolerances. Online plan adaptation during clinical MRgRT was defined as first checkpoint for human interaction.

RESULTS

OARs and targets were successfully segmented (3 min) and used for automatic plan optimization (300 min). The autonomous generated plan satisfied 12/16 dosimetric criteria, however all remained within tolerance. Without prior human validation, this baseline plan was successfully used during online MRgRT plan adaptation, where 14/16 criteria were fulfilled. As postulated, human interaction was necessary only during plan adaptation.

CONCLUSION

Autonomous, un-supervised data preparation and treatment planning was first-in-human shown to be feasible for adaptive MRgRT and successfully applied. The checkpoint for first human intervention was at the time of online MRgRT plan adaptation. Autonomous planning reduced the time delay between simulation and start of RT and may thus allow for real-time MRgRT applications in the future.

Report dose-to-medium in clinical trials where available; a consensus from the Global Harmonisation Group to maximize consistency

PURPOSE

To promote consistency in clinical trials by recommending a uniform framework as it relates to radiation transport and dose calculation in water versus in medium.

METHODS

The Global Quality Assurance of Radiation Therapy Clinical Trials Harmonisation Group (GHG; www.rtqaharmonization.org) compared the differences between dose to water in water (D_w,w), dose to water in medium (D_w,m), and dose to medium in medium (D_m,m). This was done based on a review of historical frameworks, existing literature and standards, clinical issues in the context of clinical trials, and the trajectory of radiation dose calculations. Based on these factors, recommendations were developed.

RESULTS

No framework was found to be ideal or perfect given the history, complexity, and current status of radiation therapy. Nevertheless, based on the evidence available, the GHG established a recommendation preferring dose to medium in medium (D_m,m).

CONCLUSIONS

Dose to medium in medium (D_m,m) is the preferred dose calculation and reporting framework. If an institution's planning system can only calculate dose to water in water (D_w,w), this is acceptable.

Technical Challenges of Real-Time Adaptive MR-Guided Radiotherapy

In the past few years, radiotherapy (RT) has experienced a major technological innovation with the development of hybrid machines combining magnetic resonance (MR) imaging and linear accelerators. This new technology for MR-guided cancer treatment has the potential to revolutionize the field of adaptive RT due to the opportunity to provide high-resolution, real-time MR imaging before and during treatment application. However, from a technical point of view, several challenges remain which need to be tackled to ensure safe and robust real-time adaptive MR-guided RT delivery. In this manuscript, several technical challenges to MR-guided RT are discussed. Starting with magnetic field strength tradeoffs, the potential and limitations for purely MR-based RT workflows are discussed. Furthermore, the current status of real-time 3D MR imaging and its potential for real-time RT are summarized. Finally, the potential of quantitative MR imaging for future biological RT adaptation is highlighted.

Potential Value of Radiomics in the Identification of Stage T3 and T4a Esophagogastric Junction Adenocarcinoma Based on Contrast-Enhanced CT Images

Purpose

This study was designed to evaluate the predictive performance of contrast-enhanced CT-based radiomic features for the personalized, differential diagnosis of esophagogastric junction (EGJ) adenocarcinoma at stages T3 and T4a.

Methods

Two hundred patients with T3 (n = 44) and T4a (n = 156) EGJ adenocarcinoma lesions were enrolled in this study. Traditional computed tomography (CT) features were obtained from contrast-enhanced CT images, and the traditional model was constructed using a multivariate logistic regression analysis. A radiomic model was established based on radiomic features from venous CT images, and the radiomic score (Radscore) of each patient was calculated. A combined nomogram diagnostic model was constructed based on Radscores and traditional features. The diagnostic performances of these three models (traditional model, radiomic model, and nomogram) were assessed with receiver operating characteristics curves. Sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and areas under the curve (AUC) of models were calculated, and the performances of the models were evaluated and compared. Finally, the clinical effectiveness of the three models was evaluated by conducting a decision curve analysis (DCA).

Results

An eleven-feature combined radiomic signature and two traditional CT features were constructed as the radiomic and traditional feature models, respectively. The Radscore was significantly different between patients with stage T3 and T4a EGJ adenocarcinoma. The combined nomogram performed the best and has potential clinical usefulness.

Conclusions

The developed combined nomogram might be useful in differentiating T3 and T4a stages of EGJ adenocarcinoma and may facilitate the decision-making process for the treatment of T3 and T4a EGJ adenocarcinoma.

Multicentre, deep learning, synthetic-CT generation for ano-rectal MR-only radiotherapy treatment planning

BACKGROUND AND PURPOSE

Comprehensive dosimetric analysis is required prior to the clinical implementation of pelvic MR-only sites, other than prostate, due to the limited number of site specific synthetic-CT (sCT) dosimetric assessments in the literature. This study aims to provide a comprehensive assessment of a deep learning-based, conditional generative adversarial network (cGAN) model for a large ano-rectal cancer cohort. The following challenges were investigated; T2-SPACE MR sequences, patient data from multiple centres and the impact of sex and cancer site on sCT quality.

METHOD

RT treatment position CT and T2-SPACE MR scans, from two centres, were collected for 90 ano-rectal patients. A cGAN model trained using a focal loss function, was trained and tested on 46 and 44 CT-MR ano-rectal datasets, paired using deformable registration, respectively. VMAT plans were created on CT and recalculated on sCT. Dose differences and gamma indices assessed sCT dosimetric accuracy. A linear mixed effect (LME) model assessed the impact of centre, sex and cancer site.

RESULTS

A mean PTV D95% dose difference of 0.1% (range: -0.5% to 0.7%) was found between CT and sCT. All gamma index (1%/1 mm threshold) measurements were >99.0%. The LME model found the impact of modality, cancer site, sex and centre was clinically insignificant (effect ranges: -0.4% and 0.3%). The mean dose difference for all OAR constraints was 0.1%.

CONCLUSION

Focal loss cGAN models using T2-SPACE MR sequences from multiple centres can produce generalisable, dosimetrically accurate sCTs for ano-rectal cancers.

Automatic dosimetric verification of online adapted plans on the Unity MR-Linac using 3D EPID dosimetry

BACKGROUND AND PURPOSE

The Unity MR-Linac is equipped with an EPID, the images from which contain information about the dose delivered to the patient. The purpose of this study was to introduce a framework for the automatic dosimetric verification of online adapted plans using 3D EPID dosimetry and to present the obtained dosimetric results.

MATERIALS AND METHODS

The framework was active during the delivery of 1207 online adapted plans corresponding to 127 clinical IMRT treatments (74 prostate, 19 rectum, 19 liver and 15 lymph node oligometastases). EPID reconstructed dose distributions in the patient geometry were calculated automatically and then compared to the dose distributions calculated online by the treatment planning system (TPS). The comparison was performed by γ-analysis (3% global/2mm/10% threshold) and by the difference in median dose to the high-dose volume (ΔHDV_D50). 85% for γ-pass rate and 5% for ΔHDV_D50 were used as tolerance limit values.

RESULTS

93% of the online plans were verified automatically by the framework. Missing EPID data was the reason for automation failure. 91% of the verified plans were within tolerance.

CONCLUSION

Automatic dosimetric verification of online adapted plans on the Unity MR-Linac is feasible using in vivo 3D EPID dosimetry. Almost all online adapted plans were approved automatically by the framework. This newly developed framework is a major step forward towards the clinical implementation of a permanent safety net for the entire online adaptive workflow.

Proof-of-concept delivery of intensity modulated arc therapy on the Elekta Unity 1.5 T MR-linac

In this work we present the first delivery of intensity modulated arc therapy on the Elekta Unity 1.5 T MR-linac. The machine's current intensity modulated radiation therapy based control system was modified suitably to enable dynamic delivery of radiation, for the purpose of exploring MRI-guided radiation therapy adaptation modes in a research setting. The proof-of-concept feasibility was demonstrated by planning and delivering two types of plans, each investigating the performance of different parts of a dynamic treatment. A series of fixed-speed arc plans was used to show the high-speed capabilities of the gantry during radiation, while several fully modulated prostate plans-optimised following the volumetric modulated arc therapy approach-were delivered in order to establish the performance of its multi-leaf collimator and diaphragms. These plans were delivered to Delta⁴ Phantom+ MR and film phantoms passing the clinical quality assurance criteria used in our clinic. In addition, we also performed some initial MR imaging experiments during dynamic therapy, demonstrating that the impact of radiation and moving gantry/collimator components on the image quality is negligible. These results show that arc therapy is feasible on the Elekta Unity system. The machine's high performance components enable dynamic delivery during fast gantry rotation and can be controlled in a stable fashion to deliver fully modulated plans.

