Artificial Intelligence in magnetic Resonance guided Radiotherapy: Medical and physical considerations on state of art and future perspectives

Radiomics-enhanced early regression index for predicting treatment response in rectal cancer: a multi-institutional 0.35 T MRI-guided radiotherapy study

PURPOSE

The accurate prediction of treatment response in locally advanced rectal cancer (LARC) patients undergoing MRI-guided radiotherapy (MRIgRT) is essential for optimising treatment strategies. This multi-institutional study aimed to investigate the potential of radiomics in enhancing the predictive power of a known radiobiological parameter (Early Regression Index, ERITCP) to evaluate treatment response in LARC patients treated with MRIgRT.

METHODS

Patients from three international sites were included and divided into training and validation sets. 0.35 T T2*/T1-weighted MR images were acquired during simulation and at each treatment fraction. The biologically effective dose (BED) conversion was used to account for different radiotherapy schemes: gross tumour volume was delineated on the MR images corresponding to specific BED levels and radiomic features were then extracted. Multiple logistic regression models were calculated, combining ERITCP with other radiomic features. The predictive performance of the different models was evaluated on both training and validation sets by calculating the receiver operating characteristic (ROC) curves.

RESULTS

A total of 91 patients was enrolled: 58 were used as training, 33 as validation. Overall, pCR was observed in 25 cases. The model showing the highest performance was obtained combining ERITCP at BED = 26 Gy with a radiomic feature (10th percentile of grey level histogram, 10GLH) calculated at BED = 40 Gy. The area under ROC curve (AUC) of this combined model was 0.98 for training set and 0.92 for validation set, significantly higher (p = 0.04) than the AUC value obtained using ERITCP alone (0.94 in training and 0.89 in validation set).

CONCLUSION

The integration of the radiomic analysis with ERITCP improves the pCR prediction in LARC patients, offering more precise predictive models to further personalise 0.35 T MRIgRT treatments of LARC patients.

Impact of bias field correction on 0.35 T pelvic MR images: evaluation on generative adversarial network-based OARs' auto-segmentation and visual grading assessment

Purpose

Magnetic resonance imaging (MRI)-guided radiotherapy enables adaptive treatment plans based on daily anatomical changes and accurate organ visualization. However, the bias field artifact can compromise image quality, affecting diagnostic accuracy and quantitative analyses. This study aims to assess the impact of bias field correction on 0.35 T pelvis MRIs by evaluating clinical anatomy visualization and generative adversarial network (GAN) auto-segmentation performance.

Materials and methods

3D simulation MRIs from 60 prostate cancer patients treated on MR-Linac (0.35 T) were collected and preprocessed with the N4ITK algorithm for bias field correction. A 3D GAN architecture was trained, validated, and tested on 40, 10, and 10 patients, respectively, to auto-segment the organs at risk (OARs) rectum and bladder. The GAN was trained and evaluated either with the original or the bias-corrected MRIs. The Dice similarity coefficient (DSC) and 95th percentile Hausdorff distance (HD95th) were computed for the segmented volumes of each patient. The Wilcoxon signed-rank test assessed the statistical difference of the metrics within OARs, both with and without bias field correction. Five radiation oncologists blindly scored 22 randomly chosen patients in terms of overall image quality and visibility of boundaries (prostate, rectum, bladder, seminal vesicles) of the original and bias-corrected MRIs. Bennett's S score and Fleiss' kappa were used to assess the pairwise interrater agreement and the interrater agreement among all the observers, respectively.

Results

In the test set, the GAN trained and evaluated on original and bias-corrected MRIs showed DSC/HD95th of 0.92/5.63 mm and 0.92/5.91 mm for the bladder and 0.84/10.61 mm and 0.83/9.71 mm for the rectum. No statistical differences in the distribution of the evaluation metrics were found neither for the bladder (DSC: p = 0.07; HD95th: p = 0.35) nor for the rectum (DSC: p = 0.32; HD95th: p = 0.63). From the clinical visual grading assessment, the bias-corrected MRI resulted mostly in either no change or an improvement of the image quality and visualization of the organs' boundaries compared with the original MRI.

Conclusion

The bias field correction did not improve the anatomy visualization from a clinical point of view and the OARs' auto-segmentation outputs generated by the GAN.

Intra-frame motion deterioration effects and deep-learning-based compensation in MR-guided radiotherapy

BACKGROUND

Current commercially available hybrid magnetic resonance linear accelerators (MR-Linac) use 2D+t cine MR imaging to provide intra-fractional motion monitoring. However, given the limited temporal resolution of cine MR imaging, target intra-frame motion deterioration effects, resulting in effective time latency and motion artifacts in the image domain, can be appreciable, especially in the case of fast breathing.

PURPOSE

The aim of this work is to investigate intra-frame motion deterioration effects in MR-guided radiotherapy (MRgRT) by simulating the motion-corrupted image acquisition, and to explore the feasibility of deep-learning-based compensation approaches, relying on the intra-frame motion information which is spatially and temporally encoded in the raw data (k-space).

METHODS

An intra-frame motion model was defined to simulate motion-corrupted MR images, with 4D anthropomorphic digital phantoms being exploited to provide ground truth 2D+t cine MR sequences. A total number of 10 digital phantoms were generated for lung cancer patients, with randomly selected eight patients for training or validation and the remaining two for testing. The simulation code served as the data generator, and a dedicated motion pattern perturbation scheme was proposed to build the intra-frame motion database, where three degrees of freedom were designed to guarantee the diversity of intra-frame motion trajectories, enabling a thorough exploration in the domain of the potential anatomical structure positions. U-Nets with three types of loss functions: L1 or L2 loss defined in image or Fourier domain, referred to as NNImgLoss-L1 , NNFloss-L1 and NNL2-Loss were trained to extract information from the motion-corrupted image and used to estimate the ground truth final-position image, corresponding to the end of the acquisition. Images before and after compensation were evaluated in terms of (i) image mean-squared error (MSE) and mean absolute error (MAE), and (ii) accuracy of gross tumor volume (GTV) contouring, based on optical-flow image registration.

RESULTS

Image degradation caused by intra-frame motion was observed: for a linearly and fully acquired Cartesian readout k-space trajectory, intra-frame motion resulted in an imaging latency of approximately 50% of the acquisition time; in comparison, the motion artifacts exhibited only a negligible contribution to the overall geometric errors. All three compensation models led to a decrease in image MSE/MAE and GTV position offset compared to the motion-corrupted image. In the investigated testing dataset for GTV contouring, the average dice similarity coefficients (DSC) improved from 88% to 96%, and the 95th percentile Hausdorff distance (HD95 ) dropped from 4.8 mm to 2.1 mm. Different models showed slight performance variations across different intra-frame motion amplitude categories: NNImgLoss-L1 excelled for small/medium amplitudes, whereas NNFloss-L1 demonstrated higher DSC median values at larger amplitudes. The saliency maps of the motion-corrupted image highlighted the major contribution of the later acquired k-space data, as well as the edges of the moving anatomical structures at their final positions, during the model inference stage.

CONCLUSIONS

Our results demonstrate the deep-learning-based approaches have the potential to compensate for intra-frame motion by utilizing the later acquired data to drive the convergence of the earlier acquired k-space components.

Auto-segmentation of pelvic organs at risk on 0.35T MRI using 2D and 3D Generative Adversarial Network models

PURPOSE

Manual recontouring of targets and Organs At Risk (OARs) is a time-consuming and operator-dependent task. We explored the potential of Generative Adversarial Networks (GAN) to auto-segment the rectum, bladder and femoral heads on 0.35T MRIs to accelerate the online MRI-guided-Radiotherapy (MRIgRT) workflow.

METHODS

3D planning MRIs from 60 prostate cancer patients treated with 0.35T MR-Linac were collected. A 3D GAN architecture and its equivalent 2D version were trained, validated and tested on 40, 10 and 10 patients respectively. The volumetric Dice Similarity Coefficient (DSC) and 95th percentile Hausdorff Distance (HD95th) were computed against expert drawn ground-truth delineations. The networks were also validated on an independent external dataset of 16 patients.

RESULTS

In the internal test set, the 3D and 2D GANs showed DSC/HD95th of 0.83/9.72 mm and 0.81/10.65 mm for the rectum, 0.92/5.91 mm and 0.85/15.72 mm for the bladder, and 0.94/3.62 mm and 0.90/9.49 mm for the femoral heads. In the external test set, the performance was 0.74/31.13 mm and 0.72/25.07 mm for the rectum, 0.92/9.46 mm and 0.88/11.28 mm for the bladder, and 0.89/7.00 mm and 0.88/10.06 mm for the femoral heads. The 3D and 2D GANs required on average 1.44 s and 6.59 s respectively to generate the OARs' volumetric segmentation for a single patient.

