New developments in MRI for target volume delineation in radiotherapy

Improvement of 2D cine image quality using 3D priors and cycle generative adversarial network for low field MRI-guided radiation therapy

BACKGROUND

Cine magnetic resonance (MR) images have been used for real-time MR guided radiation therapy (MRgRT). However, the onboard MR systems with low-field strength face the problem of limited image quality.

PURPOSE

To improve the quality of cine MR images in MRgRT using prior image information provided by the patient planning and positioning MR images.

METHODS

This study employed MR images from 18 pancreatic cancer patients who received MR-guided stereotactic body radiation therapy. Planning 3D MR images were acquired during the patient simulation, and positioning 3D MR images and 2D sagittal cine MR images were acquired before and during the beam delivery, respectively. A deep learning-based framework consisting of two cycle generative adversarial networks (CycleGAN), Denoising CycleGAN and Enhancement CycleGAN, was developed to establish the mapping between the 3D and 2D MR images. The Denoising CycleGAN was trained to first denoise the cine images using the time domain cine image series, and the Enhancement CycleGAN was trained to enhance the spatial resolution and contrast by taking advantage of the prior image information from the planning and positioning images. The denoising performance was assessed by signal-to-noise ratio (SNR), structural similarity index measure, peak SNR, blind/reference-less image spatial quality evaluator (BRISQUE), natural image quality evaluator, and perception-based image quality evaluator scores. The quality enhancement performance was assessed by the BRISQUE and physician visual scores. In addition, the target contouring was evaluated on the original and processed images.

RESULTS

Significant differences were found for all evaluation metrics after Denoising CycleGAN processing. The BRISQUE and visual scores were also significantly improved after sequential Denoising and Enhancement CycleGAN processing. In target contouring evaluation, Dice similarity coefficient, centroid distance, Hausdorff distance, and average surface distance values were significantly improved on the enhanced images. The whole processing time was within 20 ms for a typical input image size of 512 × 512.

CONCLUSION

Taking advantage of the prior high-quality positioning and planning MR images, the deep learning-based framework enhanced the cine MR image quality significantly, leading to improved accuracy in automatic target contouring. With the merits of both high computational efficiency and considerable image quality enhancement, the proposed method may hold important clinical implication for real-time MRgRT.

Clinical implementation of a commercial synthetic computed tomography solution for radiotherapy treatment of glioblastoma

Background and Purpose

Magnetic resonance (MR)-only radiotherapy (RT) workflow eliminates uncertainties due to computed tomography (CT)-MR image registration, by using synthetic CT (sCT) images generated from MR. This study describes the clinical implementation process, from retrospective commissioning to prospective validation stage of a commercial artificial intelligence (AI)-based sCT product. Evaluation of the dosimetric performance of the sCT is presented, with emphasis on the impact of voxel size differences between image modalities.

Materials and methods

sCT performance was assessed in glioblastoma RT planning. Dose differences for 30 patients in both commissioning and validation cohorts were calculated at various dose-volume-histogram (DVH) points for target and organs-at-risk (OAR). A gamma analysis was conducted on regridded image plans. Quality assurance (QA) guidelines were established based on commissioning phase results.

Results

Mean dose difference to target structures was found to be within ± 0.7 % regardless of image resolution and cohort. OARs' mean dose differences were within ± 1.3 % for plans calculated on regridded images for both cohorts, while differences were higher for plans with original voxel size, reaching up to -4.2 % for chiasma D2% in the commissioning cohort. Gamma passing rates for the brain structure using the criteria 1 %/1mm, 2 %/2mm and 3 %/3mm were 93.6 %/99.8 %/100 % and 96.6 %/99.9 %/100 % for commissioning and validation cohorts, respectively.

Conclusions

Dosimetric outcomes in both commissioning and validation stages confirmed sCT's equivalence to CT. The large patient cohort in this study aided in establishing a robust QA program for the MR-only workflow, now applied in glioblastoma RT at our center.

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.

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.

Commissioning and implementing a Quality Assurance program for dedicated radiation oncology MRI scanners

PURPOSE

ACR and AAPM task group's guidelines addressing commissioning for dedicated MR simulators were recently published. The goal of the current paper is to present the authors' 2-year experience regarding the commissioning and introduction of a QA program based on these guidelines and an associated automated workflow.

METHODS

All mandatory commissioning tests suggested by AAPM report 284 were performed and results are reported for two MRI scanners (MAGNETOM Sola and Aera). Visual inspection, vendor clinical or service platform, third-party software, or in-house python-based code were used. Automated QA and data analysis was performed via vendor, in-house or third-party software. QATrack+ was used for QA data logging and storage. 3D geometric distortion, B0 inhomogeneity, EPI, and parallel imaging performance were evaluated.

RESULTS

Contrasting with AAPM report 284 recommendations, homogeneity and RF tests were performed monthly. The QA program allowed us to detect major failures over time (shimming, gradient calibration and RF interference). Automated QA, data analysis, and logging allowed fast ACR analysis daily and monthly QA to be performed in 3 h. On the Sola, the average distortion is 1 mm for imaging radii of 250 mm or less. For radii of up to 200 mm, the maximum, average (standard deviation) distortion is 1.2  and 0.4 mm (0.3 mm). Aera values are roughly double the Sola for radii up to 200 mm. EPI geometric distortion, ghosting ratio, and long-term stability were found to be under the maximum recommended values. Parallel imaging SNR ratio was stable and close to the theoretical value (ideal g-factor). No major failures were detected during commissioning.

CONCLUSION

An automated workflow and enhanced QA program allowed to automatically track machine and environmental changes over time and to detect periodic failures and errors that might otherwise have gone unnoticed. The Sola is more geometrically accurate, with a more homogenous B0 field than the Aera.

CT synthesis from MR images using frequency attention conditional generative adversarial network

Magnetic resonance (MR) image-guided radiotherapy is widely used in the treatment planning of malignant tumors, and MR-only radiotherapy, a representative of this technique, requires synthetic computed tomography (sCT) images for effective radiotherapy planning. Convolutional neural networks (CNN) have shown remarkable performance in generating sCT images. However, CNN-based models tend to synthesize more low-frequency components and the pixel-wise loss function usually used to optimize the model can result in blurred images. To address these problems, a frequency attention conditional generative adversarial network (FACGAN) is proposed in this paper. Specifically, a frequency cycle generative model (FCGM) is designed to enhance the inter-mapping between MR and CT and extract more rich tissue structure information. Additionally, a residual frequency channel attention (RFCA) module is proposed and incorporated into the generator to enhance its ability in perceiving the high-frequency image features. Finally, high-frequency loss (HFL) and cycle consistency high-frequency loss (CHFL) are added to the objective function to optimize the model training. The effectiveness of the proposed model is validated on pelvic and brain datasets and compared with state-of-the-art deep learning models. The results show that FACGAN produces higher-quality sCT images while retaining clearer and richer high-frequency texture information.

4D-MRI assisted stereotactic body radiation therapy for unresectable colorectal cancer liver metastases

This study evaluated the feasibilities and outcomes following four-dimensional magnetic resonance imaging (4D-MRI) assisted stereotactic body radiation therapy (SBRT) for unresectable colorectal liver metastases (CRLMs). From March 2018 to January 2022, we identified 76 unresectable CRLMs patients with 123 lesions who received 4D-MRI guided SBRT in our institution. 4D-MRI simulation with or without abdominal compression was conducted for all patients. The prescription dose was 50-65 Gy in 5-12 fractions. The image quality of computed tomography (CT) and MRI were compared using the Clarity Score. Clinical outcomes and toxicity profiles were evaluated. 4D-MRI improved the image quality compared with CT images (mean Clarity Score: 1.67 vs 2.88, P < 0.001). The abdominal compression reduced motions in cranial-caudal direction (P = 0.03) with two phase T2 weighted images assessing tumor motion. The median follow-up time was 12.5 months. For 98 lesions assessed for best response, the complete response, partial response and stable disease rate were 57.1 %, 30.6 % and 12.2 %, respectively. The local control (LC) rate at 1 year was 97.3 %. 46.1 % of patients experienced grade 1-2 toxicities and only 2.6 % patients experienced grade 3 hematologic toxicities. The 4D-MRI technique allowed accurate target delineation and motion tracking in unresectable CRLMs patients. Favorable LC rate and mild toxicities were achieved. This study provided evidence for using 4D-MRI assisted SBRT as an alternative treatment in unresectable CRLMs.

Generation of tissues outside the field of view (FOV) of radiation therapy simulation imaging based on machine learning and patient body outline (PBO)

BACKGROUND

It is not unusual to see some parts of tissues are excluded in the field of view of CT simulation images. A typical mitigation is to avoid beams entering the missing body parts at the cost of sub-optimal planning.