Marker-less online MR-guided stereotactic body radiotherapy of liver metastases at a 1.5 T MR-Linac - Feasibility, workflow data and patient acceptance

INTRODUCTION

Stereotactic body radiotherapy (SBRT) is an established ablative treatment for liver tumors with excellent local control rates. Magnetic resonance imaging guided radiotherapy (MRgRT) provides superior soft tissue contrast and may therefore facilitate a marker-less liver SBRT workflow. The goal of the present study was to investigate feasibility, workflow parameters, toxicity and patient acceptance of MRgSBRT on a 1.5 T MR-Linac.

METHODS

Ten consecutive patients with liver metastases treated on a 1.5 T MR-Linac were included in this prospective trial. Tumor delineation was performed on four-dimensional computed tomography scans and both exhale triggered and free-breathing T2 MRI scans from the MR-Linac. An internal target volume based approach was applied. Organ at risk constraints were based on the UKSABR guidelines (Version 6.1). Patient acceptance regarding device specific aspects was assessed and toxicity was scored according to the common toxicity criteria of adverse events, version 5.

RESULTS

Nine of ten tumors were clearly visible on the 1.5 T MR-Linac. No patient had fiducial markers placed for treatment. All patients were treated with three or five fractions. Median dose to 98% of the gross tumor volume was 38.5 Gy. The median time from "patient identity check" until "beam-off" was 31 min. Median beam on time was 9.6 min. Online MRgRT was well accepted in general and no treatment had to be interrupted on patient request. No event of symptomatic radiation induced liver disease was observed after a median follow-up of ten month (range 3-17 months).

CONCLUSION

Our early experience suggests that online 1.5 T MRgSBRT of liver metastases represents a promising new non-invasive marker-free treatment modality based on high image quality, clinically reasonable in-room times and high patient acceptance. Further studies are necessary to assess clinical outcome, to validate advanced motion management and to explore the benefit of online response adaptive liver SBRT.

1.5 T MR-linac planning study to compare two different strategies of rectal boost irradiation

PURPOSE

To compare treatment plans of two different rectal boost strategies: up-front versus adaptive boost at the 1.5 T MR-Linac.

METHODS

Patients with locally advanced rectal cancer (LARC) underwent standard neoadjuvant radiochemotherapy with 50.4 Gy in 28 fractions. T2-weighted MRI prior and after the treatment session were acquired to contour gross tumor volumes (GTVs) and organs at risk (OARs). The datasets were used to simulate four different boost strategies (all with 15 Gy/5 fractions in addition to 50.4 Gy): up-front boost (5 daily fractions in the first week of treatment) and an adaptive boost (one boost fraction per week). Both strategies were planned using standard and reduced PTV margins. Intra-fraction motion was assessed by pre- and post-treatment MRI-based contours.

RESULTS

Five patients were included and a total of 44 MRI sets were evaluated. The median PTV volumes of the adaptive boost were significantly smaller than for the up-front boost (81.4 cm3 vs 44.4 cm3 for PTV with standard margins; 31.2 cm3 vs 15 cm3 for PTV with reduced margins; p = 0.031). With reduced margins the rectum was significantly better spared with an adaptive boost rather than with an up-front boost: V60Gy and V65Gy were 41.2% and 24.8% compared with 59% and 29.9%, respectively (p = 0.031). Median GTV intra-fractional motion was 2 mm (range 0-8 mm).

CONCLUSIONS

The data suggest that the adaptive boost strategy exploiting tumor-shrinkage and reduced margin might result in better sparing of rectum and anal canal. Individual margin assessment, motion management and real-time adaptive radiotherapy appear attractive applications of the 1.5 T MR-Linac for further testing of individualized and safe dose escalation in patients with rectal cancer.

MRI-guided Radiation Therapy: An Emerging Paradigm in Adaptive Radiation Oncology

Radiation therapy (RT) continues to be one of the mainstays of cancer treatment. Considerable efforts have been recently devoted to integrating MRI into clinical RT planning and monitoring. This integration, known as MRI-guided RT, has been motivated by the superior soft-tissue contrast, organ motion visualization, and ability to monitor tumor and tissue physiologic changes provided by MRI compared with CT. Offline MRI is already used for treatment planning at many institutions. Furthermore, MRI-guided linear accelerator systems, allowing use of MRI during treatment, enable improved adaptation to anatomic changes between RT fractions compared with CT guidance. Efforts are underway to develop real-time MRI-guided intrafraction adaptive RT of tumors affected by motion and MRI-derived biomarkers to monitor treatment response and potentially adapt treatment to physiologic changes. These developments in MRI guidance provide the basis for a paradigm change in treatment planning, monitoring, and adaptation. Key challenges to advancing MRI-guided RT include real-time volumetric anatomic imaging, addressing image distortion because of magnetic field inhomogeneities, reproducible quantitative imaging across different MRI systems, and biologic validation of quantitative imaging. This review describes emerging innovations in offline and online MRI-guided RT, exciting opportunities they offer for advancing research and clinical care, hurdles to be overcome, and the need for multidisciplinary collaboration.

Advanced analytics and artificial intelligence in gastrointestinal cancer: a systematic review of radiomics predicting response to treatment

Purpose

Advanced medical image analytics is increasingly used to predict clinical outcome in patients diagnosed with gastrointestinal tumors. This review provides an overview on the value of radiomics in predicting response to treatment in patients with gastrointestinal tumors.

Methods

A systematic review was conducted, according to PRISMA guidelines. The protocol was prospectively registered (PROSPERO: CRD42019128408). PubMed, Embase, and Cochrane databases were searched. Original studies reporting on the value of radiomics in predicting response to treatment in patients with a gastrointestinal tumor were included. A narrative synthesis of results was conducted. Results were stratified by tumor type. Quality assessment of included studies was performed, according to the radiomics quality score.

Results

The comprehensive literature search identified 1360 unique studies, of which 60 articles were included for analysis. In 37 studies, radiomics models and individual radiomic features showed good predictive performance for response to treatment (area under the curve or accuracy > 0.75). Various strategies to construct predictive models were used. Internal validation of predictive models was often performed, while the majority of studies lacked external validation. None of the studies reported predictive models implemented in clinical practice.

Conclusion

Radiomics is increasingly used to predict response to treatment in patients suffering from gastrointestinal cancer. This review demonstrates its great potential to help predict response to treatment and improve patient selection and early adjustment of treatment strategy in a non-invasive manner.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00259-020-05142-w.

One layer at a time: the use of 3D printing in the fabrication of cadmium-free electron field shaping devices

Introduction:

Electron blocks are typically composed of a low melting point alloy (LMPA), which is poured into an insert frame containing a manually placed Styrofoam aperture negative used to define the desired field shape. Current implementations of the block fabrication process involve numerous steps which are subjective and prone to user error. Occasionally, bowing of the sides of the insert frame is observed, resulting in premature frame decommissioning. Recent works have investigated the feasibility of utilising 3D printing technology to replace the conventional electron block fabrication workflow; however, these approaches involved long print times, were not compatible with commonly used cadmium-free LMPAs, and did not address the problem of insert frame bowing. In this work, we sought to develop a new 3D printing technique that would remedy these issues.

Materials and Methods:

Electron cutout negatives and alignment jigs were printed using Acrylonitrile Butadiene Styrene, which does not warp at the high temperatures associated with molten cadmium-free alloys. The accuracy of the field shape produced by electron blocks fabricated using the 3D printed negatives was assessed using Gafchromic film and beam profiler measurements. As a proof-of-concept, electron blocks with off-axis apertures, as well as complex multi-aperture blocks to be used for passive electron beam intensity modulation, were also created.

Results:

Film and profiler measurements of field size were in excellent agreement with the values calculated using the Eclipse treatment planning system, showing less than a 1% difference in line profile full-width at half-maximum. The multi-aperture electron blocks produced fields with intensity modulation ≤3.2% of the theoretically predicted value. Use of the 3D printed alignment jig – which has contours designed to match those of the insert frame – was found to reduce the amount of frame bowing by factors of 1.8 and 2.1 in the lateral and superior–inferior directions, respectively.

Conclusions:

The 3D printed ABS negatives generated with our technique maintain their spatial accuracy even at the higher temperatures associated with cadmium-free LMPA. The negatives typically take between 1 and 2 hours to print and have a material cost of approximately $2 per patient.

Single-fraction prostate stereotactic body radiotherapy: Dose reconstruction with electromagnetic intrafraction motion tracking

PURPOSE

To reconstruct the dose delivered during single-fraction urethra-sparing prostate stereotactic body radiotherapy (SBRT) accounting for intrafraction motion monitored by intraprostatic electromagnetic transponders (EMT).

METHODS

We analyzed data of 15 patients included in the phase I/II "ONE SHOT" trial and treated with a single fraction of 19 Gy to the planning target volume (PTV) and 17 Gy to the urethra planning risk volume. During delivery, prostate motion was tracked with implanted EMT. SBRT was interrupted when a 3-mm threshold was trespassed and corrected unless the offset was transient. Motion-encoded reconstructed (MER) plans were obtained by splitting the original plans into multiple sub-beams with isocenter shifts based on recorded EMT positions, mimicking prostate motion during treatment. We analyzed intrafraction motion and compared MER to planned doses.