CONCLUSIONS

The proposed 3D GAN auto-segments pelvic OARs with high accuracy on 0.35T, in both the internal and the external test sets, outperforming its 2D equivalent in both segmentation robustness and volume generation time.

Multicentric characterisation of lateral beam profiles generated by 6FFF beam of three 0.35 T MR-linac systems

BACKGROUND

The physical characterisation of FFF-beam profiles in the presence of a magnetic field requires a new standardization procedure and formulation.

PURPOSE

The aim of this multicentric experience is to propose new normalisation factors to allow for the calculation of standard parameters typical of flattened beams, such as dosimetric field size and penumbra, for a 6 MV FFF beam from an MR-linac.

METHODS

The measurements were carried out on three ViewRay-MRIdiansystems. An equal set of measurements was acquired using the same equipment. Transverse beam profiles were acquired at seven different depthsand for six beam dimensions.The inflection point was estimated as the position of the maximum of a Gaussian fit obtained from the first derivative of the profiles. The position of the minimum and maximum points of the second derivative of the above Gaussian described the fall-off region, and the external peaks of the third derivative were considered as the in-field and out-field points. The profile normalisation was determined by imposing a 55% dose level at the inflection point and the renormalisation factors were calculated.

RESULTS

The position of the inflection point, and the second and third derivatives peaks were analysed,and the renormalisation factors as a function of field size and depth were determined. The values of the unflatness and the slope have been calculated for different depths and field sizes.

CONCLUSION

This study represents the first multi-centric evaluation of the profiles on different low-field MR-Linac systems and theset of renormalisation parameters to analyse the FFF-beam on that system was effectively proposed.

Deep learning application for abdominal organs segmentation on 0.35 T MR-Linac images

Introduction

Linear accelerator (linac) incorporating a magnetic resonance (MR) imaging device providing enhanced soft tissue contrast is particularly suited for abdominal radiation therapy. In particular, accurate segmentation for abdominal tumors and organs at risk (OARs) required for the treatment planning is becoming possible. Currently, this segmentation is performed manually by radiation oncologists. This process is very time consuming and subject to inter and intra operator variabilities. In this work, deep learning based automatic segmentation solutions were investigated for abdominal OARs on 0.35 T MR-images.

Methods

One hundred and twenty one sets of abdominal MR images and their corresponding ground truth segmentations were collected and used for this work. The OARs of interest included the liver, the kidneys, the spinal cord, the stomach and the duodenum. Several UNet based models have been trained in 2D (the Classical UNet, the ResAttention UNet, the EfficientNet UNet, and the nnUNet). The best model was then trained with a 3D strategy in order to investigate possible improvements. Geometrical metrics such as Dice Similarity Coefficient (DSC), Intersection over Union (IoU), Hausdorff Distance (HD) and analysis of the calculated volumes (thanks to Bland-Altman plot) were performed to evaluate the results.

Results

The nnUNet trained in 3D mode achieved the best performance, with DSC scores for the liver, the kidneys, the spinal cord, the stomach, and the duodenum of 0.96 ± 0.01, 0.91 ± 0.02, 0.91 ± 0.01, 0.83 ± 0.10, and 0.69 ± 0.15, respectively. The matching IoU scores were 0.92 ± 0.01, 0.84 ± 0.04, 0.84 ± 0.02, 0.54 ± 0.16 and 0.72 ± 0.13. The corresponding HD scores were 13.0 ± 6.0 mm, 16.0 ± 6.6 mm, 3.3 ± 0.7 mm, 35.0 ± 33.0 mm, and 42.0 ± 24.0 mm. The analysis of the calculated volumes followed the same behavior.

Discussion

Although the segmentation results for the duodenum were not optimal, these findings imply a potential clinical application of the 3D nnUNet model for the segmentation of abdominal OARs for images from 0.35 T MR-Linac.

MRI-Guided Radiation Therapy-An Emerging and Disruptive Process of Care: Healthcare Economic and Policy Considerations

MRI-guided radiation therapy (MRgRT) is an emerging, innovative technology that provides opportunities to transform and improve the current clinical care process in radiation oncology. As with many new technologies in radiation oncology, careful evaluation from a healthcare economic and policy perspective is required for its successful implementation. In this review article, we describe the current evidence surrounding MRgRT, framing it within the context of value within the healthcare system. Additionally, we highlight areas in which MRgRT may disrupt the current process of care, and discuss the evidence thresholds and timeline required for the widespread adoption of this promising technology.

The Future of MR-Guided Radiation Therapy

Magnetic resonance image guided radiation therapy (MRIgRT) is a relatively new technology that has already shown outcomes benefits but that has not yet reached its clinical potential. The improved soft-tissue contrast provided with MR, coupled with the immediacy of image acquisition with respect to the treatment, enables expansion of on-table adaptive protocols, currently at a cost of increased treatment complexity, use of human resources, and longer treatment slot times, which translate to decreased throughput. Many approaches are being investigated to meet these challenges, including the development of artificial intelligence (AI) algorithms to accelerate and automate much of the workflow and improved technology that parallelizes workflow tasks, as well as improvements in image acquisition speed and quality. This article summarizes limitations of current available integrated MRIgRT systems and gives an outlook about scientific developments to further expand the use of MRIgRT.

Real-time motion management in MRI-guided radiotherapy: Current status and AI-enabled prospects

MRI-guided radiotherapy (MRIgRT) is a highly complex treatment modality, allowing adaptation to anatomical changes occurring from one treatment day to the other (inter-fractional), but also to motion occurring during a treatment fraction (intra-fractional). In this vision paper, we describe the different steps of intra-fractional motion management during MRIgRT, from imaging to beam adaptation, and the solutions currently available both clinically and at a research level. Furthermore, considering the latest developments in the literature, a workflow is foreseen in which motion-induced over- and/or under-dosage is compensated in 3D, with minimal impact to the radiotherapy treatment time. Considering the time constraints of real-time adaptation, a particular focus is put on artificial intelligence (AI) solutions as a fast and accurate alternative to conventional algorithms.

The role of artificial intelligence in radiotherapy clinical practice

This review article visits the current state of artificial intelligence (AI) in radiotherapy clinical practice. We will discuss how AI has a place in the modern radiotherapy workflow at the level of automatic segmentation and planning, two applications which have seen real-work implementation. A special emphasis will be placed on the role AI can play in online adaptive radiotherapy, such as performed at MR-linacs, where online plan adaptation is a procedure which could benefit from automation to reduce on-couch time for patients. Pseudo-CT generation and AI for motion tracking will be introduced in the scope of online adaptive radiotherapy as well. We further discuss the use of AI for decision-making and response assessment, for example for personalized prescription and treatment selection, risk stratification for outcomes and toxicities, and AI for quantitative imaging and response assessment. Finally, the challenges of generalizability and ethical aspects will be covered. With this, we provide a comprehensive overview of the current and future applications of AI in radiotherapy.

Translation of AI into oncology clinical practice

Artificial intelligence (AI) is a transformative technology that is capturing popular imagination and can revolutionize biomedicine. AI and machine learning (ML) algorithms have the potential to break through existing barriers in oncology research and practice such as automating workflow processes, personalizing care, and reducing healthcare disparities. Emerging applications of AI/ML in the literature include screening and early detection of cancer, disease diagnosis, response prediction, prognosis, and accelerated drug discovery. Despite this excitement, only few AI/ML models have been properly validated and fewer have become regulated products for routine clinical use. In this review, we highlight the main challenges impeding AI/ML clinical translation. We present different clinical use cases from the domains of radiology, radiation oncology, immunotherapy, and drug discovery in oncology. We dissect the unique challenges and opportunities associated with each of these cases. Finally, we summarize the general requirements for successful AI/ML implementation in the clinic, highlighting specific examples and points of emphasis including the importance of multidisciplinary collaboration of stakeholders, role of domain experts in AI augmentation, transparency of AI/ML models, and the establishment of a comprehensive quality assurance program to mitigate risks of training bias and data drifts, all culminating toward safer and more beneficial AI/ML applications in oncology labs and clinics.