METHODS

This study is to solve the problem by developing 3 methods, (1) deep learning (DL) mechanism for missing tissue generation, (2) using patient body outline (PBO) based on surface imaging, and (3) hybrid method combining DL and PBO. The DL model was built upon a Globally and Locally Consistent Image Completion to learn features by Convolutional Neural Networks-based inpainting, based on Generative Adversarial Network. The database used comprised 10,005 CT training slices of 322 lung cancer patients and 166 CT evaluation test slices of 15 patients. CT images were from the publicly available database of the Cancer Imaging Archive. Since existing data were used PBOs were acquired from the CT images. For evaluation, Structural Similarity Index Metric (SSIM), Root Mean Square Error (RMSE) and Peak signal-to-noise ratio (PSNR) were evaluated. For dosimetric validation, dynamic conformal arc plans were made with the ground truth images and images generated by the proposed method. Gamma analysis was conducted at relatively strict criteria of 1%/1 mm (dose difference/distance to agreement) and 2%/2 mm under three dose thresholds of 1%, 10% and 50% of the maximum dose in the plans made on the ground truth image sets.

RESULTS

The average SSIM in generation part only was 0.06 at epoch 100 but reached 0.86 at epoch 1500. Accordingly, the average SSIM in the whole image also improved from 0.86 to 0.97. At epoch 1500, the average values of RMSE and PSNR in the whole image were 7.4 and 30.9, respectively. Gamma analysis showed excellent agreement with the hybrid method (equal to or higher than 96.6% of the mean of pass rates for all scenarios).

CONCLUSIONS

It was first demonstrated that missing tissues in simulation imaging could be generated with high similarity, and dosimetric limitation could be overcome. The benefit of this study can be significantly enlarged when MR-only simulation is considered.

High-dose rate endorectal brachytherapy for rectal cancer: A state-of-the-art review

Rectal cancer is a common malignancy that requires multidisciplinary treatment. By utilizing the dose-response relationship in rectal cancer radiotherapy, increasing the radiotherapy dose can improve clinical complete remission rates. High-dose rate endorectal brachytherapy (HDREBT) is a novel technique that delivers high doses of radiotherapy directly to the tumor via an endorectal applicator, sparing the adjacent normal tissues from excessive radiation exposure. HDREBT includes contact X-ray brachytherapy and high-dose-rate intracavitary brachytherapy. We introduce the latest developments in applicators and imaging techniques for HDREBT in rectal cancer and summarize the current evidence on the efficacy, safety, and feasibility of HDREBT as a neoadjuvant, definitive, or palliative treatment option for all stages of rectal cancer patients. We also discuss the potential advantages and challenges of HDREBT in achieving organ preservation and improving the quality of life of rectal cancer patients. HDREBT has shown promising results in achieving high complete response rates, enabling nonoperative management, improving organ preservation rates, and providing effective palliation in rectal cancer patients. More studies are needed to optimize its dose and fractionation schemes in different clinical scenarios.

Comprehensive MR imaging QA of 0.35 T MR-Linac using a multi-purpose large FOV phantom: A single-institution experience

PURPOSE

Magnetic resonance-guided radiotherapy (MRgRT) is desired for the treatment of diseases in the abdominothoracic region, which has a broad imaging area and continuous motion. To ensure accurate treatment delivery, an effective image quality assurance (QA) program, with a phantom that covers the field of view (FOV) similar to a human torso, is required. However, routine image QA for a large FOV is not readily available at many MRgRT centers. In this work, we present the clinical experience of the large FOV MRgRT Insight phantom for periodic daily and monthly comprehensive magnetic resonance imaging (MRI)-QA and its feasibility compared to the existing institutional routine MRI-QA procedures in 0.35 T MRgRT.

METHODS

Three phantoms; ViewRay cylindrical water phantom, Fluke 76-907 uniformity and linearity phantom, and Modus QA large FOV MRgRT Insight phantom, were imaged on the 0.35 T MR-Linac. The measurements were made in MRI mode with the true fast imaging with steady-state free precession (TRUFI) sequence. The ViewRay cylindrical water phantom was imaged in a single-position setup whereas the Fluke phantom and Insight phantom were imaged in three different orientations: axial, sagittal, and coronal. Additionally, the phased array coil QA was performed using the horizontal base plate of the Insight phantom by placing the desired coil around the base section which was compared to an in-house built Polyurethane foam phantom for reference.

RESULT

The Insight phantom captured image artifacts across the entire planar field of view, up to 400 mm, in a single image acquisition, which is beyond the FOV of the conventional phantoms. The geometric distortion test showed a similar distortion of 0.45 ± 0.01  and 0.41 ± 0.01 mm near the isocenter, that is, within 300 mm lengths for Fluke and Insight phantoms, respectively, but showed higher geometric distortion of 0.8 ± 0.4 mm in the peripheral region between 300 and 400 mm of the imaging slice for the Insight phantom. The Insight phantom with multiple image quality features and its accompanying software utilized the modulation transform function (MTF) to evaluate the image spatial resolution. The average MTF values were 0.35 ± 0.01, 0.35 ± 0.01, and 0.34 ± 0.03 for axial, coronal, and sagittal images, respectively. The plane alignment and spatial accuracy of the ViewRay water phantom were measured manually. The phased array coil test for both the Insight phantom and the Polyurethane foam phantoms ensured the proper functionality of each coil element.

CONCLUSION

The multifunctional large FOV Insight phantom helps in tracking MR imaging quality of the system to a larger extent compared to the routine daily and monthly QA phantoms currently used in our institute. Also, the Insight phantom is found to be more feasible for routine QA with easy setup.

Deep neural network-based synthetic image digital fluoroscopy using digitally reconstructed tomography

We developed a deep neural network (DNN) to generate X-ray flat panel detector (FPD) images from digitally reconstructed radiographic (DRR) images. FPD and treatment planning CT images were acquired from patients with prostate and head and neck (H&N) malignancies. The DNN parameters were optimized for FPD image synthesis. The synthetic FPD images' features were evaluated to compare to the corresponding ground-truth FPD images using mean absolute error (MAE), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM). The image quality of the synthetic FPD image was also compared with that of the DRR image to understand the performance of our DNN. For the prostate cases, the MAE of the synthetic FPD image was improved (= 0.12 ± 0.02) from that of the input DRR image (= 0.35 ± 0.08). The synthetic FPD image showed higher PSNRs (= 16.81 ± 1.54 dB) than those of the DRR image (= 8.74 ± 1.56 dB), while SSIMs for both images (= 0.69) were almost the same. All metrics for the synthetic FPD images of the H&N cases were improved (MAE 0.08 ± 0.03, PSNR 19.40 ± 2.83 dB, and SSIM 0.80 ± 0.04) compared to those for the DRR image (MAE 0.48 ± 0.11, PSNR 5.74 ± 1.63 dB, and SSIM 0.52 ± 0.09). Our DNN successfully generated FPD images from DRR images. This technique would be useful to increase throughput when images from two different modalities are compared by visual inspection.

On the effect of training database size for MR-based synthetic CT generation in the head

Generation of computed tomography (CT) images from magnetic resonance (MR) images using deep learning methods has recently demonstrated promise in improving MR-guided radiotherapy and PET/MR imaging.

PURPOSE

To investigate the performance of unsupervised training using a large number of unpaired data sets as well as the potential gain in performance after fine-tuning with supervised training using spatially registered data sets in generation of synthetic computed tomography (sCT) from magnetic resonance (MR) images.

MATERIALS AND METHODS

A cycleGAN method consisting of two generators (residual U-Net) and two discriminators (patchGAN) was used for unsupervised training. Unsupervised training utilized unpaired T1-weighted MR and CT images (2061 sets for each modality). Five supervised models were then fine-tuned starting with the generator of the unsupervised model for 1, 10, 25, 50, and 100 pairs of spatially registered MR and CT images. Four supervised training models were also trained from scratch for 10, 25, 50, and 100 pairs of spatially registered MR and CT images using only the residual U-Net generator. All models were evaluated on a holdout test set of spatially registered images from 253 patients, including 30 with significant pathology. sCT images were compared against the acquired CT images using mean absolute error (MAE), Dice coefficient, and structural similarity index (SSIM). sCT images from 60 test subjects generated by the unsupervised, and most accurate of the fine-tuned and supervised models were qualitatively evaluated by a radiologist.