RESULTS

The median EMT motion range (±SD) during delivery was 0.26 ± 0.09, 0.22 ± 0.14 and 0.18 ± 0.10 cm in the antero-posterior, supero-inferior, and left-right axes, respectively. Treatment interruptions were needed for 8 patients because of target motion beyond limits in the antero-posterior (n = 6) and/or supero-inferior directions (n = 4). Comparing MER vs. original plan there was a median relative dose difference of -1.9% (range, -7.9 to -1.0%) and of +0.5% (-0.3-1.7%) for PTV D_98% and D_2%, respectively. The clinical target volume remained sufficiently covered with a median D_98% difference of -0.3% (-1.6-0.5%). Bladder and rectum dosimetric parameters showed significant differences between original and MER plans, but mostly remained within acceptable limits.

CONCLUSIONS

The dosimetric impact of intrafraction prostate motion was minimal for target coverage for single-fraction prostate SBRT with real-time electromagnetic tracking combined with beam gating.

Plan quality assessment of modern radiotherapy delivery techniques in left-sided breast cancer: an analysis stratified by target delineation guidelines

OBJECTIVE

This study compares planning techniques stratified by consensus delineation guidelines in patients undergoing whole-breast radiotherapy based on an objective plan quality assessment scale.

METHODS

10 patients with left-sided breast cancer were randomly selected, and target delineation for intact breast was performed using Tangent (RTOG 0413), ESTRO, and RTOG guidelines. Consensus Plan Quality Metric (PQM) scoring was defined and communicated to the physicist before commencing treatment planning. Field-in-field IMRT (FiF), inverse IMRT (IMRT) and volumetric modulated arc therapy (VMAT) plans were created for each delineation. Statistical analyses utilised a two-way repeated measures analysis of variance, after applying a Bonferroni correction.

RESULTS

Total PQM score of plans for Tangent and ESTRO were comparable for FiF and IMRT techniques (FiF vs IMRT for Tangent, p = 0.637; FiF vs IMRT for ESTRO, p = 0.304), and were also significantly higher compared to VMAT. Total PQM score of plans for RTOG revealed that IMRT planning achieved a significantly higher score compared to both FiF and VMAT (IMRT vs FiF, p < 0.001; IMRT vs VMAT, p < 0.001).

CONCLUSIONS

Total PQM scores were equivalent for FiF and IMRT for both Tangent and ESTRO delineations, whereas IMRT was best suited for RTOG delineation.

ADVANCES IN KNOWLEDGE

FiF and IMRT planning techniques are best suited for ESTRO or Tangent delineations. IMRT also yields better results with RTOG delineation.

Selective Pressure-Free Treatments for COVID-19

The new severe acute respiratory syndrome coronavirus (SARS-CoV-2) has caused more than 40 million human infections since December 2019, when a cluster of unexplained pneumonia cases was first reported in Wuhan, China. Just a few days after the coronavirus was officially recognized, it was identified as the causative agent of this mysterious pneumonia. This paper discusses the pros and cons of antiviral drugs from the selective pressure and possible drug resistance point of view. We also address the key advantages of potential selective pressure-free treatment methods such as the use of sparsely and densely ionizing low-dose radiation (LDR). It is known that LDR has the capacity to modulate excessive inflammatory responses, regulate lymphocyte counts and control bacterial co-infections in patients with COVID-19 and different modalities. Substantial evidence shows that viruses are constantly mutating and evolving. When an antiviral immune response is unable to eliminate a virus, viral evolution is promoted. Therefore, it is of crucial importance to limit the use of antivirals/vaccines against SARS-CoV-2 when their effects on viral fitness are not fully understood. Furthermore, to limit the spread of the virus, it is essential to develop a vaccine that is available for as many people as possible. However, with the advent of vaccines or new therapies, the new situation may force the virus to evolve. Given this consideration, selective pressure-free treatments for COVID-19 are of great importance.

Evaluation of a cycle-generative adversarial network-based cone-beam CT to synthetic CT conversion algorithm for adaptive radiation therapy

PURPOSE

Image-guided radiation therapy could benefit from implementing adaptive radiation therapy (ART) techniques. A cycle-generative adversarial network (cycle-GAN)-based cone-beam computed tomography (CBCT)-to-synthetic CT (sCT) conversion algorithm was evaluated regarding image quality, image segmentation and dosimetric accuracy for head and neck (H&N), thoracic and pelvic body regions.

METHODS

Using a cycle-GAN, three body site-specific models were priorly trained with independent paired CT and CBCT datasets of a kV imaging system (XVI, Elekta). sCT were generated based on first-fraction CBCT for 15 patients of each body region. Mean errors (ME) and mean absolute errors (MAE) were analyzed for the sCT. On the sCT, manually delineated structures were compared to deformed structures from the planning CT (pCT) and evaluated with standard segmentation metrics. Treatment plans were recalculated on sCT. A comparison of clinically relevant dose-volume parameters (D₉₈, D₅₀ and D₂ of the target volume) and 3D-gamma (3%/3mm) analysis were performed.

RESULTS

The mean ME and MAE were 1.4, 29.6, 5.4 Hounsfield units (HU) and 77.2, 94.2, 41.8 HU for H&N, thoracic and pelvic region, respectively. Dice similarity coefficients varied between 66.7 ± 8.3% (seminal vesicles) and 94.9 ± 2.0% (lungs). Maximum mean surface distances were 6.3 mm (heart), followed by 3.5 mm (brainstem). The mean dosimetric differences of the target volumes did not exceed 1.7%. Mean 3D gamma pass rates greater than 97.8% were achieved in all cases.

CONCLUSIONS

The presented method generates sCT images with a quality close to pCT and yielded clinically acceptable dosimetric deviations. Thus, an important prerequisite towards clinical implementation of CBCT-based ART is fulfilled.

A new dose quantity for evaluation and optimisation of MV photon dose distributions when using advanced algorithms: proof of concept and potential applications

Advanced algorithms used in MV photon radiotherapy model radiation transport in any media. They represent a step forward but introduce new uncertainties and questions, including whether to report the doses to water (D_w,m) or medium (D_m,m) voxels, and the impact of fluence changes introduced by surrounding media. These aspects can compromise consistency between both reporting modes and with previous algorithms in which clinical experience is based. This study introduces a new dose quantity, the dose-to-reference-like medium, to overcome the aforementioned shortcomings. It is linked to a reference medium, water in this study (D_w,m*), and defined as the absorbed dose in a voxel of this reference medium surrounded by a reference-like medium with the same radiation transport characteristics as the original one. We propose to derive D_w,m* distributions by post-processing D_w,m or D_m,m applying a correction factor (CF) to each voxel which depends on its composition. We present and justify a simple and straightforward method to obtain CFs that only involves two phantoms with the same density: one with the considered composition and the other with that of the reference medium. A proof of concept was performed in a clinical environment for Acuros XB (AXB) advanced algorithm and 6 MV photon beams. The CFs were derived for the tissues characterised in Acuros. D_w,m* was compared to D_w,m, D_m,m, and D_w,w from AAA analytical algorithm for some virtual and clinical cases. All the previous quantities presented limitations that can be solved by D_w,m*. This new quantity allows the applicability of evaluation parameters, traceability to clinical experience, and isolation of heterogeneity effects to identify optimum plans, offering useful characteristics for plan evaluation and optimisation in clinical practice. Additionally, it also has potential applications in automated treatment planning and multi-centre activities such as clinical trials, audits, benchmarking, and shared models for automation.

Evaluation of deep learning-based auto-segmentation algorithms for delineating clinical target volume and organs at risk involving data for 125 cervical cancer patients

Objective

To evaluate the accuracy of a deep learning‐based auto‐segmentation mode to that of manual contouring by one medical resident, where both entities tried to mimic the delineation "habits" of the same clinical senior physician.

Methods

This study included 125 cervical cancer patients whose clinical target volumes (CTVs) and organs at risk (OARs) were delineated by the same senior physician. Of these 125 cases, 100 were used for model training and the remaining 25 for model testing. In addition, the medical resident instructed by the senior physician for approximately 8 months delineated the CTVs and OARs for the testing cases. The dice similarity coefficient (DSC) and the Hausdorff Distance (HD) were used to evaluate the delineation accuracy for CTV, bladder, rectum, small intestine, femoral‐head‐left, and femoral‐head‐right.