Recent Advances and Adaptive Strategies in Image Guidance for Cervical Cancer Radiotherapy

The standard of care for locally advanced cervical cancer is external beam radiotherapy (EBRT) with simultaneous chemotherapy followed by an internal radiation boost. New imaging methods such as positron-emission tomography and magnetic resonance imaging have been implemented into daily practice for better tumor delineation in radiotherapy planning. The method of delivering radiation has changed with technical advances in qualitative imaging and treatment delivery. Image-guided radiotherapy (IGRT) plays an important role in minimizing treatment toxicity of pelvic radiation and provides a superior conformality for sparing the organs at risk (OARs) such as bone marrow, bowel, rectum, and bladder. Similarly, three-dimensional image-guided adaptive brachytherapy (3D-IGABT) with computed tomography (CT) or magnetic resonance imaging (MRI) has been reported to improve target coverage and reduce the dose to normal tissues. Brachytherapy is a complementary part of radiotherapy treatment for cervical cancer and, over the past 20 years, 3D-image-based brachytherapy has rapidly evolved and established itself as the gold standard. With new techniques and adaptive treatment in cervical cancer, the concept of personalized medicine is introduced with an enhanced comprehension of the therapeutic index not only in terms of volume (three-dimensional) but during treatment too (four-dimensional). Current data show promising results with integrated IGRT and IGABT in clinical practice and, therefore, better local control and overall survival while reducing treatment-related morbidity. This review gives an overview of the substantial impact that occurred in the progress of image-guided adaptive external beam radiotherapy and brachytherapy.

MRI-Guided Radiation Therapy for Prostate Cancer: The Next Frontier in Ultrahypofractionation

Technological advances in MRI-guided radiation therapy (MRIgRT) have improved real-time visualization of the prostate and its surrounding structures over CT-guided radiation therapy. Seminal studies have demonstrated safe dose escalation achieved through ultrahypofractionation with MRIgRT due to planning target volume (PTV) margin reduction and treatment gating. On-table adaptation with MRI-based technologies can also incorporate real-time changes in target shape and volume and can reduce high doses of radiation to sensitive surrounding structures that may move into the treatment field. Ongoing clinical trials seek to refine ultrahypofractionated radiotherapy treatments for prostate cancer using MRIgRT. Though these studies have the potential to demonstrate improved biochemical control and reduced side effects, limitations concerning patient treatment times and operational workflows may preclude wide adoption of this technology outside of centers of excellence. In this review, we discuss the advantages and limitations of MRIgRT for prostate cancer, as well as clinical trials testing the efficacy and toxicity of ultrafractionation in patients with localized or post-prostatectomy recurrent prostate cancer.

Effectiveness of bladder filling control during online MR-guided adaptive radiotherapy for rectal cancer

BACKGROUND

Magnetic resonance-guided adaptive radiotherapy (MRgART) treatment sessions at MR-Linac are time-consuming and changes in organs at risk volumes can impact the treatment dosimetry. This study aims to evaluate the feasibility to control bladder filling during the rectum MRgART online session and its effectiveness on plan dosimetry.

METHODS

A total of 109 online adaptive sessions of 24 rectum cancer patients treated at Unity 1.5 T MR-Linac with a short course radiotherapy (25 Gy, 5 Gy × 5) for whom the adaptive plan was optimized and recalculated online based on the daily magnetic resonance imaging (MRI) were analysed. Patients were fitted with a bladder catheter to control bladder filling; the bladder is emptied and then partially filled with a known amount of saline at the beginning and end of the online session. A first MRI ([Formula: see text]) acquired at the beginning of the session was used for plan adaptation and the second ([Formula: see text]) was acquired while approving the adapted plan and rigidly registered with the first to ensure the appropriateness of the isodoses on the ongoing delivery treatment. For each fraction, the time interval between the two MRIs and potential bladder changes were assessed with independent metrics, and the impact on the plan dosimetry was evaluated by comparing target and organs at risk dose volume histogram cut-off points of the plan adapted on [Formula: see text] and recalculated on [Formula: see text].

RESULTS

Median bladder volume variations, DSC, and HD of 8.17%, 0.922, and 2.92 mm were registered within a median time of 38 min between [Formula: see text] and [Formula: see text]; dosimetric differences < 0.65% were registered for target coverage, and < 0.5% for bladder, small bowel and femoral heads constraints, with a p value > 0.05.

CONCLUSION

The use of a bladder filling control procedure can help ensure the dosimetric accuracy of the online adapted treatment delivered.

Deep learning based automatic segmentation of organs-at-risk for 0.35 T MRgRT of lung tumors

BACKGROUND AND PURPOSE

Magnetic resonance imaging guided radiotherapy (MRgRT) offers treatment plan adaptation to the anatomy of the day. In the current MRgRT workflow, this requires the time consuming and repetitive task of manual delineation of organs-at-risk (OARs), which is also prone to inter- and intra-observer variability. Therefore, deep learning autosegmentation (DLAS) is becoming increasingly attractive. No investigation of its application to OARs in thoracic magnetic resonance images (MRIs) from MRgRT has been done so far. This study aimed to fill this gap.

MATERIALS AND METHODS

122 planning MRIs from patients treated at a 0.35 T MR-Linac were retrospectively collected. Using an 80/19/23 (training/validation/test) split, individual 3D U-Nets for segmentation of the left lung, right lung, heart, aorta, spinal canal and esophagus were trained. These were compared to the clinically used contours based on Dice similarity coefficient (DSC) and Hausdorff distance (HD). They were also graded on their clinical usability by a radiation oncologist.

RESULTS

Median DSC was 0.96, 0.96, 0.94, 0.90, 0.88 and 0.78 for left lung, right lung, heart, aorta, spinal canal and esophagus, respectively. Median 95th percentile values of the HD were 3.9, 5.3, 5.8, 3.0, 2.6 and 3.5 mm, respectively. The physician preferred the network generated contours over the clinical contours, deeming 85 out of 129 to not require any correction, 25 immediately usable for treatment planning, 15 requiring minor and 4 requiring major corrections.

CONCLUSIONS

We trained 3D U-Nets on clinical MRI planning data which produced accurate delineations in the thoracic region. DLAS contours were preferred over the clinical contours.

Deep learning auto-segmentation on multi-sequence magnetic resonance images for upper abdominal organs

Introduction

Multi-sequence multi-parameter MRIs are often used to define targets and/or organs at risk (OAR) in radiation therapy (RT) planning. Deep learning has so far focused on developing auto-segmentation models based on a single MRI sequence. The purpose of this work is to develop a multi-sequence deep learning based auto-segmentation (mS-DLAS) based on multi-sequence abdominal MRIs.

Materials and methods

Using a previously developed 3DResUnet network, a mS-DLAS model using 4 T1 and T2 weighted MRI acquired during routine RT simulation for 71 cases with abdominal tumors was trained and tested. Strategies including data pre-processing, Z-normalization approach, and data augmentation were employed. Additional 2 sequence specific T1 weighted (T1-M) and T2 weighted (T2-M) models were trained to evaluate performance of sequence-specific DLAS. Performance of all models was quantitatively evaluated using 6 surface and volumetric accuracy metrics.

Results

The developed DLAS models were able to generate reasonable contours of 12 upper abdomen organs within 21 seconds for each testing case. The 3D average values of dice similarity coefficient (DSC), mean distance to agreement (MDA mm), 95 percentile Hausdorff distance (HD95% mm), percent volume difference (PVD), surface DSC (sDSC), and relative added path length (rAPL mm/cc) over all organs were 0.87, 1.79, 7.43, -8.95, 0.82, and 12.25, respectively, for mS-DLAS model. Collectively, 71% of the auto-segmented contours by the three models had relatively high quality. Additionally, the obtained mS-DLAS successfully segmented 9 out of 16 MRI sequences that were not used in the model training.

Conclusion

We have developed an MRI-based mS-DLAS model for auto-segmenting of upper abdominal organs on MRI. Multi-sequence segmentation is desirable in routine clinical practice of RT for accurate organ and target delineation, particularly for abdominal tumors. Our work will act as a stepping stone for acquiring fast and accurate segmentation on multi-contrast MRI and make way for MR only guided radiation therapy.

CARdioimaging in Lung Cancer PatiEnts Undergoing Radical RadioTherapy: CARE-RT Trial

BACKGROUND

Non-small-cell lung cancer (NSCLC) is a common, steady growing lung tumour that is often discovered when a surgical approach is forbidden. For locally advanced inoperable NSCLC, the clinical approach consists of a combination of chemotherapy and radiotherapy, eventually followed by adjuvant immunotherapy, a treatment that is useful but may cause several mild and severe adverse effect. Chest radiotherapy, specifically, may affect the heart and coronary artery, impairing heart function and causing pathologic changes in myocardial tissues. The aim of this study is to evaluate the damage coming from these therapies with the aid of cardiac imaging.