RESULTS

While unsupervised training produced realistic-appearing sCT images, addition of even one set of registered images improved quantitative metrics. Addition of more paired data sets to the training further improved image quality, with the best results obtained using the highest number of paired data sets (n=100). Supervised training was found to be superior to unsupervised training, while fine-tuned training showed no clear benefit over supervised learning, regardless of the training sample size.

CONCLUSION

Supervised learning (using either fine tuning or full supervision) leads to significantly higher quantitative accuracy in the generation of sCT from MR images. However, fine-tuned training using both a large number of unpaired image sets was generally no better than supervised learning using registered image sets alone, suggesting the importance of well registered paired data set for training compared to a large set of unpaired data.

The first patients treated with MR-CBCT soft-tissue matching in a MR-only prostate radiotherapy pathway

INTRODUCTION

Magnetic Resonance (MR)-only radiotherapy for prostate cancer has previously been reported using fiducial markers for on-treatment verification. MR-Cone Beam Computed Tomography (CBCT) soft-tissue matching does not require invasive fiducial markers and enables MR-only treatments to other pelvic cancers. This study evaluated the first clinical implementation of MR-only prostate radiotherapy using MR-CBCT soft-tissue matching.

METHODS

Twenty prostate patients were treated with MR-only radiotherapy using a synthetic (s)CT-optimised plan with MR-CBCT soft-tissue matching. Two MR sequences were acquired: small Field Of View (FOV) for target delineation and large FOV for organs at risk delineation, sCT generation and on-treatment verification. Patients also received a CT for validation. The prostate was independently contoured on the small FOV MR, copied to the registered CT and modified if there were MR-CT soft-tissue alignment differences (MR-CT volume). This was compared to the MR-only volume with a paired t-test. The treatment plan was recalculated on CT and the doses compared. Independent offline CT-CBCT matches for 5/20 fractions were performed by three therapeutic radiographers using the MR-only contours and compared to the online MR-CBCT matches using two one-sided paired t-tests for equivalence within ±1 mm.

RESULTS

The MR-only volumes were significantly smaller than MR-CT (p = 0.003), with a volume ratio 0.92 ± 0.02 (mean ± standard error). The sCT isocentre dose difference to CT was 0.2 ± 0.1%. MR-CBCT soft-tissue matching was equivalent to CT-CBCT (p < 0.001), with differences of 0.1 ± 0.2 mm (vertical), -0.1 ± 0.2 mm (longitudinal) and 0.0 ± 0.1 mm (lateral).

CONCLUSIONS

MR-only radiotherapy with soft-tissue matching has been successfully clinically implemented. It produced significantly smaller target volumes with high dosimetric and on-treatment matching accuracy.

IMPLICATIONS FOR PRACTICE

MR-only prostate radiotherapy can be safely delivered without using invasive fiducial markers. This enables MR-only radiotherapy to be extended to other pelvic cancers where fiducial markers cannot be used.

Compensation cycle consistent generative adversarial networks (Comp-GAN) for synthetic CT generation from MR scans with truncated anatomy

BACKGROUND

MR scans used in radiotherapy can be partially truncated due to the limited field of view (FOV), affecting dose calculation accuracy in MR-based radiation treatment planning.

PURPOSE

We proposed a novel Compensation-cycleGAN (Comp-cycleGAN) by modifying the cycle-consistent generative adversarial network (cycleGAN), to simultaneously create synthetic CT (sCT) images and compensate the missing anatomy from the truncated MR images.

METHODS

Computed tomography (CT) and T1 MR images with complete anatomy of 79 head-and-neck patients were used for this study. The original MR images were manually cropped 10-25 mm off at the posterior head to simulate clinically truncated MR images. Fifteen patients were randomly chosen for testing and the rest of the patients were used for model training and validation. Both the truncated and original MR images were used in the Comp-cycleGAN training stage, which enables the model to compensate for the missing anatomy by learning the relationship between the truncation and known structures. After the model was trained, sCT images with complete anatomy can be generated by feeding only the truncated MR images into the model. In addition, the external body contours acquired from the CT images with full anatomy could be an optional input for the proposed method to leverage the additional information of the actual body shape for each test patient. The mean absolute error (MAE) of Hounsfield units (HU), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM) were calculated between sCT and real CT images to quantify the overall sCT performance. To further evaluate the shape accuracy, we generated the external body contours for sCT and original MR images with full anatomy. The Dice similarity coefficient (DSC) and mean surface distance (MSD) were calculated between the body contours of sCT and original MR images for the truncation region to assess the anatomy compensation accuracy.

RESULTS

The average MAE, PSNR, and SSIM calculated over test patients were 93.1 HU/91.3 HU, 26.5 dB/27.4 dB, and 0.94/0.94 for the proposed Comp-cycleGAN models trained without/with body-contour information, respectively. These results were comparable with those obtained from the cycleGAN model which is trained and tested on full-anatomy MR images, indicating the high quality of the sCT generated from truncated MR images by the proposed method. Within the truncated region, the mean DSC and MSD were 0.85/0.89 and 1.3/0.7 mm for the proposed Comp-cycleGAN models trained without/with body contour information, demonstrating good performance in compensating the truncated anatomy.

CONCLUSIONS

We developed a novel Comp-cycleGAN model that can effectively create sCT with complete anatomy compensation from truncated MR images, which could potentially benefit the MRI-based treatment planning.

An Analysis of the Effects of In Vitro Photodynamic Therapy on Prostate Cancer Tissue by Histopathological Examination and Magnetic Resonance Imaging

Prostate cancer can significantly shorten the lifetime of a patient, even if he is diagnosed at an early stage. The development of minimally-invasive focal therapies such as photodynamic therapy to reduce the number of neoplastic cells while sparing delicate structures is extremely advantageous for treating prostate cancer. This study investigates the effect of photodynamic therapy performed in prostate tissue samples in vitro, using quantitative magnetic resonance imaging and histopathological analysis. Prostate tissue samples were treated with oxygenated solutions of Rose Bengal (RB) or protoporphyrin IX disodium salt (PpIX), illuminated with visible light, and then analyzed for changes in morphology by microscopy and by measurement of spin–lattice and spin–spin relaxation times at 1.5 Tesla. In the treated prostate tissue samples, histopathological images revealed chromatin condensation and swelling of the stroma, and in some cases, thrombotic necrosis and swelling of the stroma accompanied by pyknotic nuclei occurred. Several samples had protein fragments in the stroma. Magnetic resonance imaging of the treated prostate tissue samples revealed differences in the spin–lattice and spin–spin relaxation times prior to and post photodynamic action.

Evolution of Radiation Therapy in Pancreas Cancer Management toward MRI-Guided Adaptive Radiation Therapy

Pancreas cancer has a poor prognosis despite aggressive treatment and is the fourth leading cause of cancer death in the United States. At diagnosis, most patients have either metastatic or locally advanced disease. In this article, we review the evolution of treatments in locally advanced pancreas cancer (LAPC) and discuss the various radiation therapy fractionation schemes. Furthermore, we examine the data supporting dose escalation and the delivery of ablative biologically effective doses in the setting of LAPC. Finally, we review the role of MRI-guided radiation therapy in escalating dose while sparing organs at risk in the era of stereotactic magnetic resonance-guided adaptive radiation therapy.

Reduction of inter-observer variability using MRI and CT fusion in delineating of primary tumor for radiotherapy in lung cancer with atelectasis

Purpose

To compare the difference between magnetic resonance imaging (MRI) and computed tomography (CT) in delineating the target area of lung cancer with atelectasis.

Method

A retrospective analysis was performed on 15 patients with lung cancer accompanied by atelectasis. All positioning images were transferred to Eclipse treatment planning systems (TPSs). Six MRI sequences (T1WI, T1WI+C, T1WI+C Delay, T1WI+C 10 minutes, T2WI, DWI) were registered with positioning CT. Five radiation oncologists delineated the tumor boundary to obtain the gross tumor volume (GTV). Conformity index (CI) and dice coefficient (DC) were used to measure differences among observers.

Results

The differences in delineation mean volumes, CI, and DC among CT and MRIs were significant. Multiple comparisons were made between MRI sequences and CT. Among them, DWI, T2WI, and T1WI+C 10 minutes sequences were statistically significant with CT in mean volumes, DC, and CI. The mean volume of DWI, T2WI, and T1WI+C 10 minutes sequence in the target area is significantly smaller than that on the CT sequence, but the consistency is higher than that of CT sequences.

Conclusions

The recognition of atelectasis by MRI was better than that by CT, which could reduce interobserver variability of primary tumor delineation in lung cancer with atelectasis. Among them, DWI, T2WI, T1WI+C 10 minutes may be a better choice to improve the GTV delineation of lung cancer patients with atelectasis.