Results

The DSC values of the auto‐segmentation model and manual contouring by the resident were, respectively, 0.86 and 0.83 for the CTV (P < 0.05), 0.91 and 0.91 for the bladder (P > 0.05), 0.88 and 0.84 for the femoral‐head‐right (P < 0.05), 0.88 and 0.84 for the femoral‐head‐left (P < 0.05), 0.86 and 0.81 for the small intestine (P < 0.05), and 0.81 and 0.84 for the rectum (P > 0.05). The HD (mm) values were, respectively, 14.84 and 18.37 for the CTV (P < 0.05), 7.82 and 7.63 for the bladder (P > 0.05), 6.18 and 6.75 for the femoral‐head‐right (P > 0.05), 6.17 and 6.31 for the femoral‐head‐left (P > 0.05), 22.21 and 26.70 for the small intestine (P > 0.05), and 7.04 and 6.13 for the rectum (P > 0.05). The auto‐segmentation model took approximately 2 min to delineate the CTV and OARs while the resident took approximately 90 min to complete the same task.

Conclusion

The auto‐segmentation model was as accurate as the medical resident but with much better efficiency in this study. Furthermore, the auto‐segmentation approach offers additional perceivable advantages of being consistent and ever improving when compared with manual approaches.

Multivariate normal tissue complication probability models for rectal and bladder morbidity in prostate cancer patients treated with proton therapy

BACKGROUND AND PURPOSE

Normal tissue complication probability (NTCP) models applied for model-based patient selection to proton therapy (PT) have usually been derived using dose/volume histogram (DVH) parameters from photon-based radiotherapy. This study aimed to derive PT-specific multivariate NTCP models that also accounted for the spatial dose distribution (rectum only) as well as non-dose/volume related factors.

MATERIALS AND METHODS

The study included rectum and bladder DVHs, 2D rectal dose maps and relevant patient/treatment characteristics from 1151 prostate cancer cases treated with PT. Prospectively scored Grade 2 late rectal bleeding (CTCAE v3.0, also procedural interventions separately) (n = 156 (15%)) and Grade 3+ GU morbidity (n = 51 (4%)) were entered into a multivariate logistic regression analysis. Model evaluation included assessment of the area under the receiver operating characteristic curve (AUC).

RESULTS

Anticoagulant use was a dominant predictor, chosen in four of the six rectum models and in the bladder model. Age was a dominant predictor in all procedural only rectum models while prostate volume, bladder D5% and V75Gy were predictors in the bladder model. The selection frequency of the dose/volume predictors varied widely, where the percentage of the anterior rectum surface receiving >=75 Gy was the most robust. AUC values ranged from 0.58 to 0.70 across all models, with no clear difference between the DVH- and spatial-based models for the rectum.

CONCLUSION

Anticoagulant use and age were the most prominent predictors in the NTCP models. V75Gy of the rectal wall and the bladder was a predictor in the DVH-based models of the rectum and bladder respectively.

Dose-Volume Effects and Risk Factors for Late Diarrhea in Cervix Cancer Patients After Radiochemotherapy With Image Guided Adaptive Brachytherapy in the EMBRACE I Study

PURPOSE

To evaluate patient- and treatment-related risk factors associated with incidence and persistence of late diarrhea after radiochemotherapy and image guided adaptive brachytherapy (IGABT) in locally advanced cervical cancer.

MATERIALS AND METHODS

Of 1416 patients from the EMBRACE I study, 1199 were prospectively evaluated using physician-reported (Common Terminology Criteria for Adverse Events version 3 [CTCAEv3]) assessment for diarrhea; median follow-up 48 months. Patient-reported outcome (EORTC) was available in 900 patients. Incidence of CTCAE G≥2, G≥3, and EORTC "very much" diarrhea was analyzed with Cox proportional hazards regression. Binary logistic regression was used for analysis of persistent G≥1 and EORTC "quite a bit" - "very much" (≥"quite a bit") diarrhea, defined if present in at least half of all follow-ups.

RESULTS

Crude incidences of G≥2 and G≥3 diarrhea were 8.3% and 1.5%, respectively, and 8% of patients reported "very much" diarrhea. Persistent G≥1 and ≥"quite a bit" diarrhea was present in 16% and 7%, respectively. Patient-related risk factors were baseline diarrhea, smoking, and diabetes with hazard ratios of 1.4 to 7.3. Treatment-related risk factors included prescribed dose, V43 Gy, V57 Gy (lymph node boost), and para-aortic irradiation for external beam radiation therapy (EBRT). G≥2 diarrhea at 3 years increased from 9.5% to 19.9% with prescribed dose 45 Gy versus 50 Gy, 8.7% to 14.0% with V43 Gy <2500 cm³ versus >3000 cm³ and 9.4% to 19.0% with V57 Gy <165 cm³ versus ≥165 cm³. Brachytherapy-related bowel and rectum D_2cm3 were also associated with diarrhea.

CONCLUSION

Dose and volume effects have been established for late diarrhea after radiochemotherapy and IGABT in both CTCAE and EORTC reporting. The risk of diarrhea was lower with a pelvic EBRT prescription of 45 Gy, and higher with larger lymph node boosts volumes (ie, ≥165 cm³). The importance of EBRT volumes as determinants of late toxicity underline the need for continuous quality assurance of target contouring, dose planning, and conformity. The findings of brachytherapy dosimetric factors related to the intestines may become more important with highly conformal EBRT.

The role of computational methods for automating and improving clinical target volume definition

Treatment planning in radiotherapy distinguishes three target volume concepts: the gross tumor volume (GTV), the clinical target volume (CTV), and the planning target volume (PTV). Over time, GTV definition and PTV margins have improved through the development of novel imaging techniques and better image guidance, respectively. CTV definition is sometimes considered the weakest element in the planning process. CTV definition is particularly complex since the extension of microscopic disease cannot be seen using currently available in-vivo imaging techniques. Instead, CTV definition has to incorporate knowledge of the patterns of tumor progression. While CTV delineation has largely been considered the domain of radiation oncologists, this paper, arising from a 2019 ESTRO Physics research workshop, discusses the contributions that medical physics and computer science can make by developing computational methods to support CTV definition. First, we overview the role of image segmentation algorithms, which may in part automate CTV delineation through segmentation of lymph node stations or normal tissues representing anatomical boundaries of microscopic tumor progression. The recent success of deep convolutional neural networks has also enabled learning entire CTV delineations from examples. Second, we discuss the use of mathematical models of tumor progression for CTV definition, using as example the application of glioma growth models to facilitate GTV-to-CTV expansion for glioblastoma that is consistent with neuroanatomy. We further consider statistical machine learning models to quantify lymphatic metastatic progression of tumors, which may eventually improve elective CTV definition. Lastly, we discuss approaches to incorporate uncertainty in CTV definition into treatment plan optimization as well as general limitations of the CTV concept in the case of infiltrating tumors without natural boundaries.

Identification of treatment error types for lung cancer patients using convolutional neural networks and EPID dosimetry

BACKGROUND/PURPOSE

Electronic portal imaging device (EPID) dosimetry aims to detect treatment errors, potentially leading to treatment adaptation. Clinically used threshold classification methods for detecting errors lead to loss of information (from multi-dimensional EPID data to a few numbers) and cannot be used for identifying causes of errors. Advanced classification methods, such as deep learning, can use all available information. In this study, convolutional neural networks (CNNs) were trained to detect and identify error type and magnitude of simulated treatment errors in lung cancer patients. The purpose of this simulation study is to provide a proof-of-concept of CNNs for error identification using EPID dosimetry in an in vivo scenario.

MATERIALS AND METHODS

Clinically realistic ranges of anatomical changes, positioning errors and mechanical errors were simulated for lung cancer patients. Predicted portal dose images (PDIs) containing errors were compared to error-free PDIs using the widely used gamma analysis. CNNs were trained to classify errors using 2D gamma maps. Three classification levels were assessed: Level 1 (main error type, e.g., anatomical change), Level 2 (error subtype, e.g., tumor regression) and Level 3 (error magnitude, e.g., >50% tumor regression).

RESULTS

CNNs showed good performance for all classification levels (training/test accuracy 99.5%/96.1%, 92.5%/86.8%, 82.0%/72.9%). For Level 3, overfitting became more apparent.

CONCLUSION

This simulation study indicates that deep learning is a promising powerful tool for identifying types and magnitude of treatment errors with EPID dosimetry, providing additional information not currently available from EPID dosimetry. This is a first step towards rapid, automated models for identification of treatment errors using EPID dosimetry.

Quantitative magnetic resonance imaging on hybrid magnetic resonance linear accelerators: Perspective on technical and clinical validation

Many preclinical and clinical observations support that functional magnetic resonance imaging (MRI), such as diffusion weighted (DW) and dynamic contrast enhanced (DCE) MRI, might have a predictive value for radiotherapy. The aim of this review was to assess the current status of quantitative MRI on hybrid MR-Linacs. In a literature research, four publications were identified, investigating technical feasibility, accuracy, repeatability and reproducibility of DW and DCE-MRI in phantoms and first patients. Accuracy and short term repeatability was < 5% for DW-MRI in current MR-Linac systems. Consequently, quantitative imaging providing accurate and reproducible functional information seems possible in MR-Linacs.