METHODS

This is a single-centre, prospective clinical trial. Patients with NSCLC who are enrolled will undergo computed tomography (CT) and magnetic resonance imaging (MRI) before chemotherapy 3 months, 6 months, and 9-12 months after the treatment. We expect to enrol 30 patients in 2 years.

CONCLUSIONS

Our clinical trial will be an opportunity not only to highlight the timing and the radiation dose needed for pathological cardiac tissue changes to happen but will also provide useful data to set new follow-up schedules and strategies, keeping in mind that, more often than not, patients affected by NSCLC may present other heart- and lung-related pathological conditions.

AI in MRI: Computational Frameworks for a Faster, Optimized, and Automated Imaging Workflow

Over the last decade, artificial intelligence (AI) has made an enormous impact on a wide range of fields, including science, engineering, informatics, finance, and transportation [...].

Stereotactic Magnetic Resonance-Guided Adaptive and Non-Adaptive Radiotherapy on Combination MR-Linear Accelerators: Current Practice and Future Directions

Stereotactic body radiotherapy (SBRT) is an effective radiation therapy technique that has allowed for shorter treatment courses, as compared to conventionally dosed radiation therapy. As its name implies, SBRT relies on daily image guidance to ensure that each fraction targets a tumor, instead of healthy tissue. Magnetic resonance imaging (MRI) offers improved soft-tissue visualization, allowing for better tumor and normal tissue delineation. MR-guided RT (MRgRT) has traditionally been defined by the use of offline MRI to aid in defining the RT volumes during the initial planning stages in order to ensure accurate tumor targeting while sparing critical normal tissues. However, the ViewRay MRIdian and Elekta Unity have improved upon and revolutionized the MRgRT by creating a combined MRI and linear accelerator (MRL), allowing MRgRT to incorporate online MRI in RT. MRL-based MR-guided SBRT (MRgSBRT) represents a novel solution to deliver higher doses to larger volumes of gross disease, regardless of the proximity of at-risk organs due to the (1) superior soft-tissue visualization for patient positioning, (2) real-time continuous intrafraction assessment of internal structures, and (3) daily online adaptive replanning. Stereotactic MR-guided adaptive radiation therapy (SMART) has enabled the safe delivery of ablative doses to tumors adjacent to radiosensitive tissues throughout the body. Although it is still a relatively new RT technique, SMART has demonstrated significant opportunities to improve disease control and reduce toxicity. In this review, we included the current clinical applications and the active prospective trials related to SMART. We highlighted the most impactful clinical studies at various tumor sites. In addition, we explored how MRL-based multiparametric MRI could potentially synergize with SMART to significantly change the current treatment paradigm and to improve personalized cancer care.

Patient-specific transfer learning for auto-segmentation in adaptive 0.35 T MRgRT of prostate cancer: a bi-centric evaluation

BACKGROUND

Online adaptive radiation therapy (RT) using hybrid magnetic resonance linear accelerators (MR-Linacs) can administer a tailored radiation dose at each treatment fraction. Daily MR imaging followed by organ and target segmentation adjustments allow to capture anatomical changes, improve target volume coverage, and reduce the risk of side effects. The introduction of automatic segmentation techniques could help to further improve the online adaptive workflow by shortening the re-contouring time and reducing intra- and inter-observer variability. In fractionated RT, prior knowledge, such as planning images and manual expert contours, is usually available before irradiation, but not used by current artificial intelligence-based autocontouring approaches.

PURPOSE

The goal of this study was to train convolutional neural networks (CNNs) for automatic segmentation of bladder, rectum (organs at risk, OARs), and clinical target volume (CTV) for prostate cancer patients treated at 0.35 T MR-Linacs. Furthermore, we tested the CNNs generalization on data from independent facilities and compared them with the MR-Linac treatment planning system (TPS) propagated structures currently used in clinics. Finally, expert planning delineations were utilized for patient- (PS) and facility-specific (FS) transfer learning to improve auto-segmentation of CTV and OARs on fraction images.

METHODS

In this study, data from fractionated treatments at 0.35 T MR-Linacs were leveraged to develop a 3D U-Net-based automatic segmentation. Cohort C1 had 73 planning images and cohort C2 had 19 planning and 240 fraction images. The baseline models (BMs) were trained solely on C1 planning data using 53 MRIs for training and 10 for validation. To assess their accuracy, the models were tested on three data subsets: (i) 10 C1 planning images not used for training, (ii) 19 C2 planning, and (iii) 240 C2 fraction images. BMs also served as a starting point for FS and PS transfer learning, where the planning images from C2 were used for network parameter fine tuning. The segmentation output of the different trained models was compared against expert ground truth by means of geometric metrics. Moreover, a trained physician graded the network segmentations as well as the segmentations propagated by the clinical TPS.

RESULTS

The BMs showed dice similarity coefficients (DSC) of 0.88(4) and 0.93(3) for the rectum and the bladder, respectively, independent of the facility. CTV segmentation with the BM was the best for intermediate- and high-risk cancer patients from C1 with DSC=0.84(5) and worst for C2 with DSC=0.74(7). The PS transfer learning brought a significant improvement in the CTV segmentation, yielding DSC=0.72(4) for post-prostatectomy and low-risk patients and DSC=0.88(5) for intermediate- and high-risk patients. The FS training did not improve the segmentation accuracy considerably. The physician's assessment of the TPS-propagated versus network-generated structures showed a clear advantage of the latter.

CONCLUSIONS

The obtained results showed that the presented segmentation technique has potential to improve automatic segmentation for MR-guided RT.

Evaluation of real-time tumor contour prediction using LSTM networks for MR-guided radiotherapy

BACKGROUND AND PURPOSE

Magnetic resonance imaging guided radiotherapy (MRgRT) with deformable multileaf collimator (MLC) tracking would allow to tackle both rigid displacement and tumor deformation without prolonging treatment. However, the system latency must be accounted for by predicting future tumor contours in real-time. We compared the performance of three artificial intelligence (AI) algorithms based on long short-term memory (LSTM) modules for the prediction of 2D-contours 500ms into the future.

MATERIALS AND METHODS

Models were trained (52 patients, 3.1h of motion), validated (18 patients, 0.6h) and tested (18 patients, 1.1h) with cine MRs from patients treated at one institution. Additionally, we used three patients (2.9h) treated at another institution as second testing set. We implemented 1) a classical LSTM network (LSTM-shift) predicting tumor centroid positions in superior-inferior and anterior-posterior direction which are used to shift the last observed tumor contour. The LSTM-shift model was optimized both in an offline and online fashion. We also implemented 2) a convolutional LSTM model (ConvLSTM) to directly predict future tumor contours and 3) a convolutional LSTM combined with spatial transformer layers (ConvLSTM-STL) to predict displacement fields used to warp the last tumor contour.

RESULTS

The online LSTM-shift model was found to perform slightly better than the offline LSTM-shift and significantly better than the ConvLSTM and ConvLSTM-STL. It achieved a 50% Hausdorff distance of 1.2mm and 1.0mm for the two testing sets, respectively. Larger motion ranges were found to lead to more substantial performance differences across the models.

CONCLUSION

LSTM networks predicting future centroids and shifting the last tumor contour are the most suitable for tumor contour prediction. The obtained accuracy would allow to reduce residual tracking errors during MRgRT with deformable MLC-tracking.

A body mass index-based method for "MR-only" abdominal MR-guided adaptive radiotherapy

PURPOSE

Dose calculation for MR-guided radiotherapy (MRgRT) at the 0.35 T MR-Linac is currently based on deformation of planning CTs (defCT) acquired for each patient. We present a simple and robust bulk density overwrite synthetic CT (sCT) method for abdominal treatments in order to streamline clinical workflows.

METHOD

Fifty-six abdominal patient treatment plans were retrospectively evaluated. All patients had been treated at the MR-Linac using MR datasets for treatment planning and plan adaption and defCT for dose calculation. Bulk density CTs (4M-sCT) were generated from MR images with four material compartments (bone, lung, air, soft tissue). The relative electron densities (RED) for bone and lung were extracted from contoured CT structure average REDs. For soft tissue, a correlation between BMI and RED was evaluated. Dose was recalculated on 4M-sCT and compared to dose distributions on defCTs assessing dose differences in the PTV and organs at risk (OAR).