Development of a vendor neutral MRI distortion quality assurance workflow

With the utilization of magnetic resonance (MR) imaging in radiotherapy increasing, routine quality assurance (QA) of these systems is necessary. The assessment of geometric distortion in images used for radiotherapy treatment planning needs to be quantified and monitored over time. This work presents an adaptable methodology for performing routine QA for systematic MRI geometric distortion. A software tool and compatible protocol (designed to work with any CT and MR compatible phantom on any scanner) were developed to quantify geometric distortion via deformable image registration. The MR image is deformed to the CT, generating a deformation field, which is sampled, quantifying geometric distortion as a function of distance from scanner isocenter. Configurability of the QA tool was tested, and results compared to those provided from commercial solutions. Registration accuracy was investigated by repeating the deformable registration step on the initial deformed MR image to define regions with residual distortions. The geometric distortion of four clinical systems was quantified using the customisable QA method presented. Maximum measured distortions varied from 2.2 to 19.4 mm (image parameter and sampling volume dependent). The workflow was successfully customized for different phantom configurations and volunteer imaging studies. Comparison to a vendor supplied solution showed good agreement in regions where the two procedures were sampling the same imaging volume. On a large field of view phantom across various scanners, the QA tool accurately quantified geometric distortions within 17–22 cm from scanner isocenter. Beyond these regions, the geometric integrity of images in clinical applications should be considered with a higher degree of uncertainty due to increased gradient nonlinearity and B₀ inhomogeneity. This tool has been successfully integrated into routine QA of the MRI scanner utilized for radiotherapy within our department. It enables any low susceptibility MR‐CT compatible phantom to quantify the geometric distortion on any MRI scanner with a configurable, user friendly interface for ease of use and consistency in data collection and analysis.

MR-guided adaptive versus ITV-based stereotactic body radiotherapy for hepatic metastases (MAESTRO): a randomized controlled phase II trial

Background

Stereotactic body radiotherapy (SBRT) is an established local treatment method for patients with hepatic oligometastasis or oligoprogression. Liver metastases often occur in close proximity to radiosensitive organs at risk (OARs). This limits the possibility to apply sufficiently high doses needed for optimal local control. Online MR-guided radiotherapy (oMRgRT) is expected to hold potential to improve hepatic SBRT by offering superior soft-tissue contrast for enhanced target identification as well as the benefit of gating and daily real-time adaptive treatment. The MAESTRO trial therefore aims to assess the potential advantages of adaptive, gated MR-guided SBRT compared to conventional SBRT at a standard linac using an ITV (internal target volume) approach.

Methods

This trial is conducted as a prospective, randomized, three-armed phase II study in 82 patients with hepatic metastases (solid malignant tumor, 1–3 hepatic metastases confirmed by magnetic resonance imaging (MRI), maximum diameter of each metastasis ≤ 5 cm (in case of 3 metastases: sum of diameters ≤ 12 cm), age ≥ 18 years, Karnofsky Performance Score ≥ 60%). If a biologically effective dose (BED) ≥ 100 Gy (α/β = 10 Gy) is feasible based on ITV-based planning, patients will be randomized to either MRgRT or ITV-based SBRT. If a lesion cannot be treated with a BED ≥ 100 Gy, the patient will be treated with MRgRT at the highest possible dose. Primary endpoint is the non-inferiority of MRgRT at the MRIdian Linac® system compared to ITV-based SBRT regarding hepatobiliary and gastrointestinal toxicity CTCAE III or higher. Secondary outcomes investigated are local, locoregional (intrahepatic) and distant tumor control, progression-free survival, overall survival, possible increase of BED using MRgRT if the BED is limited with ITV-based SBRT, treatment-related toxicity, quality of life, dosimetric parameters of radiotherapy plans as well as morphological and functional changes in MRI. Potential prognostic biomarkers will also be evaluated.

Discussion

MRgRT is known to be both highly cost- and labor-intensive. The MAESTRO trial aims to provide randomized, higher-level evidence for the dosimetric and possible consecutive clinical benefit of MR-guided, on-table adaptive and gated SBRT for dose escalation in critically located hepatic metastases adjacent to radiosensitive OARs.

Trial registration

The study has been prospectively registered on August 30th, 2021: Clinicaltrials.gov, “Magnetic Resonance-guided Adaptive Stereotactic Body Radiotherapy for Hepatic Metastases (MAESTRO)”, NCT05027711.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13014-022-02033-2.

Multi-sequence MR generated sCT is promising for HNC MR-only RT: a comprehensive evaluation of previously developed sCT generation networks

PURPOSE

To verify the feasibility of our in-house developed multi-sequence magnetic resonance (MR)-generated synthetic computed tomography (sCT) for the accurate dose calculation and fractional positioning for head and neck MR-only radiation therapy (RT).

MATERIALS AND METHODS

Forty-five patients with nasopharyngeal carcinoma were retrospectively studied. By applying our previously in-house developed network, a patient's sCT can rapidly be generated with respect to feeding the sole T1 image, T1C image, T1DixonC image, T2 image, and their combination respectively (five pipelines in total). A k(5)-fold strategy was implemented during model establishment. Dose recalculation was performed for each pipeline generation to attain a dosimetric feasibility evaluation. Fractional positioning evaluation was performed by calculating the digitally reconstructed radiograph (DRR) of the sCT and planning CT and their offset to the portal image.

RESULTS

The dose mean absolute error values are (0.47±0.16)%, (0.48±0.15)% (p<0.05), (0.50±0.16)% (p<0.05), (0.50±0.15)% (p<0.05), and (0.45±0.16)% (p<0.05) for the T1, T1C, T1Dixon C, T2, and 4-channel generated sCT to the prescription dose, respectively. The 4-channel-generated sCT outperforms any other single-sequence pipelines. Among the single-sequence MR imaging-generated sCTs, the T1-generated shows the most accurate HU image quality and provide a reliable dose result. Quantified positioning errors with calculation of the difference to the planning CT offsets are (-0.26±0.50)mm, (-0.58±0.52)mm (p<0.05), (-0.27±0.57)mm (p>0.05), (-0.31±0.44)mm (p>0.05), and (-0.19±0.37)mm (p>0.05) at LNG and (0.34±0.53)mm, (0.48±0.56)mm (p>0.05), (0.55±0.56)mm (p>0.05), (0.37±0.61)mm (p>0.05), and (0.24±0.43)mm (p>0.05) at LAT of the anterior-posterior direction for the five pipelines.

CONCLUSION

Multi-sequence MR-generated sCT allows for accurate dose calculation and fractional positioning for head and neck MR-only RT. This article is protected by copyright. All rights reserved.

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.

Multi-organ auto-delineation in head-and-neck MRI for radiation therapy using regional convolutional neural network

Magnetic resonance imaging (MRI) allows accurate and reliable organ delineation for many disease sites in radiation therapy because MRI is able to offer superb soft-tissue contrast. Manual organ-at-risk delineation is labor-intensive and time-consuming. This study aims to develop a deep-learning-based automated multi-organ segmentation method to release the labor and accelerate the treatment planning process for head-and-neck (HN) cancer radiotherapy. A novel regional convolutional neural network (R-CNN) architecture, namely, mask scoring R-CNN, has been developed in this study. In the proposed model, a deep attention feature pyramid network is used as a backbone to extract the coarse features given by MRI, followed by feature refinement using R-CNN. The final segmentation is obtained through mask and mask scoring networks taking those refined feature maps as input. With the mask scoring mechanism incorporated into conventional mask supervision, the classification error can be highly minimized in conventional mask R-CNN architecture. A cohort of 60 HN cancer patients receiving external beam radiation therapy was used for experimental validation. Five-fold cross-validation was performed for the assessment of our proposed method. The Dice similarity coefficients of brain stem, left/right cochlea, left/right eye, larynx, left/right lens, mandible, optic chiasm, left/right optic nerve, oral cavity, left/right parotid, pharynx, and spinal cord were 0.89 ± 0.06, 0.68 ± 0.14/0.68 ± 0.18, 0.89 ± 0.07/0.89 ± 0.05, 0.90 ± 0.07, 0.67 ± 0.18/0.67 ± 0.10, 0.82 ± 0.10, 0.61 ± 0.14, 0.67 ± 0.11/0.68 ± 0.11, 0.92 ± 0.07, 0.85 ± 0.06/0.86 ± 0.05, 0.80 ± 0.13, and 0.77 ± 0.15, respectively. After the model training, all OARs can be segmented within 1 min.