Machine learning applications in radiation oncology: Current use and needs to support clinical implementation

BACKGROUND AND PURPOSE

The use of artificial intelligence (AI)/ machine learning (ML) applications in radiation oncology is emerging, however no clear guidelines on commissioning of ML-based applications exist. The purpose of this study was therefore to investigate the current use and needs to support implementation of ML-based applications in routine clinical practice.

MATERIALS AND METHODS

A survey was conducted among medical physicists in radiation oncology, consisting of four parts: clinical applications (1), model training, acceptance and commissioning (2), quality assurance (QA) in clinical practice and General Data Protection Regulation (GDPR) (3), and need for education and guidelines (4). Survey answers of medical physicists of the same radiation oncology centre were treated as a separate unique responder in case reporting on different AI applications.

RESULTS

In total, 213 medical physicists from 202 radiation oncology centres were included in the analysis. Sixty-nine percent (1 4 7) was using (37%) or preparing (32%) to use ML in clinic, mostly for contouring and treatment planning. In 86%, human observers were still involved in daily clinical use for quality check of the output of the ML algorithm. Knowledge on ethics, legislation and data sharing was limited and scattered among responders. Besides the need for (implementation) guidelines, training of medical physicists and larger databases containing multicentre data was found to be the top priority to accommodate the further introduction of ML in clinical practice.

CONCLUSION

The results of this survey indicated the need for education and guidelines on the implementation and quality assurance of ML-based applications to benefit clinical introduction.

Clinical experience and workflow challenges with magnetic resonance-only radiation therapy simulation and planning for prostate cancer

BACKGROUND AND PURPOSE

Magnetic Resonance (MR)-only planning has been implemented clinically for radiotherapy of prostate cancer. However, fewer studies exist regarding the overall success rate of MR-only workflows. We report on successes and challenges of implementing MR-only workflows for prostate.

MATERIALS AND METHODS

A total of 585 patients with prostate cancer underwent an MR-only simulation and planning between 06/2016-06/2018. MR simulation included images for contouring, synthetic-CT generation and fiducial identification. Workflow interruptions occurred that required a backup CT, a re-simulation or an update to our current quality assurance (QA) process. The challenges were prospectively evaluated and classified into syn-CT generation, motion/artifacts in the MRs, fiducial QA and bowel preparation guidelines.

RESULTS

MR-only simulation was successful in 544 (93.2 %) patients. . In seventeen patients (2.9%), reconstruction of synthetic-CT failed due to patient size, femur angulation, or failure to determine the body contour. Twenty-four patients (4.1%) underwent a repeat/backup CT scan because of artifacts on the MR such as image blur due to patient motion or biopsy/surgical artifacts that hampered identification of the implanted fiducial markers. In patients requiring large coverage due to nodal involvement, inhomogeneity artifacts were resolved by using a two-stack acquisition and adaptive inhomogeneity correction. Bowel preparation guidelines were modified to address frequent rectum/gas issues due to longer MR scan time.

CONCLUSIONS

MR-only simulation has been successfully implemented for a majority of patients in the clinic. However, MR-CT or CT-only pathway may still be needed for patients where MR-only solution fails or patients with MR contraindications.

In vivo dosimetry in brachytherapy: Requirements and future directions for research, development, and clinical practice

Brachytherapy can deliver high doses to the target while sparing healthy tissues due to its steep dose gradient leading to excellent clinical outcome. Treatment accuracy depends on several manual steps making brachytherapy susceptible to operational mistakes. Currently, treatment delivery verification is not routinely available and has led, in some cases, to systematic errors going unnoticed for years. The brachytherapy community promoted developments in in vivo dosimetry (IVD) through research groups and small companies. Although very few of the systems have been used clinically, it was demonstrated that the likelihood of detecting deviations from the treatment plan increases significantly with time-resolved methods. Time–resolved methods could interrupt a treatment avoiding gross errors which is not possible with time-integrated dosimetry. In addition, lower experimental uncertainties can be achieved by using source-tracking instead of direct dose measurements. However, the detector position in relation to the patient anatomy remains a main source of uncertainty. The next steps towards clinical implementation will require clinical trials and systematic reporting of errors and near-misses. It is of utmost importance for each IVD system that its sensitivity to different types of errors is well understood, so that end-users can select the most suitable method for their needs. This report aims to formulate requirements for the stakeholders (clinics, vendors, and researchers) to facilitate increased clinical use of IVD in brachytherapy. The report focuses on high dose-rate IVD in brachytherapy providing an overview and outlining the need for further development and research.

Radiotherapy Treatment plannINg study Guidelines (RATING): A framework for setting up and reporting on scientific treatment planning studies

Radiotherapy treatment planning studies contribute significantly  to advances and improvements in radiation treatment of cancer patients. They are a pivotal step to support and facilitate the introduction of novel techniques into clinical practice, or as a first step before clinical trials can be carried out. There have been numerous examples published in the literature that demonstrated the feasibility of such techniques as IMRT, VMAT, IMPT, or that compared different treatment methods (e.g. non-coplanar vs coplanar treatment), or investigated planning approaches (e.g. automated planning). However, for a planning study to generate trustworthy new knowledge and give confidence in applying its findings, then its design, execution and reporting all need to meet high scientific standards. This paper provides a 'quality framework' of recommendations and guidelines that can contribute to the quality of planning studies and resulting publications. Throughout the text, questions are posed and, if applicable to a specific study and if met, they can be answered positively in the provided 'RATING' score sheet. A normalised weighted-sum score can then be calculated from the answers as a quality indicator. The score sheet can also be used to suggest how the quality might be improved, e.g. by focussing on questions with high weight, or by encouraging consideration of aspects given insufficient attention. Whilst the overall aim of this framework and scoring system is to improve the scientific quality of treatment planning studies and papers, it might also be used by reviewers and journal editors to help to evaluate scientific manuscripts reporting planning studies.

Radiotherapy in oncological emergencies: fast-track treatment planning

BACKGROUND AND PURPOSE

To report on our clinical experience with a newly implemented workflow for radiotherapy (RT) emergency treatments, which allows for a fast treatment application outside the regular working-hours, and its clinical applicability.

METHODS

Treatment planning of 18 emergency RT patients was carried out using diagnostic computed tomography (CT) without a dedicated RT simulation CT. The cone-beam CT (CBCT) deviations of the first RT treatment were analyzed regarding setup accuracy. Furthermore, feasibility of the "fast-track" workflow was evaluated with respect to dose deviations caused by different Hounsfield unit (HU) to relative electron density (rED) calibrations and RT treatment couch surface shapes via 3D gamma index analysis of exemplary treatment plans. The dosimetric uncertainty introduced by different CT calibrations was quantified.

RESULTS

Mean patient setup vs. CBCT isocenter deviations were (0.49 ± 0.44) cm (x), (2.68 ± 1.63) cm (y) and (1.80 ± 1.06) cm (z) for lateral, longitudinal and vertical directions, respectively. Three out of four dose comparisons between the emergency RT plan calculated on the diagnostic CT and the same plan calculated on the treatment planning CT showed clinically acceptable gamma passing rates, when correcting for surface artifacts. The maximum difference of rED was 0.054, while most parts of the CT calibration curves coincided well.

CONCLUSION

In an emergency RT setting, the use of diagnostic CT data for treatment planning might be time-saving and was shown to be suitable for many cases, considering reproducibility of patient setup, accuracy of initial patient setup and accuracy of dose-calculation.

Impact of computed tomography (CT) reconstruction kernels on radiotherapy dose calculation

PURPOSE

To quantitatively evaluate the effect of computed tomography (CT) reconstruction kernels on various dose calculation algorithms with heterogeneity correction.

METHODS

The gammex electron density (ED) Phantom was scanned with the Siemens PET/CT Biograph20 mCT and reconstructed with twelve different kernel options. Hounsfield unit (HU) vs electron density (ED) curves were generated to compare absolute differences. Scans were repeated under head and pelvis protocols and reconstructed per H40s (head) and B40s (pelvis) kernels. In addition, raw data from a full-body patient scan were also reconstructed using the four B kernels. Per reconstruction, photon (3D and VMAT), electron (18 and 20 MeV) and proton (single field) treatment plans were generated using Varian Eclipse dose calculation algorithms. Photon and electron plans were also simulated to pass through cortical bone vs liver plugs of the phantom for kernel comparison. Treatment field monitor units (MU) and isodose volumes were compared across all scenarios.