RESULTS

Mean RED of bone was 1.17 ± 0.02, mean RED of lung 0.17 ± 0.05. The correlation between BMI and RED for soft tissue was statistically significant (p < 0.01). PTV dose differences between 4M-sCT and defCT were Dmean: -0.4 ± 1.0%, D1%: -0.3 ± 1.1% and D95%: -0.5 ± 1.0%. OARs showed D2%: -0.3 ± 1.9% and Dmean: -0.1 ± 1.4% differences. Local 3D gamma index pass rates (2%/2mm) between dose calculated using 4M-sCT and defCT were 96.8 ± 2.6% (range 89.9-99.6%).

CONCLUSION

The presented method for sCT generation enables precise dose calculation for MR-only abdominal MRgRT.

MRI-LINAC: A transformative technology in radiation oncology

Advances in radiotherapy technologies have enabled more precise target guidance, improved treatment verification, and greater control and versatility in radiation delivery. Amongst the recent novel technologies, Magnetic Resonance Imaging (MRI) guided radiotherapy (MRgRT) may hold the greatest potential to improve the therapeutic gains of image-guided delivery of radiation dose. The ability of the MRI linear accelerator (LINAC) to image tumors and organs with on-table MRI, to manage organ motion and dose delivery in real-time, and to adapt the radiotherapy plan on the day of treatment while the patient is on the table are major advances relative to current conventional radiation treatments. These advanced techniques demand efficient coordination and communication between members of the treatment team. MRgRT could fundamentally transform the radiotherapy delivery process within radiation oncology centers through the reorganization of the patient and treatment team workflow process. However, the MRgRT technology currently is limited by accessibility due to the cost of capital investment and the time and personnel allocation needed for each fractional treatment and the unclear clinical benefit compared to conventional radiotherapy platforms. As the technology evolves and becomes more widely available, we present the case that MRgRT has the potential to become a widely utilized treatment platform and transform the radiation oncology treatment process just as earlier disruptive radiation therapy technologies have done.

Synthetic computed tomography generation for abdominal adaptive radiotherapy using low-field magnetic resonance imaging

Background and Purpose

Magnetic Resonance guided Radiotherapy (MRgRT) still needs the acquisition of Computed Tomography (CT) images and co-registration between CT and Magnetic Resonance Imaging (MRI). The generation of synthetic CT (sCT) images from the MR data can overcome this limitation. In this study we aim to propose a Deep Learning (DL) based approach for sCT image generation for abdominal Radiotherapy using low field MR images.

Materials and methods

CT and MR images were collected from 76 patients treated on abdominal sites. U-Net and conditional Generative Adversarial Network (cGAN) architectures were used to generate sCT images. Additionally, sCT images composed of only six bulk densities were generated with the aim of having a Simplified sCT.Radiotherapy plans calculated using the generated images were compared to the original plan in terms of gamma pass rate and Dose Volume Histogram (DVH) parameters.

Results

sCT images were generated in 2 s and 2.5 s with U-Net and cGAN architectures respectively.Gamma pass rates for 2%/2mm and 3%/3mm criteria were 91% and 95% respectively. Dose differences within 1% for DVH parameters on the target volume and organs at risk were obtained.

Conclusion

U-Net and cGAN architectures are able to generate abdominal sCT images fast and accurately from low field MRI.

Image-based motion artifact reduction on liver dynamic contrast enhanced MRI

Liver MRI images often suffer from degraded quality due to ghosting or blurring artifacts caused by patient respiratory or bulk motion. In this study, we developed a two-stage deep learning model to reduce motion artifact on dynamic contrast enhanced (DCE) liver MRIs. The stage-I network utilized a deep residual network with a densely connected multi-resolution block (DRN-DCMB) network to remove most motion artifacts. The stage-II network applied the generative adversarial network (GAN) and perceptual loss compensation to preserve image structural features. The stage-I network served as the generator of GAN and its pretrained parameters in stage-I were further updated via backpropagation during stage-II training. The stage-I network was trained using small image patches with simulated motion artifacts including image-space rotational and translational motion, and K-space based centric and interleaved linear motion, sinusoidal, and rotational motion to mimic liver motion patterns. The stage-II network training used full-size images with the same types of simulated motion. The liver DCE-MRI image volumes without obvious motion artifacts in 10 patients were used for the training process, of which 1020 images of 8 patients were used for training and 240 images of 2 patients for validation. Finally, the whole two-stage deep learning model was tested with simulated motion images (312 clean images from 5 test patients) and patient images with real motion artifacts (28 motion images from 12 patients). The resulted images after two-stage processing demonstrated reduced motion artifacts while preserved anatomic details without image blurriness, with SSIM of 0.935 ± 0.092, MSE of 60.7 ± 9.0 × 10^-3, and PSNR of 32.054 ± 2.219.

Tuning the optimal diffusion-weighted MRI parameters on a 0.35-T MR-Linac for clinical implementation: A phantom study

PURPOSE

This study aims to assess the quality of a new diffusion-weighted imaging (DWI) sequence implemented on an MR-Linac MRIdian system, evaluating and optimizing the acquisition parameters to explore the possibility of clinically implementing a DWI acquisition protocol in a 0.35-T MR-Linac.

MATERIALS AND METHODS

All the performed analyses have been carried out on two types of phantoms: a homogeneous 24-cm diameter polymethylmethacrylate (PMMA) sphere (SP) and a homemade phantom (HMP) constating in a PMMA cylinder filled with distilled water with empty sockets into which five cylindrical vials filled with five different concentrations of methylcellulose water solutions have been inserted. SP was used to evaluate the dependence of diffusion gradient inhomogeneity artifacts on gantry position. Four diffusion sequences with b-values of 500 s/mm2 and 3 averages have been acquired: three with diffusion gradients in the three main directions (phase direction, read direction, slice direction) and one with the diffusion gradients switched off. The dependence of diffusion image uniformity and SNR on the number of averages in the MR sequences was also investigated to determine the optimal number of averages. Finally, the ADC values of HMP have been computed and then compared between images acquired in the scanners at 0.35 and 1.5 T.

RESULTS

In order to acquire high-quality artifact-free DWI images, the "slice" gradient direction has been identified to be the optimal one and 0° to be the best gradient angle. Both the SNR ratio and the uniformity increase with the number of averages. A threshold value of 80 for SNR and 85% for uniformity was adopted to choose the best number of averages. By making a compromise between time and quality and limiting the number of b-values, it is possible to reduce the acquisition time to 78 s. The Passing-Bablok test showed that the two methods, with 0.35 and 1.5 T scanners, led to similar results.

CONCLUSION

The quality of the DWI has been accurately evaluated in relation to different sequence parameters, and optimal parameters have been identified to select a clinical protocol for the acquisition of ADC maps sustainable in the workflow of a hybrid radiotherapy system with a 0.35-T MRI scanner.

A deep learning approach to generate synthetic CT in low field MR-guided radiotherapy for lung cases

INTRODUCTION

This study aims to apply a conditional Generative Adversarial Network (cGAN) to generate synthetic Computed Tomography (sCT) from 0.35 Tesla Magnetic Resonance (MR) images of the thorax.

METHODS

Sixty patients treated for lung lesions were enrolled and divided into training (32), validation (8), internal (10,T_A) and external (10,T_B) test set. Image accuracy of generated sCT was evaluated computing the mean absolute (MAE) and mean error (ME) with respect the original CT. Three treatment plans were calculated for each patient considering MRI as reference image: original CT, sCT (pure sCT) and sCT with GTV density override (hybrid sCT) were used as Electron Density (ED) map. Dose accuracy was evaluated comparing treatment plans in terms of gamma analysis and Dose Volume Histogram (DVH) parameters.

RESULTS

No significant difference was observed between the test sets for image and dose accuracy parameters. Considering the whole test cohort, a MAE of 54.9 ± 10.5 HU and a ME of 4.4 ± 7.4 HU was obtained. Mean gamma passing rates for 2%/2mm, and 3%/3mm tolerance criteria were 95.5 ± 5.9% and 98.2 ± 4.1% for pure sCT, 96.1 ± 5.1% and 98.5 ± 3.9% for hybrid sCT: the difference between the two approaches was significant (p = 0.01). As regards DVH analysis, differences in target parameters estimation were found to be within 5% using hybrid approach and 20% using pure sCT.

CONCLUSION

The DL algorithm here presented can generate sCT images in the thorax with good image and dose accuracy, especially when the hybrid approach is used. The algorithm does not suffer from inter-scanner variability, making feasible the implementation of MR-only workflows for palliative treatments.