Assessing the impact of magnetic resonance treatment simulation (MRSIM) on target volume delineation and dose to organs at risk for oropharyngeal radiotherapy

Introduction

Assessing the use of a radiation therapy (RT) planning MRI performed in the treatment position (pMRI) on target volume delineation and effect on organ at risk dose for oropharyngeal cancer patients planned with diagnostic MRI (dMRI) and CT scan.

Methods

Diagnostic MRI scans were acquired for 26 patients in a neutral patient position using a 3T scanner (dMRI). Subsequent pMRI scans were acquired on the same scanner with a flat couch top and the patient in their immobilisation mask. Each series was rigidly registered to the patients planning CT scan and volumes were first completed with the CT/dMRI. The pMRI was then made available for volume modification. For the group with revised volumes, two IMRT plans were developed to demonstrate the impact of the modification. Image and registration quality was also evaluated.

Results

The pMRI registration led to the modification of target volumes for 19 of 26 participants. The pMRI target volumes were larger in absolute volume resulting in reduced capacity for organ sparing. Predominantly, modifications occurred for the primary gross tumour volume (GTVp) with a mean Dice Similarity Coefficient (DSC) of 0.7 and the resulting high risk planning target volume, a mean DSC of 0.89. Both MRIs scored similarly for image quality, with the pMRI demonstrating improved registration quality and efficiency.

Conclusions

A pMRI provides improvement in registration efficiency, quality and a higher degree of oncologist confidence in target delineation. These results have led to a practice change within our department, where a pMRI is acquired for all eligible oropharyngeal cancer patients.

Generating pseudo-computerized tomography (P-CT) scan images from magnetic resonance imaging (MRI) images using machine learning algorithms based on fuzzy theory for radiotherapy treatment planning

PURPOSE

The substitution of computerized tomography (CT) with magnetic resonance imaging (MRI) has been investigated for external radiotherapy treatment planning. The present study aims to use pseudo-CT (P-CT) images created by MRI images to calculate the dose distribution for facilitating the treatment planning process.

METHODS

In this work, following image segmentation with a fuzzy clustering algorithm, an adaptive neuro-fuzzy algorithm was utilized to design the Hounsfield unit (HU) conversion model based on the features vector of MRI images. The model was generated on the set of extracted features from the gray-level co-occurrence matrices and the gray-level run-length matrices for 14 arbitrarily selected patients with brain malady. The performance of the algorithm was investigated on blind datasets through dose-volume histogram and isodose curve evaluations, using the RayPlan treatment planning system (TPS), along with the gamma analysis and statistical indices.

RESULTS

In the proposed approach, the mean absolute error within the range of 45.4 HU was found among the test data. Also, the relative dose difference between the planning target volume region of the CT and the P-CT was 0.5%, and the best gamma pass rate for 3%/3 mm was 97.2%.

CONCLUSION

The proposed method provides a satisfactory average error rate for the generation of P-CT data in the different parts of the brain region from a collection of MRI series. Also, dosimetric parameters evaluation shows good agreement between reference CT and related P-CT images.

Automated delineation of head and neck organs at risk using synthetic MRI-aided mask scoring regional convolutional neural network

PURPOSE

Auto-segmentation algorithms offer a potential solution to eliminate the labor-intensive, time-consuming, and observer-dependent manual delineation of organs-at-risk (OARs) in radiotherapy treatment planning. This study aimed to develop a deep learning-based automated OAR delineation method to tackle the current challenges remaining in achieving reliable expert performance with the state-of-the-art auto-delineation algorithms.

METHODS

The accuracy of OAR delineation is expected to be improved by utilizing the complementary contrasts provided by computed tomography (CT) (bony-structure contrast) and magnetic resonance imaging (MRI) (soft-tissue contrast). Given CT images, synthetic MR images were firstly generated by a pre-trained cycle-consistent generative adversarial network. The features of CT and synthetic MRI were then extracted and combined for the final delineation of organs using mask scoring regional convolutional neural network. Both in-house and public datasets containing CT scans from head-and-neck (HN) cancer patients were adopted to quantitatively evaluate the performance of the proposed method against current state-of-the-art algorithms in metrics including Dice similarity coefficient (DSC), 95th percentile Hausdorff distance (HD95), mean surface distance (MSD), and residual mean square distance (RMS).

RESULTS

Across all of 18 OARs in our in-house dataset, the proposed method achieved an average DSC, HD95, MSD, and RMS of 0.77 (0.58-0.90), 2.90 mm (1.32-7.63 mm), 0.89 mm (0.42-1.85 mm), and 1.44 mm (0.71-3.15 mm), respectively, outperforming the current state-of-the-art algorithms by 6%, 16%, 25%, and 36%, respectively. On public datasets, for all nine OARs, an average DSC of 0.86 (0.73-0.97) were achieved, 6% better than the competing methods.

CONCLUSION

We demonstrated the feasibility of a synthetic MRI-aided deep learning framework for automated delineation of OARs in HN radiotherapy treatment planning. The proposed method could be adopted into routine HN cancer radiotherapy treatment planning to rapidly contour OARs with high accuracy.

Task group 284 report: magnetic resonance imaging simulation in radiotherapy: considerations for clinical implementation, optimization, and quality assurance

The use of dedicated magnetic resonance simulation (MR-SIM) platforms in Radiation Oncology has expanded rapidly, introducing new equipment and functionality with the overall goal of improving the accuracy of radiation treatment planning. However, this emerging technology presents a new set of challenges that need to be addressed for safe and effective MR-SIM implementation. The major objectives of this report are to provide recommendations for commercially available MR simulators, including initial equipment selection, siting, acceptance testing, quality assurance, optimization of dedicated radiation therapy specific MR-SIM workflows, patient-specific considerations, safety, and staffing. Major contributions include guidance on motion and distortion management as well as MRI coil configurations to accommodate patients immobilized in the treatment position. Examples of optimized protocols and checklists for QA programs are provided. While the recommendations provided here are minimum requirements, emerging areas and unmet needs are also highlighted for future development.

Dosimetric comparison of MR-linac-based IMRT and conventional VMAT treatment plans for prostate cancer

Background

The aim of this study was to evaluate and compare the performance of intensity modulated radiation therapy (IMRT) plans, planned for low-field strength magnetic resonance (MR) guided linear accelerator (linac) delivery (labelled IMRT MRL plans), and clinical conventional volumetric modulated arc therapy (VMAT) plans, for the treatment of prostate cancer (PCa). Both plans used the original planning target volume (PTV) margins. Additionally, the potential dosimetric benefits of MR-guidance were estimated, by creating IMRT MRL plans using smaller PTV margins.

Materials and methods

20 PCa patients previously treated with conventional VMAT were considered. For each patient, two different IMRT MRL plans using the low-field MR-linac treatment planning system were created: one with original (orig.) PTV margins and the other with reduced (red.) PTV margins. Dose indices related to target coverage, as well as dose-volume histogram (DVH) parameters for the target and organs at risk (OAR) were compared. Additionally, the estimated treatment delivery times and the number of monitor units (MU) of each plan were evaluated.

Results

The dose distribution in the high dose region and the target volume DVH parameters (D_98%, D_50%, D_2% and V_95%) were similar for all three types of treatment plans, with deviations below 1% in most cases. Both IMRT MRL plans (orig. and red. PTV margins) showed similar homogeneity indices (HI), however worse values for the conformity index (CI) were also found when compared to VMAT. The IMRT MRL plans showed similar OAR sparing when the orig. PTV margins were used but a significantly better sparing was feasible when red. PTV margins were applied. Higher number of MU and longer predicted treatment delivery times were seen for both IMRT MRL plans.

Conclusions

A comparable plan quality between VMAT and IMRT MRL plans was achieved, when applying the same PTV margin. However, online MR-guided adaptive radiotherapy allows for a reduction of PTV margins. With a red. PTV margin, better sparing of the surrounding tissues can be achieved, while maintaining adequate target coverage. Nonetheless, longer treatment delivery times, characteristic for the IMRT technique, have to be expected.

Signal Enhancement of Low Magnetic Field Magnetic Resonance Image Using a Conventional- and Cyclic-Generative Adversarial Network Models With Unpaired Image Sets

In this study, the signal enhancement ratio of low-field magnetic resonance (MR) images was investigated using a deep learning-based algorithm. Unpaired image sets (0.06 Tesla and 1.5 Tesla MR images for different patients) were used in this study following three steps workflow. In the first step, the deformable registration of a 1.5 Tesla MR image into a 0.06 Tesla MR image was performed to ensure that the shapes of the unpaired set matched. In the second step, a cyclic-generative adversarial network (GAN) was used to generate a synthetic MR image of the original 0.06 Tesla MR image based on the deformed or original 1.5 Tesla MR image. Finally, an enhanced 0.06 Tesla MR image could be generated using the conventional-GAN with the deformed or synthetic MR image. The results from the optimized flow and enhanced MR images showed significant signal enhancement of the anatomical view, especially in the nasal septum, inferior nasal choncha, nasopharyngeal fossa, and eye lens. The signal enhancement ratio, signal-to-noise ratio (SNR) and correlation factor between the original and enhanced MR images were analyzed for the evaluation of the image quality. A combined method using conventional- and cyclic-GANs is a promising approach for generating enhanced MR images from low-magnetic-field MR.