RESULTS

The twelve kernels resulted in minor differences in HU, except at the extreme ends of the density curve with a maximum absolute difference of 55.2 HU. The head and pelvis scans of the phantom resulted in absolute HU differences of up to 49.1 HU for cortical bone and 45.1 HU for lung 300, which is a relative difference of 4.1% and 6.2%, respectively. MU comparisons across photon and proton calculation algorithms for the patient and phantom scans were within 1-2 MU, with a maximum difference of 5.4 MU found for the 20 MeV electron plan. The 20MeV electron plan also displayed maximum differences in isodose volumes of 20.4 cc for V90%.

CONCLUSION

Clinically insignificant differences were found among the various kernel generated plans for photon and proton plans calculated on patient and phantom scan data. However, differences in isodose volumes found for higher energy electron plans amongst the kernels may have clinical implications for prescribing dose to an isodose level.

A review of the role of MRI in diagnosis and treatment of early stage lung cancer

Despite magnetic resonance imaging (MRI) being a mainstay in the oncologic care for many disease sites, it has not routinely been used in early lung cancer diagnosis, staging, and treatment. While MRI provides improved soft tissue contrast compared to computed tomography (CT), an advantage in multiple organs, the physical properties of the lungs and mediastinum create unique challenges for lung MRI. Although multi-detector CT remains the gold standard for lung imaging, advances in MRI technology have led to its increased clinical relevance in evaluating early stage lung cancer. Even though positron emission tomography is used more frequently in this context, functional MR imaging, including diffusion-weighted MRI and dynamic contrast-enhanced MRI, are emerging as useful modalities for both diagnosis and evaluation of treatment response for lung cancer. In parallel with these advances, the development of combined MRI and linear accelerator devices (MR-linacs), has spurred the integration of MRI into radiation treatment delivery in the form of MR-guided radiotherapy (MRgRT). Despite challenges for MRgRT in early stage lung cancer radiotherapy, early data utilizing MR-linacs shows potential for the treatment of early lung cancer. In both diagnosis and treatment, MRI is a promising modality for imaging early lung cancer.

Treatment plan quality during online adaptive re-planning

BACKGROUND

Online adaptive radiotherapy is intended to prevent plan degradation caused by inter-fractional tumor volume and shape changes, but time limitations make online re-planning challenging. The aim of this study was to compare the quality of online-adapted plans to their respective reference treatment plans.

METHODS

Fifty-two patients treated on a ViewRay MRIdian Linac were included in this retrospective study. In total 238 online-adapted plans were analyzed, which were optimized with either changing of the segment weights (n = 85) or full re-optimization (n = 153). Five different treatment sites were evaluated: prostate, abdomen, liver, lung and pelvis. Dosimetric parameters of gross tumor volume (GTV), planning target volume (PTV), 2 cm ring around the PTV and organs at risk (OARs) were considered. The Wilcoxon signed-rank test was used to assess differences between online-adapted and reference treatment plans, p < 0.05 was considered significant.

RESULTS

The average duration of the online adaptation, consisting of contour editing, plan optimization and quality assurance (QA), was 24 ± 6 min. The GTV was slightly larger (average ± SD: 1.9% ± 9.0%) in the adapted plans than in the reference plans (p < 0.001). GTV-D95% exhibited no significant changes when considering all plans, but GTV-D2% increased by 0.40% ± 1.5% on average (p < 0.001). There was a very small yet significant decrease in GTV-coverage for the abdomen plans. The ring Dmean increased on average by 1.0% ± 3.6% considering all plans (p < 0.001). There was a significant reduction of the dose to the rectum of 4.7% ± 16% on average (p < 0.001) for prostate plans.

CONCLUSIONS

Dosimetric quality of online-adapted plans was comparable to reference treatment plans and OAR dose was either comparable or decreased, depending on treatment site. However, dose spillage was slightly increased.

Effects of Multileaf Collimator Design and Function When Using an Optimized Dynamic Conformal Arc Approach for Stereotactic Radiosurgery Treatment of Multiple Brain Metastases With a Single Isocenter: A Planning Study

Background Stereotactic radiosurgery (SRS) or fractionated SRS (fSRS) are effective options for the treatment of brain metastases. When treating multiple metastases with a linear accelerator-based approach, a single isocenter allows for efficient treatment delivery. In this study, we present our findings comparing dosimetric parameters of Brainlab (Munich, Germany) Elements™ Multiple Brain Mets SRS (MME) software (version 1.5 versus version 2.0) for a variety of scenarios and patients. The impact of multileaf collimator design and function on plan quality within the software was also evaluated. Materials and methods Twenty previously treated patients with a total of 58 lesions (from one to seven lesions each) were replanned with an updated version of the multiple brain Mets software solution. For each plan, the mean conformity index (CI), mean gradient index (GI), the volume of normal brain receiving 12 Gy (V12), and mean brain dose were evaluated. Additionally, all v2.0 plans were further evaluated with jaw tracking for by Elekta (Stockholm, Sweden) and HD120™ multileaf collimator by Varian Medical Systems (Palo Alto, USA). Results The new software version demonstrated improvements for CI, GI and V12 (p <0.01). For the Elekta Agility™ multileaf collimator, jaw tracking improved all dosimetric parameters except for CI (p =0.178) and mean brain dose (p =0.93). For the Varian with HD120 multileaf collimator, all parameters improved. Conclusions The software enhancements in v2.0 of the software provided improvements in planning efficiency and dosimetric parameters. Differences in multileaf collimator design may provide an additional incremental benefit in a subset of clinical scenarios.

Long-term experience of MPC across multiple TrueBeam linacs: MPC concordance with conventional QC and sensitivity to real-world faults

Machine Performance Check (MPC) is an automated Quality Control (QC) tool that is integrated into the TrueBeam and Halcyon linear accelerators (Linacs), utilizing the imaging systems to verify the Linac beam and geometry. This work compares the concordance of daily MPC results with conventional QC tests over a 3-year period for eight Linacs in order to assess the sensitivity of MPC in detecting faults. The MPC output measurements were compared with the monthly ionization chamber measurements for 6 and 10 MV photon beams and 6, 9, 12, 16, and 18 MeV electron beams. All 6 MV Beam and Geometry (6MVBG) MPC test failures were analyzed to determine the failure rate and the number of true and false negative results, using the conventional QC record as the reference. The concordance between conventional QC test failures and MPC test failures was investigated. The mean agreement across 1933 MPC output and monthly comparison chamber measurements for all beam energies was 0.2%, with 97.8% within 1.5%, and a maximum difference of 2.9%. Of the 5000-6000 MPC individual test parameter results for the 6MVBG test, the highest failure rate was BeamOutputChange (0.5%), then BeamCenterShift (0.3%), and was ≤ 0.1% for the remaining parameters. There were 50 true negative and 27 false negative out of tolerance MPC results, with false negatives resolved by repeating MPC or by independent measurement. The analysis of conventional QC failures demonstrated that MPC detected all failures, except occasions when MPC reported output within tolerance, a result of the MPC-chamber response variation. The variation in MPC output versus chamber measurement indicates MPC is appropriate for daily output constancy but not for the measurement of absolute output. The comparison of the 6MVBG results and conventional records provides evidence that MPC is a sensitive method of performing beam and mechanical checks in a clinical setting.

Comparison of CBCT-based dose calculation methods in head and neck cancer radiotherapy: from Hounsfield unit to density calibration curve to deep learning

PURPOSE

Anatomical variations occur during head and neck (H&N) radiotherapy treatment. kV cone-beam computed tomography (CBCT) images can be used for daily dose monitoring to assess dose variations owing to anatomic changes. Deep learning methods (DLMs) have recently been proposed to generate pseudo-CT (pCT) from CBCT to perform dose calculation. This study aims to evaluate the accuracy of a DLM and to compare this method with three existing methods of dose calculation from CBCT in H&N cancer radiotherapy.

METHODS

Forty-four patients received VMAT for H&N cancer (70-63-56 Gy). For each patient, reference CT (Bigbore, Philips) and CBCT images (XVI, Elekta) were acquired. The DLM was based on a generative adversarial network. The three compared methods were: (a) a method using a density to Hounsfield Unit (HU) relation from phantom CBCT image (HU-D curve method), (b) a water-air-bone density assignment method (DAM), and iii) a method using deformable image registration (DIR). The imaging endpoints were the mean absolute error (MAE) and mean error (ME) of HU from pCT and reference CT (CT_ref ). The dosimetric endpoints were dose discrepancies and 3D gamma analyses (local, 2%/2 mm, 30% dose threshold). Dose discrepancies were defined as the mean absolute differences between DVHs calculated from the CT_ref and pCT of each method.