Synthetic CT generation for MRI-guided adaptive radiotherapy in prostate cancer

Current MRI-guided adaptive radiotherapy (MRgART) workflows require fraction-specific electron and/or mass density maps, which are created by deformable image registration (DIR) between the simulation CT images and daily MR images. Manual density overrides may also be needed where DIR-produced results are inaccurate. This approach slows the adaptive radiotherapy workflow and introduces additional dosimetric uncertainties, especially in the presence of the magnetic field. This study investigated a method based on a conditional generative adversarial network (cGAN) with a multi-planar method to generate synthetic CT images from low-field MR images to improve efficiency in MRgART workflows for prostate cancer. Fifty-seven male patients, who received MRI-guided radiation therapy to the pelvis using the ViewRay MRIdian Linac, were selected. Forty-five cases were randomly assigned to the training cohort with the remaining twelve cases assigned to the validation/testing cohort. All patient datasets had a semi-paired DIR-deformed CT-sim image and 0.35T MR image acquired using a true fast imaging with steady-state precession (TrueFISP) sequence. Synthetic CT images were compared with deformed CT images to evaluate image quality and dosimetric accuracy. To evaluate the dosimetric accuracy of this method, clinical plans were recalculated on synthetic CT images in the MRIdian treatment planning system. Dose volume histograms for planning target volumes (PTVs) and organs-at-risk (OARs) and dose distributions using gamma analyses were evaluated. The mean-absolute-errors (MAEs) in CT numbers were 30.1 ± 4.2 HU, 19.6 ± 2.3 HU and 158.5 ± 26.0 HU for the whole pelvis, soft tissue, and bone, respectively. The peak signal-to-noise ratio was 35.2 ± 1.7 and the structural index similarity measure was 0.9758 ± 0.0035. The dosimetric difference was on average less than 1% for all PTV and OAR metrics. Plans showed good agreement with gamma pass rates of 99% and 99.9% for 1%/1 mm and 2%/2 mm, respectively. Our study demonstrates the potential of using synthetic CT images created with a multi-planar cGAN method from 0.35T MRI TrueFISP images for the MRgART treatment of prostate radiotherapy. Future work will validate the method in a large cohort of patients and investigate the limitations of the method in the adaptive workflow.

Clinical Validation of a Deep-Learning Segmentation Software in Head and Neck: An Early Analysis in a Developing Radiation Oncology Center

Background: Organs at risk (OARs) delineation is a crucial step of radiotherapy (RT) treatment planning workflow. Time-consuming and inter-observer variability are main issues in manual OAR delineation, mainly in the head and neck (H & N) district. Deep-learning based auto-segmentation is a promising strategy to improve OARs contouring in radiotherapy departments. A comparison of deep-learning-generated auto-contours (AC) with manual contours (MC) was performed by three expert radiation oncologists from a single center. Methods: Planning computed tomography (CT) scans of patients undergoing RT treatments for H&N cancers were considered. CT scans were processed by Limbus Contour auto-segmentation software, a commercial deep-learning auto-segmentation based software to generate AC. H&N protocol was used to perform AC, with the structure set consisting of bilateral brachial plexus, brain, brainstem, bilateral cochlea, pharyngeal constrictors, eye globes, bilateral lens, mandible, optic chiasm, bilateral optic nerves, oral cavity, bilateral parotids, spinal cord, bilateral submandibular glands, lips and thyroid. Manual revision of OARs was performed according to international consensus guidelines. The AC and MC were compared using the Dice similarity coefficient (DSC) and 95% Hausdorff distance transform (DT). Results: A total of 274 contours obtained by processing CT scans were included in the analysis. The highest values of DSC were obtained for the brain (DSC 1.00), left and right eye globes and the mandible (DSC 0.98). The structures with greater MC editing were optic chiasm, optic nerves and cochleae. Conclusions: In this preliminary analysis, deep-learning auto-segmentation seems to provide acceptable H&N OAR delineations. For less accurate organs, AC could be considered a starting point for review and manual adjustment. Our results suggest that AC could become a useful time-saving tool to optimize workload and resources in RT departments.

Machine learning in neuro-oncology: toward novel development fields

PURPOSE

Artificial Intelligence (AI) involves several and different techniques able to elaborate a large amount of data responding to a specific planned outcome. There are several possible applications of this technology in neuro-oncology.

METHODS

We reviewed, according to PRISMA guidelines, available studies adopting AI in different fields of neuro-oncology including neuro-radiology, pathology, surgery, radiation therapy, and systemic treatments.

RESULTS

Neuro-radiology presented the major number of studies assessing AI. However, this technology is being successfully tested also in other operative settings including surgery and radiation therapy. In this context, AI shows to significantly reduce resources and costs maintaining an elevated qualitative standard. Pathological diagnosis and development of novel systemic treatments are other two fields in which AI showed promising preliminary data.

CONCLUSION

It is likely that AI will be quickly included in some aspects of daily clinical practice. Possible applications of these techniques are impressive and cover all aspects of neuro-oncology.

Effect of intrafraction adaptation on PTV margins for MRI guided online adaptive radiotherapy for rectal cancer

Purpose

To determine PTV margins for intrafraction motion in MRI-guided online adaptive radiotherapy for rectal cancer and the potential benefit of performing a 2nd adaptation prior to irradiation.

Methods

Thirty patients with rectal cancer received radiotherapy on a 1.5 T MR-Linac. On T2-weighted images for adaptation (MRI_adapt), verification prior to (MRI_ver) and after irradiation (MRI_post) of 5 treatment fractions per patient, the primary tumor GTV (GTV_prim) and mesorectum CTV (CTV_meso) were delineated. The structures on MRI_adapt were expanded to corresponding PTVs. We determined the required expansion margins such that on average over 5 fractions, 98% of CTV_meso and 95% of GTV_prim on MRI_post was covered in 90% of the patients. Furthermore, we studied the benefit of an additional adaptation, just prior to irradiation, by evaluating the coverage between the structures on MRI_ver and MRI_post. A threshold to assess the need for a secondary adaptation was determined by considering the overlap between MRI_adapt and MRI_ver.

Results

PTV margins for intrafraction motion without 2nd adaptation were 6.4 mm in the anterior direction and 4.0 mm in all other directions for CTV_meso and 5.0 mm isotropically for GTV_prim. A 2nd adaptation, applied for all fractions where the motion between MRI_adapt and MRI_ver exceeded 1 mm (36% of the fractions) would result in a reduction of the PTV_meso margin to 3.2 mm/2.0 mm. For PTV_prim a margin reduction to 3.5 mm is feasible when a 2nd adaptation is performed in fractions where the motion exceeded 4 mm (17% of the fractions).

Conclusion

We studied the potential benefit of intrafraction motion monitoring and a 2nd adaptation to reduce PTV margins in online adaptive MRIgRT in rectal cancer. Performing 2nd adaptations immediately after online replanning when motion exceeded 1 mm and 4 mm for CTV_meso and GTV_prim respectively, could result in a 30–50% margin reduction with limited reduction of dose to the bowel.

Personalized auto-segmentation for magnetic resonance imaging guided adaptive radiotherapy of prostate cancer

PURPOSE

Fast and accurate delineation of organs on treatment-fraction images is critical in magnetic resonance imaging-guided adaptive radiotherapy (MRIgART). This study proposes a personalized auto-segmentation (AS) framework to assist online delineation of prostate cancer using MRIgART.

METHODS

Image data from 26 patients diagnosed with prostate cancer and treated using hypofractionated MRIgART (5 fractions per patient) were collected retrospectively. Daily pretreatment T2-weighted MRI was performed using a 1.5-T MRI system integrated into a Unity MR-linac. First-fraction image and contour data from 16 patients (80 image-sets) were used to train the population AS model, and the remaining 10 patients composed the test set. The proposed personalized AS framework contained two main steps. First, a convolutional neural network was employed to train the population model using the training set. Second, for each test patient, the population model was progressively fine-tuned with manually checked delineations of the patient's current and previous fractions to obtain a personalized model that was applied to the next fraction.

RESULTS

Compared with the population model, the personalized models substantially improved the mean Dice similarity coefficient from 0.79 to 0.93 for the prostate clinical target volume (CTV), 0.91 to 0.97 for the bladder, 0.82 to 0.92 for the rectum, 0.91 to 0.93 for the femoral heads, respectively.

CONCLUSIONS

The proposed method can achieve accurate segmentation and potentially shorten the overall online delineation time of MRIgART. This article is protected by copyright. All rights reserved.