A phantom study to contrast and compare polymer and gold fiducial markers in radiotherapy simulation imaging

To assess visibility and artifact characteristics of polymer fiducials compared to standard gold fiducials for radiotherapy CT and MRI simulation. Three gold and three polymer fiducials were inserted into a CT and MRI tissue-equivalent phantom that approximated the prostate cancer radiotherapy configuration. The phantom and fiducials were imaged on CT and MRI. Images were assessed in terms of fiducial visibility and artifact. ImageJ was employed to quantify the pixel gray-scale of each fiducial and artifact. Fiducial gray-scale histograms and profiles were generated for analysis. Objective measurements of the contrast-to-noise ratio (CNR), signal-to-noise ratio (SNR), and artifact index (AI) were calculated. The CT images showed that the gold fiducials are visually brighter, with greater contrast than the polymer. The higher peak values illustrate this in the line profiles. However, they produce bright radiating and dark shadowing artifacts. This is depicted by the greater width of line profiles and the disruption of phantom area profiles. Quantitatively this results in greater percentile ranges of the histograms. Furthermore, for CT, gold had a higher CNR than polymer, relative to the phantom. However, the gold CNR and SNR were degraded by the greater artifact and thus AI. Both fiducials were visible on MRI and had similar histograms and profiles that were also reflected in comparable CNR, SNR and AI. Polymer fiducials were well visualized in a phantom on CT and MR and produce less artifact than the gold fiducials. Polymer markers could enhance the quality and accuracy of radiotherapy co-registration and planning but require clinical confirmation.

Early Monitoring Response to Therapy in Patients with Brain Lesions Using the Cumulative SUV Histogram

Gamma Knife treatment is an alternative to traditional brain surgery and whole-brain radiation therapy for treating cancers that are inaccessible via conventional treatments. To assess the effectiveness of Gamma Knife treatments, functional imaging can play a crucial role. The aim of this study is to evaluate new prognostic indices to perform an early assessment of treatment response to therapy using positron emission tomography imaging. The parameters currently used in nuclear medicine assessments can be affected by statistical fluctuation errors and/or cannot provide information on tumor extension and heterogeneity. To overcome these limitations, the Cumulative standardized uptake value (SUV) Histogram (CSH) and Area Under the Curve (AUC) indices were evaluated to obtain additional information on treatment response. For this purpose, the absolute level of [11C]-Methionine (MET) uptake was measured and its heterogeneity distribution within lesions was evaluated by calculating the CSH and AUC indices. CSH and AUC parameters show good agreement with patient outcomes after Gamma Knife treatments. Furthermore, no relevant correlations were found between CSH and AUC indices and those usually used in the nuclear medicine environment. CSH and AUC indices could be a useful tool for assessing patient responses to therapy.

Cone beam computed tomography for dose calculation quality assurance for magnetic resonance-only radiotherapy

•

Clinical Magnetic Resonance (MR)-only radiotherapy requires a dose quality assurance method.

•

Doses calculated on Cone Beam Computed Tomography (CBCT) were within 2% of MR-only doses calculated using synthetic CT.

•

CBCT with asymmetric dose difference tolerances of [−2%,1%] appears clinically feasible for quality assurance of prostate MR-only radiotherapy.

Background and purpose

Magnetic Resonance (MR)-only prostate radiotherapy using synthetic Computed Tomography (sCT) algorithms with high dose accuracy has been clinically implemented. MR images can suffer from geometric distortions so Quality Assurance (QA) using an independent, geometrically accurate, image could be required. The first-fraction Cone Beam CT (CBCT) has demonstrated potential but has not been evaluated in a clinical MR-only pathway. This study evaluated the clinical use of CBCT for dose accuracy QA of MR-only radiotherapy.

Materials and methods

A total of 49 patients treated with MR-only prostate radiotherapy were divided into two cohorts. Cohort 1 (20 patients) received a back-up CT, whilst Cohort 2 (29 patients) did not. All patients were planned using the sCT and received daily CBCT imaging with MR-CBCT soft-tissue matching. Each CBCT was calibrated using a patient-specific stepwise Hounsfield Units-to-mass density curve. The treatment plan was recalculated on the first-fraction CBCT using the clinically applied soft-tissue match and the doses compared. For Cohort 1 the sCT was rigidly registered to the back-up CT, the plan recalculated and doses compared.

Results

Mean sCT-CBCT dose difference across both cohorts was -0.6±0.1% (standard error of the mean, range −2.3%,2.3%), with 47/49 patients within [-2%,1%]. The sCT-CBCT dose difference was systematically lower than the sCT-CT by -0.7±0.6% (±95% limits of agreement). The mean sCT-CBCT gamma pass rate (2%/2mm) was 96.1±0.4% (85.4%,99.7%).

Conclusions

CBCT-based dose accuracy QA for MR-only radiotherapy appears clinically feasible. There was a small systematic sCT-CBCT dose difference implying asymmetric tolerances of [-2%,1%] would be appropriate.

A review on medical imaging synthesis using deep learning and its clinical applications

This paper reviewed the deep learning-based studies for medical imaging synthesis and its clinical application. Specifically, we summarized the recent developments of deep learning-based methods in inter- and intra-modality image synthesis by listing and highlighting the proposed methods, study designs, and reported performances with related clinical applications on representative studies. The challenges among the reviewed studies were then summarized with discussion.

Patient-Reported Tolerance of Magnetic Resonance-Guided Radiation Therapy

Purpose

Magnetic resonance-guided radiation therapy (MRgRT) has been incorporated into a growing number of clinical practices world-wide, however, there is limited data on patient experiences with MRgRT. The purpose of this study was to prospectively evaluate patient tolerance of MRgRT using patient reported outcome questionnaires (PRO-Q).

Methods

Ninety patients were enrolled in this prospective observational study and treated with MRgRT (MRIdian Linac System, ViewRay Inc. Oakwood Village, OH, United States) between September 2018 and September 2019. Breath-hold-gated dose delivery with audiovisual feedback was completed as needed. Patients completed an in-house developed PRO-Q after the first and last fraction of MRgRT.

Results

The most commonly treated anatomic sites were the abdomen (47%) and pelvis (33%). Respiratory gating was utilized in 62% of the patients. Patients rated their experience as positive or at least tolerable with mean scores of 1.0–2.8. The most common complaint was the temperature in the room (61%) followed by paresthesias (57%). The degree of anxiety reported by 45% of the patients significantly decreased at the completion of treatment (mean score 1.54 vs. 1.36, p = 0.01). Forty-three percent of the patients reported some degree of disturbing noise which was improved considerably by use of music. All patients appreciated their active role during the treatment.

Conclusion

This evaluation of PROs indicates that MRgRT was well-tolerated by our patients. Patients’ experience may further improve with adjustment of room temperature and noise reduction.

Applicator visualization using ultrashort echo time MRI for high-dose-rate endorectal brachytherapy

PURPOSE

The individual channels in an endorectal applicator for high-dose-rate endorectal brachytherapy are not visible on standard MRI sequences. The aim of this study was to test whether an ultrashort echo time (UTE) MRI sequence could be used to visualize the individual channels to enable MR-only treatment planning for rectal cancer.

METHODS AND MATERIALS

We used a radial three-dimensional (3D) UTE pulse sequence and acquired images of phantoms and two patients with rectal cancer. We rigidly registered a UTE image and CT scan of an applicator phantom, based on the outline of the applicator. One observer compared channel positions on the UTE image and CT scan in five slices spaced 25 mm apart. To quantify geometric distortions, we scanned a commercial 3D geometric quality assurance phantom and calculated the difference between detected marker positions on the UTE image and corresponding marker positions on two 3D T₁-weighted images with opposing readout directions.

RESULTS

On the UTE images, there is sufficient contrast to discern the individual channels. The difference in channel positions on the UTE image compared with the CT was on average -0.1 ± 0.1 mm (left-right) and 0.1 ± 0.3 mm (anteroposterior). After rigid registration to the 3D T₁-weighted sequences, the residual 95th percentile of the geometric distortion inside a 550-mm-diameter sphere was 1.0 mm (left-right), 0.9 mm (anteroposterior), and 0.9 mm (craniocaudal).