RESULTS

In the entire body, the MAEs and MEs of the DLM, HU-D curve method, DAM, and DIR method were 82.4 and 17.1 HU, 266.6 and 208.9 HU, 113.2 and 14.2 HU, and 95.5 and -36.6 HU, respectively. The MAE obtained using the DLM differed significantly from those of other methods (Wilcoxon, P ≤ 0.05). The DLM dose discrepancies were 7 ± 8 cGy (maximum = 44 cGy) for the ipsilateral parotid gland D_mean and 5 ± 6 cGy (max = 26 cGy) for the contralateral parotid gland mean dose (D_mean ). For the parotid gland D_mean , no significant dose difference was observed between the DLM and other methods. The mean 3D gamma pass rate ± standard deviation was 98.1 ± 1.2%, 91.0 ± 5.3%, 97.9 ± 1.6%, and 98.8 ± 0.7% for the DLM, HU-D method, DAM, and DIR method, respectively. The gamma pass rates and mean gamma results of the HU-D curve method, DAM, and DIR method differed significantly from those of the DLM.

CONCLUSIONS

For H&N radiotherapy, DIR method and DLM appears as the most appealing CBCT-based dose calculation methods among the four methods in terms of dose accuracy as well as calculation time. Using the DIR method or DLM with CBCT images enables dose monitoring in the parotid glands during the treatment course and may be used to trigger replanning.

First system for fully-automated multi-criterial treatment planning for a high-magnetic field MR-Linac applied to rectal cancer

Background and purpose: In this study we developed a workflow for fully-automated generation of deliverable IMRT plans for a 1.5 T MR-Linac (MRL) based on contoured CT scans, and we evaluated automated MRL planning for rectal cancer.Methods: The Monte Carlo dose calculation engine used in the clinical MRL TPS (Monaco, Elekta AB, Stockholm, Sweden), suited for high accuracy dose calculations in a 1.5 T magnetic field, was coupled to our in-house developed Erasmus-iCycle optimizer. Clinically deliverable plans for 23 rectal cancer patients were automatically generated in a two-step process, i.e., multi-criterial fluence map optimization with Erasmus-iCycle followed by a conversion into a deliverable IMRT plan in the clinical TPS. Automatically generated plans (AUTOplans) were compared to plans that were manually generated with the clinical TPS (MANplans).Results: With AUTOplanning large reductions in planning time and workload were obtained; 4-6 h mainly hands-on planning for MANplans vs ∼1 h of mainly computer computation time for AUTOplans. For equal target coverage, the bladder and bowel bag D_mean was reduced in the AUTOplans by 1.3 Gy (6.9%) on average with a maximum reduction of 4.5 Gy (23.8%). Dosimetric measurements at the MRL demonstrated clinically acceptable delivery accuracy for the AUTOplans.Conclusions: A system for fully automated multi-criterial planning for a 1.5 T MR-Linac was developed and tested for rectal cancer patients. Automated planning resulted in major reductions in planning workload and time, while plan quality improved. Negative impact of the high magnetic field on the dose distributions could be avoided.

Recent advanced in Surface Guided Radiation Therapy

The growing acceptance and recognition of Surface Guided Radiation Therapy (SGRT) as a promising imaging technique has supported its recent spread in a large number of radiation oncology facilities. Although this technology is not new, many aspects of it have only recently been exploited. This review focuses on the latest SGRT developments, both in the field of general clinical applications and special techniques.SGRT has a wide range of applications, including patient positioning with real-time feedback, patient monitoring throughout the treatment fraction, and motion management (as beam-gating in free-breathing or deep-inspiration breath-hold). Special radiotherapy modalities such as accelerated partial breast irradiation, particle radiotherapy, and pediatrics are the most recent SGRT developments.The fact that SGRT is nowadays used at various body sites has resulted in the need to adapt SGRT workflows to each body site. Current SGRT applications range from traditional breast irradiation, to thoracic, abdominal, or pelvic tumor sites, and include intracranial localizations.Following the latest SGRT applications and their specifications/requirements, a stricter quality assurance program needs to be ensured. Recent publications highlight the need to adapt quality assurance to the radiotherapy equipment type, SGRT technology, anatomic treatment sites, and clinical workflows, which results in a complex and extensive set of tests.Moreover, this review gives an outlook on the leading research trends. In particular, the potential to use deformable surfaces as motion surrogates, to use SGRT to detect anatomical variations along the treatment course, and to help in the establishment of personalized patient treatment (optimized margins and motion management strategies) are increasingly important research topics. SGRT is also emerging in the field of patient safety and integrates measures to reduce common radiotherapeutic risk events (e.g. facial and treatment accessories recognition).This review covers the latest clinical practices of SGRT and provides an outlook on potential applications of this imaging technique. It is intended to provide guidance for new users during the implementation, while triggering experienced users to further explore SGRT applications.

Treatment of radiation-induced maculopathy with fluocinolone acetonide

PURPOSE

Chronic macular oedema is a well-known presentation of radiation-induced maculopathy (RM) following external beam photon therapy, plaque radiotherapy and proton beam radiotherapy for choroidal tumours. Current therapies vary in respect of efficacy and clinical benefit. The potential of fluocinolone acetonide (FAc) slow-release implants is unknown. We hypothesised that local continuous delivery of low-dose corticosteroids might improve symptoms of RM.

METHODS

Five-two male and three female-patients from 37 to 68 years presented with RM following ¹⁰⁶Ru-plaque brachytherapy or stereotactic radiation therapy (STx) with photons using a hypofractionated schedule of 5 × 10 Gy. All were treated with triamcinolone injections in first line and proofed to be refractory to steroids. In addition, two patients had received Ozurdex® implants as a second-line treatment, though without any clinical benefit. FAc slow-release implants were injected, and patients were followed up to monitor clinical improvement.

RESULTS

All patients responded to therapy by means of a decrease in macular oedema. In four of five (80%) patients, visual acuity improved, and one patient showed stable visual acuity. No toxic effects or complications were observed.

CONCLUSION

Slow-release implants of FAc are a promising therapeutic potent steroid treatment option to benefit anatomical structures of the fovea and visual function. Slow-release implants with FAc reduce the frequency of intravitreal injections and the therapeutic burden.

Prostate intrafraction motion during the preparation and delivery of MR-guided radiotherapy sessions on a 1.5T MR-Linac

PURPOSE

To evaluate prostate intrafraction motion using MRI during the full course of online adaptive MR-Linac radiotherapy (RT) fractions, in preparation of MR-guided extremely hypofractionated RT.

MATERIAL AND METHODS

Five low and intermediate risk prostate cancer patients were treated with 20 × 3.1 Gy fractions on a 1.5T MR-Linac. Each fraction, initial MRI (Pre) scans were obtained at the start of every treatment session. Pre-treatment planning MRI contours were propagated and adapted to this Pre scan after which plan re-optimization was started in the treatment planning system followed by dose delivery. 3D Cine-MR imaging was started simultaneously with beam-on and acquired over the full beam-on period. Prostate intrafraction motion in this cine-MR was determined with a previously validated soft-tissue contrast based tracking algorithm. In addition, absolute accuracy of the method was determined using a 4D phantom.

RESULTS

Prostate motion was completely automatically determined over the full on-couch period (approx. 45 min) with no identified mis-registrations. The translation 95% confidence intervals are within clinically applied margins of 5 mm, and plan adaption for intrafraction motion was required in only 4 out of 100 fractions.

CONCLUSION

This is the first study to investigate prostate intrafraction motions during entire MR-guided RT sessions on an MR-Linac. We have shown that high quality 3D cine-MR imaging and prostate tracking during RT is feasible with beam-on. The clinically applied margins of 5 mm have proven to be sufficient for these treatments and may potentially be further reduced using intrafraction plan adaptation guided by cine-MR imaging.

Delivered dose quantification in prostate radiotherapy using online 3D cine imaging and treatment log files on a combined 1.5T magnetic resonance imaging and linear accelerator system

BACKGROUND AND PURPOSE

Monitoring the intrafraction motion and its impact on the planned dose distribution is of crucial importance in radiotherapy. In this work we quantify the delivered dose for the first prostate patients treated on a combined 1.5T Magnetic Resonance Imaging (MRI) and linear accelerator system in our clinic based on online 3D cine-MR and treatment log files.

MATERIALS AND METHODS

A prostate intrafraction motion trace was obtained with a soft-tissue based rigid registration method with six degrees of freedom from 3D cine-MR dynamics with a temporal resolution of 8.5-16.9 s. For each fraction, all dynamics were also registered to the daily MR image used during the online treatment planning, enabling the mapping to this reference point. Moreover, each fraction's treatment log file was used to extract the timestamped machine parameters during delivery and assign it to the appropriate dynamic volume. These partial plans to dynamic volume combinations were calculated and summed to yield the delivered fraction dose. The planned and delivered dose distributions were compared among all patients for a total of 100 fractions.

RESULTS

The clinical target volume underwent on average a decrease of 2.2% ± 2.9% in terms of D99% coverage while bladder V62Gy was increased by 1.6% ± 2.3% and rectum V62Gy decreased by 0.2% ± 2.2%.