Offline and online LSTM networks for respiratory motion prediction in MR-guided radiotherapy

Objective. Gated beam delivery is the current clinical practice for respiratory motion compensation in MR-guided radiotherapy, and further research is ongoing to implement tracking. To manage intra-fractional motion using multileaf collimator tracking the total system latency needs to be accounted for in real-time. In this study, long short-term memory (LSTM) networks were optimized for the prediction of superior–inferior tumor centroid positions extracted from clinically acquired 2D cine MRIs. Approach. We used 88 patients treated at the University Hospital of the LMU Munich for training and validation (70 patients, 13.1 h), and for testing (18 patients, 3.0 h). Three patients treated at Fondazione Policlinico Universitario Agostino Gemelli were used as a second testing set (1.5 h). The performance of the LSTMs in terms of root mean square error (RMSE) was compared to baseline linear regression (LR) models for forecasted time spans of 250 ms, 500 ms and 750 ms. Both the LSTM and the LR were trained with offline (offline LSTM and offline LR) and online schemes (offline+online LSTM and online LR), the latter to allow for continuous adaptation to recent respiratory patterns. Main results. We found the offline+online LSTM to perform best for all investigated forecasts. Specifically, when predicting 500 ms ahead it achieved a mean RMSE of 1.20 mm and 1.00 mm, while the best performing LR model achieved a mean RMSE of 1.42 mm and 1.22 mm for the LMU and Gemelli testing set, respectively. Significance. This indicates that LSTM networks have potential as respiratory motion predictors and that continuous online re-optimization can enhance their performance.

Dosimetric Impact of Inter-Fraction Variability in the Treatment of Breast Cancer: Towards New Criteria to Evaluate the Appropriateness of Online Adaptive Radiotherapy

Purpose

As a discipline in its infancy, online adaptive RT (ART) needs new ontologies and ad hoc criteria to evaluate the appropriateness of its use in clinical practice. In this experience, we propose a predictive model able to quantify the dosimetric impact due to daily inter-fraction variability in a standard RT breast treatment, to identify in advance the treatment fractions where patients might benefit from an online ART approach.

Methods

The study was focused on right breast cancer patients treated using standard adjuvant RT on an artificial intelligence (AI)-based linear accelerator. Patients were treated with daily CBCT images and without online adaptation, prescribing 40.05 Gy in 15 fractions, with four IMRT tangential beams. ESTRO guidelines were followed for the delineation on planning CT (pCT) of organs at risk and targets. For each patient, all the CBCT images were rigidly aligned to pCT: CTV and PTV were manually re-contoured and the original treatment plan was recalculated. Various radiological parameters were measured on CBCT images, to quantify inter-fraction variability present in each RT fraction after the couch shifts compensation. The variation of these parameters was correlated with the variation of V95% of PTV (ΔV95%) using the Wilcoxon Mann–Whitney test. Fractions where ΔV95% > 2% were considered as adverse events. A logistic regression model was calculated considering the most significant parameter, and its performance was quantified with a receiver operating characteristic (ROC) curve.

Results

A total of 75 fractions on 5 patients were analyzed. The body variation between daily CBCT and pCT along the beam axis with the highest MU was identified as the best predictor (p = 0.002). The predictive model showed an area under ROC curve of 0.86 (95% CI, 0.82–0.99) with a sensitivity of 85.7% and a specificity of 83.8% at the best threshold, which was equal to 3 mm.

Conclusion

A novel strategy to identify treatment fractions that may benefit online ART was proposed. After image alignment, the measure of body difference between daily CBCT and pCT can be considered as an indirect estimator of V95% PTV variation: a difference larger than 3 mm will result in a V95% decrease larger than 2%. A larger number of observations is needed to confirm the results of this hypothesis-generating study.

Robustness and reproducibility of radiomics in T2 weighted images from magnetic resonance image guided linear accelerator in a phantom study

PURPOSE

Quantitative radiomics features extracted from medical images have been shown to provide value in predicting clinical outcomes. The study for robustness and reproducibility of radiomics features obtained with magnetic resonance image guided linear accelerator (MR-Linac) is insufficient. The objective of this work was to investigate the stability of radiomics features extracted from T2-weighted images of MR-Linac for five common effect factors.

MATERIALS AND METHOD

In this work, ten jellies, five fruits/vegetables, and a dynamic phantom were used to evaluate the impact of test-retest, intraobserver, varied thicknesses, radiation, and motion. These phantoms were scanned on a 1.5 T MRI system of MR-Linac. For test-retest data, the phantoms were scanned twice with repositioning within 15 min. To assess for intraobserver comparison, the segmentation of MR images was repeated by one observer in a double-blind manner. Three slice thicknesses (1.2 mm, 2.4 mm, and 4.8 mm) were used to select robust features that were insensitive to different thicknesses. The effect of radiation on features was studied by acquiring images when the beam was on. Common movement images of patients during radiotherapy were simulated by a dynamic phantom with five motion states to study the motion effect. A total of 1409 radiomics features, including shape features, first-order features, and texture features, were extracted from the original, wavelet, square, logarithmic, exponential and gradient images. The robustness and reproducibility features were evaluated using the concordance correlation coefficient (CCC).

RESULT

The intraobserver group had the most robust features (936/1079, 86.7%), while the group of motion effects had the lowest robustness (56/936, 6.0%), followed by the group of different thickness cohorts (374/936, 40.0%). The stability of features in the test-retest and radiation groups was 1072 of 1312 (81.7%) and 810 of 936 (86.5%), respectively. Overall, 25 of 1409 (2.4%) radiomics features remained robust in all five tests, mostly focusing on the image type of the wavelet. The number of stable features extracted from when the beam was on was less than that extracted when the beam was off. Shape features were the most robust of all of the features in all of the groups, excluding the motion group.

CONCLUSION

Compared with other factors fewer features remained robust to the effect of motion. This result emphasizes the need to consider the effect of respiration motion. The study for T2-weighted images from MR-Linac under different conditions will help us to build a robust predictive model applicable for radiotherapy.

Toward MR-only proton therapy planning for pediatric brain tumors: synthesis of relative proton stopping power images with multiple sequence MRI and development of an online quality assurance tool

PURPOSE

To generate synthetic relative proton-stopping-power (sRPSP) images from MRI sequence(s) and develop an online quality assurance (QA) tool for sRPSP to facilitate safe integration of MR-only proton planning into clinical practice.

MATERIALS AND METHODS

Planning CT and MR images of 195 pediatric brain tumor patients were utilized (training: 150, testing: 45). Seventeen consistent-cycle Generative Adversarial Network (ccGAN) models were trained separately using paired CT-converted RPSP and MRI datasets to transform a subject's MRI into sRPSP. T1-weighted (T1W), T2-weighted (T2W), and FLAIR MRI were permutated to form 17 combinations, with or without preprocessing, for determining the optimal training sequence(s). For evaluation, sRPSP images were converted to synthetic CT (sCT) and compared to the real CT in terms of mean absolute error (MAE) in HU. For QA, sCT was deformed and compared to a reference template built from training dataset to produce a flag map, highlighting pixels that deviate by >100 HU and fall outside the mean ± standard deviation reference intensity. The gamma intensity analysis (10%/3mm) of the deformed sCT against the QA template on the intensity difference was investigated as a surrogate of sCT accuracy.

RESULTS

The sRPSP images generated from a single T1W or T2W sequence outperformed that generated from multi-MRI sequences in terms of MAE (all P<0.05). Preprocessing with N4 bias and histogram matching reduced MAE of T2W MRI-based sCT (54±21 HU vs. 42±13 HU, P = .002). The gamma intensity analysis of sCT against the QA template was highly correlated with the MAE of sCT against the real CT in the testing cohort (r = -0.89 for T1W sCT; r = -0.93 for T2W sCT).

CONCLUSION

Accurate sRPSP images can be generated from T1W/T2W MRI for proton planning. A QA tool highlights regions of inaccuracy, flagging problematic cases unsuitable for clinical use. This article is protected by copyright. All rights reserved.

THUNDER 2: THeragnostic Utilities for Neoplastic DisEases of the Rectum by MRI guided radiotherapy

Background

Neoadjuvant chemoradiation therapy (nCRT) is the standard treatment modality in locally advanced rectal cancer (LARC). Since response to radiotherapy (RT) is dose dependent in rectal cancer, dose escalation may lead to higher complete response rates. The possibility to predict patients who will achieve complete response (CR) is fundamental. Recently, an early tumour regression index (ERI) was introduced to predict pathological CR (pCR) after nCRT in LARC patients.

The primary endpoints will be the increase of CR rate and the evaluation of feasibility of delta radiomics-based predictive MRI guided Radiotherapy (MRgRT) model.