CONCLUSIONS

With a UTE sequence, the endorectal applicator and individual channels can be adequately visualized in both phantom and patients. The geometrical fidelity is within an acceptable range.

Estimation and evaluation of pseudo-CT images using linear regression models and texture feature extraction from MRI images in the brain region to design external radiotherapy planning

Aim

The aim of this study is to construct and evaluate Pseudo-CT images (P-CTs) for electron density calculation to facilitate external radiotherapy treatment planning.

Background

Despite numerous benefits, computed tomography (CT) scan does not provide accurate information on soft tissue contrast, which often makes it difficult to precisely differentiate target tissues from the organs at risk and determine the tumor volume. Therefore, MRI imaging can reduce the variability of results when registering with a CT scan.

Materials and methods

In this research, a fuzzy clustering algorithm was used to segment images into different tissues, also linear regression methods were used to design the regression model based on the feature extraction method and the brightness intensity values. The results of the proposed algorithm for dose-volume histogram (DVH), Isodose curves, and gamma analysis were investigated using the RayPlan treatment planning system, and VeriSoft software. Furthermore, various statistical indices such as Mean Absolute Error (MAE), Mean Error (ME), and Structural Similarity Index (SSIM) were calculated.

Results

The MAE of a range of 45-55 was found from the proposed methods. The relative difference error between the PTV region of the CT and the Pseudo-CT was 0.5, and the best gamma rate was 95.4% based on the polar coordinate feature and proposed polynomial regression model.

Conclusion

The proposed method could support the generation of P-CT data for different parts of the brain region from a collection of MRI series with an acceptable average error rate by different evaluation criteria.

Quantitative imaging for radiotherapy purposes

Quantitative imaging biomarkers show great potential for use in radiotherapy. Quantitative images based on microscopic tissue properties and tissue function can be used to improve contouring of the radiotherapy targets. Furthermore, quantitative imaging biomarkers might be used to predict treatment response for several treatment regimens and hence be used as a tool for treatment stratification, either to determine which treatment modality is most promising or to determine patient-specific radiation dose. Finally, patient-specific radiation doses can be further tailored to a tissue/voxel specific radiation dose when quantitative imaging is used for dose painting. In this review, published standards, guidelines and recommendations on quantitative imaging assessment using CT, PET and MRI are discussed. Furthermore, critical issues regarding the use of quantitative imaging for radiation oncology purposes and resultant pending research topics are identified.

Multi-sequence MR image-based synthetic CT generation using a generative adversarial network for head and neck MRI-only radiotherapy

PURPOSE

The purpose of this study is to investigate the effect of different magnetic resonance (MR) sequences on the accuracy of deep learning-based synthetic computed tomography (sCT) generation in the complex head and neck region.

METHODS

Four sequences of MR images (T1, T2, T1C, and T1DixonC-water) were collected from 45 patients with nasopharyngeal carcinoma. Seven conditional generative adversarial network (cGAN) models were trained with different sequences (single channel) and different combinations (multi-channel) as inputs. To further verify the cGAN performance, we also used a U-net network as a comparison. Mean absolute error, structural similarity index, peak signal-to-noise ratio, dice similarity coefficient, and dose distribution were evaluated between the actual CTs and sCTs generated from different models.

RESULTS

The results show that the cGAN model with multi-channel (i.e., T1 + T2 + T1C + T1DixonC-water) as input to predict sCT achieves higher accuracy than any single MR sequence model. The T1-weighted MR model achieves better results than T2, T1C, and T1DixonC-water models. The comparison between cGAN and U-net shows that the sCTs predicted by cGAN retains additional image details are less blurred and more similar to the actual CT.

CONCLUSIONS

Conditional generative adversarial network with multiple MR sequences as model input shows the best accuracy. The T1-weighted MR images provide sufficient image information and are suitable for sCT prediction in clinical scenarios with limited acquisition sequences or limited acquisition time.

Field-map correction in read-out segmented echo planar imaging for reduced spatial distortion in prostate DWI for MRI-guided radiotherapy applications

Diffusion-weighted echo planar imaging (DW-EPI) suffers from geometric distortion due to low phase-encoding bandwidth. Read-out segmented echo planar imaging (RS-EPI) reduces distortion but residual distortion remains in extreme cases. Additional corrections need to be applied, especially for radiotherapy applications where a high degree of accuracy is needed. In this study the use of magnetic field map corrections are assessed in DW-EPI and RS-EPI, to reduce geometric uncertainty for MRI-guided radiotherapy applications. Magnetic field maps were calculated from gradient echo images and distortion corrections were applied to RS-EPI images. Distortions were assessed in a prostate phantom by comparing to the known geometry, and in vivo using a modified Hausdorff distance metric using a T2-weighted spin echo as ground truth. Across 10 patients, field map-corrected RS-EPI reduced maximum distortion by 5 mm on average compared to DW-EPI (σ = 1.9 mm). Geometric distortions were also reduced significantly using field mapping with RS-EPI, compared to RS-EPI alone (p ≤ 0.05). The increased geometric accuracy of these techniques can potentially allow diffusion-weighted images to be fused with other MR or CT images for radiotherapy treatment purposes.

Magnetic resonance imaging sequence evaluation of an MR Linac system; early clinical experience

OBJECTIVES

To systematically identify the preferred magnetic resonance imaging (MRI) sequences following volunteer imaging on a 1.5 Tesla (T) MR-Linear Accelerator (MR Linac) for future protocol development.

METHODS

Non-patient volunteers were recruited to a Research and Ethics committee approved prospective MR-only imaging study on a 1.5T MR Linac system. Volunteers attended 1-3 imaging sessions that included a combination of mDixon, T1w, T2w sequences using 2-dimensional (2D) and 3-dimensional (3D) acquisitions. Each sequence was acquired over 2-7 minutes and reviewed by a panel of 3 observers to evaluate image quality using a visual grading analysis based on a 4-point Likert scale. Sequences were acquired and modified iteratively until deemed fit for purpose (online image matching or re-planning) and all observers agreed they were suitable in 3 volunteers.

RESULTS

26 volunteers underwent 31 imaging sessions of six general anatomical regions. Images were acquired in one or two of six general anatomical regions: male pelvis (n = 9), female pelvis (n = 4), chestwall/breast (n = 5), lung/oesophagus (n = 5), abdomen (n = 3) and head and neck (n = 5). Images were acquired using a pre-defined exam-card that on average, included six sequences (range 2-10), with a maximum scan time of approximately one hour. The majority of observers preferred T2-weighted sequences. The thorax teams were the only groups to prefer T1-weighted imaging.

CONCLUSIONS

An iterative process identified sequence agreement in all anatomical regions. These sequences will now be evaluated in patient volunteers.

ADVANCES IN KNOWLEDGE

This manuscript is the first publication sharing the results of the first systematic selection of MRI sequences for use in on-board MRI-guided radiotherapy by end-users (therapeutic radiographers and clinical oncologists) in healthy volunteers.

The accuracy of Magnetic Resonance - Cone Beam Computed Tomography soft-tissue matching for prostate radiotherapy

BACKGROUND AND PURPOSE

Magnetic Resonance (MR)-Only radiotherapy requires a method for matching image with on-treatment Cone Beam Computed Tomography (CBCT). This study aimed to investigate the accuracy of MR-CBCT soft-tissue matching for prostate MR-only radiotherapy.

MATERIALS AND METHODS

Three patient cohorts were used, with all patients receiving MR and CT scans. For the first cohort (10 patients) the first fraction CBCT was automatically rigidly registered to the CT and MR scans and the MR-CT registration predicted using the MR-CBCT and CT-CBCT registrations. This was compared to the automatic MR-CT registration. For the second and third cohorts (five patients each) the first fraction CBCT was independently matched to the CT and MR by four radiographers, the MR-CBCT and CT-CBCT matches compared and the inter-observer variability assessed. The second cohort used a CT-based structure set and the third a MR-based structure set with the MR relabelled as a 'CT'.

RESULTS

The mean difference between predicted and actual MR-CT registrations was Δ R All = - 0.1 ± 0.2 mm (s.e.m.). Radiographer MR-CBCT registrations were not significantly different to CT-CBCT, with mean differences in soft-tissue match ⩽ 0.2 mm and all except one difference ⩽ 3.3 mm . This was less than the MR-CBCT inter-observer limits of agreement [ 3.5 , 2.4 , 0.9 ] mm (vertical, longitudinal, lateral), which were similar ( ⩽ 0.5 mm ) to CT-CBCT.