CONCLUSIONS

The first MR-linac dose reconstruction results based on prostate tracking from intrafraction 3D cine-MR and treatment log files are presented. Such a pipeline is essential for online adaptation especially as we progress to MRI-guided extremely hypofractionated treatments.

A study of the clinical, treatment planning and dosimetric feasibility of dose painting in external beam radiotherapy of prostate cancer

BACKGROUND AND PURPOSE

Radiotherapy dose painting is a promising technique which enables dose escalation to areas of higher tumour cell density within the prostate which are associated with radioresistance, known as dominant intraprostatic lesions (DILs). The aim of this study was to determine factors affecting the feasibility of radiotherapy dose painting in patients with high and intermediate risk prostate cancer.

MATERIALS & METHODS

Twenty patients were recruited into the study for imaging using a 3 T magnetic resonance imaging (MRI) scanner. Identified DILs were outlined and the scan registered with the planning computed tomography (CT) dataset. Intensity-modulated plans were produced and evaluated to determine the effect of the organ-at-risk constraints on the dose that could be delivered to the DILs. Measurements were made to verify that the distribution could be safely delivered.

RESULTS

MRI scans were obtained for nineteen patients. Fourteen patients had one to two DILs with ten overlapping the urethra and/or rectum. The target boost of 86 Gy was achieved in seven plans but was limited to 80 Gy for five patients whose boost volume overlapped or abutted the urethra. Dosimetric measurements gave a satisfactory gamma pass rate at 3%/3 mm.

CONCLUSIONS

It was feasible to produce dose-painted plans for a boost of 86 Gy for approximately half the patients with DILs. The main limiting factor was the proximity of the urethra to the boost volumes. For a small proportion of patients, rigid registration between CT and MRI images was not adequate for planning purposes.

Single-Fraction Stereotactic Body Radiation Therapy: A Paradigm During the Coronavirus Disease 2019 (COVID-19) Pandemic and Beyond?

PURPOSE

Owing to the coronavirus disease 2019 (COVID-19) pandemic, radiation oncology departments have adopted various strategies to deliver radiation therapy safely and efficiently while minimizing the risk of severe acute respiratory syndrome coronavirus-2 transmission among patients and health care providers. One practical strategy is to deliver stereotactic body radiation therapy (SBRT) in a single fraction, which has been well established for treating bone metastases, although it has been infrequently used for other extracranial sites.

METHODS AND MATERIALS

A PubMed search of published articles in English related to single-fraction SBRT was performed. A critical review was performed of the articles that described clinical outcomes of single-fraction SBRT for treatment of primary extracranial cancers and oligometastatic extraspinal disease.

RESULTS

Single-fraction SBRT for peripheral early-stage non-small cell lung cancer is supported by randomized data and is strongly endorsed during the COVID-19 pandemic by the European Society for Radiotherapy and Oncology-American Society for Radiation Oncology practice guidelines. Prospective and retrospective studies supporting a single-fraction regimen are limited, although outcomes are promising for renal cell carcinoma, liver metastases, and adrenal metastases. Data are immature for primary prostate cancer and demonstrate excess late toxicity in primary pancreatic cancer.

CONCLUSIONS

Single-fraction SBRT should be strongly considered for peripheral early-stage non-small cell lung cancer during the COVID-19 pandemic to mitigate the potentially severe consequences of severe acute respiratory syndrome coronavirus-2 transmission. Although single-fraction SBRT is promising for the definitive treatment of other primary or oligometastatic cancers, multi-fraction SBRT should be the preferred regimen owing to the need for additional prospective evaluation to determine long-term efficacy and safety.

J. Kotte AN, Bol GH, T. van den Berg CA. Clinical implementation of MRI-based organs-at-risk auto-segmentation with convolutional networks for prostate radiotherapy

BACKGROUND

Structure delineation is a necessary, yet time-consuming manual procedure in radiotherapy. Recently, convolutional neural networks have been proposed to speed-up and automatise this procedure, obtaining promising results. With the advent of magnetic resonance imaging (MRI)-guided radiotherapy, MR-based segmentation is becoming increasingly relevant. However, the majority of the studies investigated automatic contouring based on computed tomography (CT).

PURPOSE

In this study, we investigate the feasibility of clinical use of deep learning-based automatic OARs delineation on MRI.

MATERIALS AND METHODS

We included 150 patients diagnosed with prostate cancer who underwent MR-only radiotherapy. A three-dimensional (3D) T1-weighted dual spoiled gradient-recalled echo sequence was acquired with 3T MRI for the generation of the synthetic-CT. The first 48 patients were included in a feasibility study training two 3D convolutional networks called DeepMedic and dense V-net (dV-net) to segment bladder, rectum and femurs. A research version of an atlas-based software was considered for comparison. Dice similarity coefficient, 95% Hausdorff distances (HD95), and mean distances were calculated against clinical delineations. For eight patients, an expert RTT scored the quality of the contouring for all the three methods. A choice among the three approaches was made, and the chosen approach was retrained on 97 patients and implemented for automatic use in the clinical workflow. For the successive 53 patients, Dice, HD95 and mean distances were calculated against the clinically used delineations.

RESULTS

DeepMedic, dV-net and the atlas-based software generated contours in 60 s, 4 s and 10-15 min, respectively. Performances were higher for both the networks compared to the atlas-based software. The qualitative analysis demonstrated that delineation from DeepMedic required fewer adaptations, followed by dV-net and the atlas-based software. DeepMedic was clinically implemented. After retraining DeepMedic and testing on the successive patients, the performances slightly improved.

CONCLUSION

High conformality for OARs delineation was achieved with two in-house trained networks, obtaining a significant speed-up of the delineation procedure. Comparison of different approaches has been performed leading to the succesful adoption of one of the neural networks, DeepMedic, in the clinical workflow. DeepMedic maintained in a clinical setting the accuracy obtained in the feasibility study.

MRI image analysis methods and applications: an algorithmic perspective using brain tumors as an exemplar

The use of magnetic resonance imaging (MRI) in healthcare and the emergence of radiology as a practice are both relatively new compared with the classical specialties in medicine. Having its naissance in the 1970s and later adoption in the 1980s, the use of MRI has grown exponentially, consequently engendering exciting new areas of research. One such development is the use of computational techniques to analyze MRI images much like the way a radiologist would. With the advent of affordable, powerful computing hardware and parallel developments in computer vision, MRI image analysis has also witnessed unprecedented growth. Due to the interdisciplinary and complex nature of this subfield, it is important to survey the current landscape and examine the current approaches for analysis and trend trends moving forward.

MR-PROTECT: Clinical feasibility of a prostate MRI-only radiotherapy treatment workflow and investigation of acceptance criteria

BACKGROUND

Retrospective studies on MRI-only radiotherapy have been presented. Widespread clinical implementations of MRI-only workflows are however limited by the absence of guidelines. The MR-PROTECT trial presents an MRI-only radiotherapy workflow for prostate cancer using a new single sequence strategy. The workflow incorporated the commercial synthetic CT (sCT) generation software MriPlanner™ (Spectronic Medical, Helsingborg, Sweden). Feasibility of the workflow and limits for acceptance criteria were investigated for the suggested workflow with the aim to facilitate future clinical implementations.

METHODS

An MRI-only workflow including imaging, post imaging tasks, treatment plan creation, quality assurance and treatment delivery was created with questionnaires. All tasks were performed in a single MR-sequence geometry, eliminating image registrations. Prospective CT-quality assurance (QA) was performed prior treatment comparing the PTV mean dose between sCT and CT dose-distributions. Retrospective analysis of the MRI-only gold fiducial marker (GFM) identification, DVH- analysis, gamma evaluation and patient set-up verification using GFMs and cone beam CT were performed.

RESULTS

An MRI-only treatment was delivered to 39 out of 40 patients. The excluded patient was too large for the predefined imaging field-of-view. All tasks could successfully be performed for the treated patients. There was a maximum deviation of 1.2% in PTV mean dose was seen in the prospective CT-QA. Retrospective analysis showed a maximum deviation below 2% in the DVH-analysis after correction for rectal gas and gamma pass-rates above 98%. MRI-only patient set-up deviation was below 2 mm for all but one investigated case and a maximum of 2.2 mm deviation in the GFM-identification compared to CT.

CONCLUSIONS

The MR-PROTECT trial shows the feasibility of an MRI-only prostate radiotherapy workflow. A major advantage with the presented workflow is the incorporation of a sCT-generation method with multi-vendor capability. The presented single sequence approach are easily adapted by other clinics and the general implementation procedure can be replicated. The dose deviation and the gamma pass-rate acceptance criteria earlier suggested was achievable, and these limits can thereby be confirmed. GFM-identification acceptance criteria are depending on the choice of identification method and slice thickness. Patient positioning strategies needs further investigations to establish acceptance criteria.