Methods

Patients affected by LARC cT2-3, N0-2 or cT4 for anal sphincter involvement N0-2a, M0 without high risk features will be enrolled in the trial. Neoadjuvant CRT will be administered using MRgRT. The initial RT treatment will consist in delivering 55 Gy in 25 fractions on Gross Tumor Volume (GTV) plus the corresponding mesorectum and 45 Gy in 25 fractions on the drainage nodes. Chemotherapy with 5-fluoracil (5-FU) or oral capecitabine will be administered continuously.

A 0.35 Tesla MRI will be acquired at simulation and every day during MRgRT. At fraction 10, ERI will be calculated: if ERI will be inferior than 13.1, the patient will continue the original treatment; if ERI will be higher than 13.1 the treatment plan will be reoptimized, intensifying the dose to the residual tumor at the 11^th fraction to reach 60.1 Gy.

At the end of nCRT instrumental examinations are to be performed in order to restage patients. In case of stable disease or progression, the patient will undergo surgery. In case of major or complete clinical response, conservative approaches may be chosen. Patients will be followed up to evaluate toxicity and quality of life.

The number of cases to be enrolled will be 63: all the patients will be treated at Fondazione Policlinico Universitario A. Gemelli IRCCS in Rome.

Discussion

This clinical trial investigates the impact of RT dose escalation in poor responder LARC patients identified using ERI, with the aim of increasing the probability of CR and consequently an organ preservation benefit in this group of patients.

Trial registration

ClinicalTrials.gov Identifier: NCT04815694 (25/03/2021).

Locally tuned deformation fields combination for 2D cine-MRI-based driving of 3D motion models

PURPOSE

To target mobile tumors in radiotherapy with the recent MR-Linac hardware solutions, research is being conducted to drive a 3D motion model with 2D cine-MRI to reproduce the breathing motion in 4D. This work presents a method to combine several deformation fields using local measures to better drive 3D motion models.

METHODS

The method uses weight maps, each representing the proximity with a specific area of interest. The breathing state is evaluated on cine-MRI frames in these areas and a different deformation field is estimated for each using a 2D to 3D motion model. The different deformation fields are multiplied by their respective weight maps and combined to form the final field to apply to a reference image. A global motion model is adjusted locally on the selected areas and creates a 3DCT for each cine-MRI frame.

RESULTS

The 13 patients on which it was tested showed on average an improvement of the accuracy of our model of 0.71 mm for areas selected to drive the model and 0.5 mm for other areas compared to our previous method without local adjustment. The additional computation time for each region was around 40 ms on a modern laptop.

CONCLUSION

The method improves the accuracy of the2D-based driving of 3D motion models. It can be used on top of existing methods relying on deformation fields. It does add some computation time but, depending on the area to deform and the number of regions of interests, offers the potential of online use.

Enhancing the impact of Artificial Intelligence in Medicine: A joint AIFM-INFN Italian initiative for a dedicated cloud-based computing infrastructure

Artificial Intelligence (AI) techniques have been implemented in the field of Medical Imaging for more than forty years. Medical Physicists, Clinicians and Computer Scientists have been collaborating since the beginning to realize software solutions to enhance the informative content of medical images, including AI-based support systems for image interpretation. Despite the recent massive progress in this field due to the current emphasis on Radiomics, Machine Learning and Deep Learning, there are still some barriers to overcome before these tools are fully integrated into the clinical workflows to finally enable a precision medicine approach to patients' care. Nowadays, as Medical Imaging has entered the Big Data era, innovative solutions to efficiently deal with huge amounts of data and to exploit large and distributed computing resources are urgently needed. In the framework of a collaboration agreement between the Italian Association of Medical Physicists (AIFM) and the National Institute for Nuclear Physics (INFN), we propose a model of an intensive computing infrastructure, especially suited for training AI models, equipped with secure storage systems, compliant with data protection regulation, which will accelerate the development and extensive validation of AI-based solutions in the Medical Imaging field of research. This solution can be developed and made operational by Physicists and Computer Scientists working on complementary fields of research in Physics, such as High Energy Physics and Medical Physics, who have all the necessary skills to tailor the AI-technology to the needs of the Medical Imaging community and to shorten the pathway towards the clinical applicability of AI-based decision support systems.

When your MR linac is down: Can an automated pipeline bail you out of trouble?

PURPOSE

The unique treatment delivery technique provided by magnetic resonance guided radiotherapy (MRgRT) can represent a significant drawback when system fail occurs. This retrospective study proposes and evaluates a pipeline to completely automate the workflow necessary to shift a MRgRT treatment to a traditional radiotherapy linac.

MATERIAL AND METHODS

Patients undergoing treatment during the last MRgRT system failure were retrospectively included in this study. The core of the proposed pipeline was based on a tool able to mimic the original MR linac dose distribution. The so obtained dose distribution (AUTO) has been compared with the distribution obtained in the conventional radiotherapy linac (MAN). Plan comparison has been performed in terms of time required to obtain the final dose distribution, DVH parameters, dosimetric indices and visual analogue scales scoring by radiation oncologists.

RESULTS

AUTO plans generation has been obtained within 10 min for all the considered cases. All AUTO plans were found to be within clinical tolerance, showing a mean target coverage variation of 1.7% with a maximum value of 4.3% and a minimum of 0.6% when compared with MAN plans. The highest OARs mean variation has been found for rectum V60 (6.7%). Dosimetric indices showed no relevant differences, with smaller gradient measure in favour of AUTO plans. Visual analogue scales scoring has confirmed comparable plan quality for AUTO plans.

CONCLUSION

The proposed workflow allows a fast and accurate generation of automatic treatment plans. AUTO plans can be considered equivalent to MAN ones, with limited clinical impact in the worst-case scenario.

Comparison of different deep learning architectures for synthetic CT generation from MR images

PURPOSE

Among the different available methods for synthetic CT generation from MR images for the task of MR-guided radiation planning, the deep learning algorithms have and do outperform their conventional counterparts. In this study, we investigated the performance of some most popular deep learning architectures including eCNN, U-Net, GAN, V-Net, and Res-Net for the task of sCT generation. As a baseline, an atlas-based method is implemented to which the results of the deep learning-based model are compared.

METHODS

A dataset consisting of 20 co-registered MR-CT pairs of the male pelvis is applied to assess the different sCT production methods' performance. The mean error (ME), mean absolute error (MAE), Pearson correlation coefficient (PCC), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR) metrics were computed between the estimated sCT and the ground truth (reference) CT images.

RESULTS

The visual inspection revealed that the sCTs produced by eCNN, V-Net, and ResNet, unlike the other methods, were less noisy and greatly resemble the ground truth CT image. In the whole pelvis region, the eCNN yielded the lowest MAE (26.03 ± 8.85 HU) and ME (0.82 ± 7.06 HU), and the highest PCC metrics were yielded by the eCNN (0.93 ± 0.05) and ResNet (0.91 ± 0.02) methods. The ResNet model had the highest PSNR of 29.38 ± 1.75 among all models. In terms of the Dice similarity coefficient, the eCNN method revealed superior performance in major tissue identification (air, bone, and soft tissue).

CONCLUSIONS

All in all, the eCNN and ResNet deep learning methods revealed acceptable performance with clinically tolerable quantification errors.

New developments in MRI: System characterization, technical advances and radiotherapy applications

A Special Issue of Physica Medica - European Journal of Medical Physics, focused on some important points of contact between the world of magnetic resonance and that of medical physics, was published during 2021. This Editorial describes and comments on the content of this Focus Issue, which contains articles from leading groups invited by the Guest Editors.

A review of artificial intelligence applications for motion tracking in radiotherapy

During radiotherapy, the organs and tumour move as a result of the dynamic nature of the body; this is known as intrafraction motion. Intrafraction motion can result in tumour underdose and healthy tissue overdose, thereby reducing the effectiveness of the treatment while increasing toxicity to the patients. There is a growing appreciation of intrafraction target motion management by the radiation oncology community. Real-time image-guided radiation therapy (IGRT) can track the target and account for the motion, improving the radiation dose to the tumour and reducing the dose to healthy tissue. Recently, artificial intelligence (AI)-based approaches have been applied to motion management and have shown great potential. In this review, four main categories of motion management using AI are summarised: marker-based tracking, markerless tracking, full anatomy monitoring and motion prediction. Marker-based and markerless tracking approaches focus on tracking the individual target throughout the treatment. Full anatomy algorithms monitor for intrafraction changes in the full anatomy within the field of view. Motion prediction algorithms can be used to account for the latencies due to the time for the system to localise, process and act.