CONCLUSIONS

MR-CBCT soft-tissue matching is not significantly different to CT-CBCT. Relabelling the MR as a 'CT' does not appear to change the automatic registration. This suggests that MR-CBCT soft-tissue matching is feasible and accurate.

Evaluation of proton and photon dose distributions recalculated on 2D and 3D Unet-generated pseudoCTs from T1-weighted MR head scans

Introduction: The recent developments of magnetic resonance (MR) based adaptive strategies for photon and, potentially for proton therapy, require a fast and reliable conversion of MR images to X-ray computed tomography (CT) values. CT values are needed for photon and proton dose calculation. The improvement of conversion results employing a 3D deep learning approach is evaluated. Material and methods: A database of 89 T1-weighted MR head scans with about 100 slices each, including rigidly registered CTs, was created. Twenty-eight validation patients were randomly sampled, and four patients were selected for application. The remaining patients were used to train a 2D and a 3D U-shaped convolutional neural network (Unet). A stack size of 32 slices was used for 3D training. For all application cases, volumetric modulated arc therapy photon and single-field uniform dose pencil-beam scanning proton plans at four different gantry angles were optimized for a generic target on the CT and recalculated on 2D and 3D Unet-based pseudoCTs. Mean (absolute) error (MAE/ME) and a gradient sharpness estimate were used to quantify the image quality. Three-dimensional gamma and dose difference analyses were performed for photon (gamma criteria: 1%, 1 mm) and proton dose distributions (gamma criteria: 2%, 2 mm). Range (80% fall off) differences for beam's eye view profiles were evaluated for protons. Results: Training 36 h for 1000 epochs in 3D (6 h for 200 epochs in 2D) yielded a maximum MAE of 147 HU (135 HU) for the application patients. Except for one patient gamma pass rates for photon and proton dose distributions were above 96% for both Unets. Slice discontinuities were reduced for 3D training at the cost of sharpness. Conclusions: Image analysis revealed a slight advantage of 2D Unets compared to 3D Unets. Similar dose calculation performance was reached for the 2D and 3D network.

MRI-Based Proton Treatment Planning for Base of Skull Tumors

Purpose

To introduce a novel, deep-learning method to generate synthetic computed tomography (SCT) scans for proton treatment planning and evaluate its efficacy.

Materials and Methods

50 Patients with base of skull tumors were divided into 2 nonoverlapping training and study cohorts. Computed tomography and magnetic resonance imaging pairs for patients in the training cohort were used for training our novel 3-dimensional generative adversarial network (cycleGAN) algorithm. Upon completion of the training phase, SCT scans for patients in the study cohort were predicted based on their magnetic resonance images only. The SCT scans obtained were compared against the corresponding original planning computed tomography scans as the ground truth, and mean absolute errors (in Hounsfield units [HU]) and normalized cross-correlations were calculated. Proton plans of 45 Gy in 25 fractions with 2 beams per plan were generated for the patients based on their planning computed tomographies and recalculated on SCT scans. Dose-volume histogram endpoints were compared. A γ-index analysis along 3 cardinal planes intercepting at the isocenter was performed. Proton distal range along each beam was calculated.

Results

Image quality metrics show agreement between the generated SCT scans and the ground truth with mean absolute error values ranging from 38.65 to 65.12 HU and an average of 54.55 ± 6.81 HU and a normalized cross-correlation average of 0.96 ± 0.01. The dosimetric evaluation showed no statistically significant differences (p > 0.05) within planning target volumes for dose-volume histogram endpoints and other metrics studied, with the exception of the dose covering 95% of the target volume, with a relative difference of 0.47%. The γ-index analysis showed an average passing rate of 98% with a 10% threshold and 2% and 2-mm criteria. Proton ranges of 48 of 50 beams (96%) in this study were within clinical tolerance adopted by 4 institutions.

Conclusions

This study shows our method is capable of generating SCT scans with acceptable image quality, dose distribution agreement, and proton distal range compared with the ground truth. Our results set a promising approach for magnetic resonance imaging-based proton treatment planning.

The rationale for MR-only treatment planning for external radiotherapy

•MR-only treatment planning could improve the spatial accuracy of radiotherapy.•The benefit compared to a mixed MR-CT workflow will vary between patient groups.•Further development of QA tools is needed before the procedure will save resources.

MRI basics for radiation oncologists

•

Issues of MRI that are relevant for radiation oncologists are addressed.

•

Radiation oncology requires dedicated scan protocols.

•

Use of diagnostic protocols is not recommended for radiotherapy.

•

MR images must be made in treatment position with the standard positioning devices.

•

Safety screening prior to entering the MRI room is crucial.

MRI is increasingly used in radiation oncology to facilitate tumor and organ-at-risk delineation and image guidance. In this review, we address issues of MRI that are relevant for radiation oncologists when interpreting MR images offered for radiotherapy. Whether MRI is used in combination with CT or in an MRI-only workflow, it is generally necessary to ensure that MR images are acquired in treatment position, using the positioning and fixation devices that are commonly applied in radiotherapy. For target delineation, often a series of separate image sets are used with distinct image contrasts, acquired within a single exam. MR images can suffer from image distortions. While this can be avoided with dedicated scan protocols, in a diagnostic setting geometrical fidelity is less relevant and is therefore less accounted for. Since geometrical fidelity is of utmost importance in radiation oncology, it requires dedicated scan protocols. The strong magnetic field of an MRI scanner and the use of radiofrequency radiation can cause safety hazards if not properly addressed. Safety screening is crucial for every patient and every operator prior to entering the MRI room.

Deep learning approaches using 2D and 3D convolutional neural networks for generating male pelvic synthetic computed tomography from magnetic resonance imaging

PURPOSE

The improved soft tissue contrast of magnetic resonance imaging (MRI) compared to computed tomography (CT) makes it a useful imaging modality for radiotherapy treatment planning. Even when MR images are acquired for treatment planning, the standard clinical practice currently also requires a CT for dose calculation and x-ray-based patient positioning. This increases workloads, introduces uncertainty due to the required inter-modality image registrations, and involves unnecessary irradiation. While it would be beneficial to use exclusively MR images, a method needs to be employed to estimate a synthetic CT (sCT) for generating electron density maps and patient positioning reference images. We investigated 2D and 3D convolutional neural networks (CNNs) to generate a male pelvic sCT using a T1-weighted MR image and compare their performance.

METHODS

A retrospective study was performed using CTs and T1-weighted MR images of 20 prostate cancer patients. CTs were deformably registered to MR images to create CT-MR pairs for training networks. The proposed 2D CNN, which contained 27 convolutional layers, was modified from the state-of-the-art 2D CNN to save computational memory and prepare for building the 3D CNN. The proposed 2D and 3D models were trained from scratch to map intensities of T1-weighted MR images to CT Hounsfield Unit (HU) values. Each sCT was generated in a fivefold cross-validation framework and compared with the corresponding deformed CT (dCT) using voxel-wise mean absolute error (MAE). The sCT geometric accuracy was evaluated by comparing bone regions, defined by thresholding at 150 HU in the dCTs and the sCTs, using dice similarity coefficient (DSC), recall, and precision. To evaluate sCT patient positioning accuracy, bone regions in dCTs and sCTs were rigidly registered to the corresponding cone-beam CTs. The resulting paired Euler transformation vectors were compared by calculating translation vector distances and absolute differences of Euler angles. Statistical tests were performed to evaluate the differences among the proposed models and Han's model.

RESULTS

Generating a pelvic sCT required approximately 5.5 s using the proposed models. The average MAEs within the body contour were 40.5 ± 5.4 HU (mean ± SD) and 37.6 ± 5.1 HU for the 2D and 3D CNNs, respectively. The average DSC, recall, and precision for the bone region (thresholding the CT at 150 HU) were 0.81 ± 0.04, 0.85 ± 0.04, and 0.77 ± 0.09 for the 2D CNN, and 0.82 ± 0.04, 0.84 ± 0.04, and 0.80 ± 0.08 for the 3D CNN, respectively. For both models, mean translation vector distances are less than 0.6 mm with mean absolute differences of Euler angles less than 0.5°.

CONCLUSIONS

The 2D and 3D CNNs generated accurate pelvic sCTs for the 20 patients using T1-weighted MR images. Statistical tests indicated that the proposed 3D model was able to generate sCTs with smaller MAE and higher bone region precision compared to 2D models. Results of patient alignment tests suggested that sCTs generated by the proposed CNNs can provide accurate patient positioning. The accuracy of the dose calculation using generated sCTs will be tested and compared for the proposed models in the future.