A review of substitute CT generation for MRI-only radiation therapy

Generating synthetic computed tomography for radiotherapy: SynthRAD2023 challenge report

Radiation therapy plays a crucial role in cancer treatment, necessitating precise delivery of radiation to tumors while sparing healthy tissues over multiple days. Computed tomography (CT) is integral for treatment planning, offering electron density data crucial for accurate dose calculations. However, accurately representing patient anatomy is challenging, especially in adaptive radiotherapy, where CT is not acquired daily. Magnetic resonance imaging (MRI) provides superior soft-tissue contrast. Still, it lacks electron density information, while cone beam CT (CBCT) lacks direct electron density calibration and is mainly used for patient positioning. Adopting MRI-only or CBCT-based adaptive radiotherapy eliminates the need for CT planning but presents challenges. Synthetic CT (sCT) generation techniques aim to address these challenges by using image synthesis to bridge the gap between MRI, CBCT, and CT. The SynthRAD2023 challenge was organized to compare synthetic CT generation methods using multi-center ground truth data from 1080 patients, divided into two tasks: (1) MRI-to-CT and (2) CBCT-to-CT. The evaluation included image similarity and dose-based metrics from proton and photon plans. The challenge attracted significant participation, with 617 registrations and 22/17 valid submissions for tasks 1/2. Top-performing teams achieved high structural similarity indices (≥0.87/0.90) and gamma pass rates for photon (≥98.1%/99.0%) and proton (≥97.3%/97.0%) plans. However, no significant correlation was found between image similarity metrics and dose accuracy, emphasizing the need for dose evaluation when assessing the clinical applicability of sCT. SynthRAD2023 facilitated the investigation and benchmarking of sCT generation techniques, providing insights for developing MRI-only and CBCT-based adaptive radiotherapy. It showcased the growing capacity of deep learning to produce high-quality sCT, reducing reliance on conventional CT for treatment planning.

CycleSGAN: A cycle-consistent and semantics-preserving generative adversarial network for unpaired MR-to-CT image synthesis

CycleGAN has been leveraged to synthesize a CT image from an available MR image after trained on unpaired data. Due to the lack of direct constraints between the synthetic and the input images, CycleGAN cannot guarantee structural consistency and often generates inaccurate mappings that shift the anatomy, which is highly undesirable for downstream clinical applications such as MRI-guided radiotherapy treatment planning and PET/MRI attenuation correction. In this paper, we propose a cycle-consistent and semantics-preserving generative adversarial network, referred as CycleSGAN, for unpaired MR-to-CT image synthesis. Our design features a novel and generic way to incorporate semantic information into CycleGAN. This is done by designing a pair of three-player games within the CycleGAN framework where each three-player game consists of one generator and two discriminators to formulate two distinct types of adversarial learning: appearance adversarial learning and structure adversarial learning. These two types of adversarial learning are alternately trained to ensure both realistic image synthesis and semantic structure preservation. Results on unpaired hip MR-to-CT image synthesis show that our method produces better synthetic CT images in both accuracy and visual quality as compared to other state-of-the-art (SOTA) unpaired MR-to-CT image synthesis methods.

Deep learning-based pseudo-CT synthesis from zero echo time MR sequences of the pelvis

OBJECTIVES

To generate pseudo-CT (pCT) images of the pelvis from zero echo time (ZTE) MR sequences and compare them to conventional CT.

METHODS

Ninety-one patients were prospectively scanned with CT and MRI including ZTE sequences of the pelvis. Eleven ZTE image volumes were excluded due to implants and severe B1 field inhomogeneity. Out of the 80 data sets, 60 were used to train and update a deep learning (DL) model for pCT image synthesis from ZTE sequences while the remaining 20 cases were selected as an evaluation cohort. CT and pCT images were assessed qualitatively and quantitatively by two readers.

RESULTS

Mean pCT ratings of qualitative parameters were good to perfect (2-3 on a 4-point scale). Overall intermodality agreement between CT and pCT was good (ICC = 0.88 (95% CI: 0.85-0.90); p < 0.001) with excellent interreader agreements for pCT (ICC = 0.91 (95% CI: 0.88-0.93); p < 0.001). Most geometrical measurements did not show any significant difference between CT and pCT measurements (p > 0.05) with the exception of transverse pelvic diameter measurements and lateral center-edge angle measurements (p = 0.001 and p = 0.002, respectively). Image quality and tissue differentiation in CT and pCT were similar without significant differences between CT and pCT CNRs (all p > 0.05).

CONCLUSIONS

Using a DL-based algorithm, it is possible to synthesize pCT images of the pelvis from ZTE sequences. The pCT images showed high bone depiction quality and accurate geometrical measurements compared to conventional CT. CRITICAL RELEVANCE STATEMENT: pCT images generated from MR sequences allow for high accuracy in evaluating bone without the need for radiation exposure. Radiological applications are broad and include assessment of inflammatory and degenerative bone disease or preoperative planning studies.

KEY POINTS

pCT, based on DL-reconstructed ZTE MR images, may be comparable with true CT images. Overall, the intermodality agreement between CT and pCT was good with excellent interreader agreements for pCT. Geometrical measurements and tissue differentiation were similar in CT and pCT images.

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.

Evaluation of magnetic resonance imaging derived synthetic computed tomography for proton therapy planning in prostate cancer

Background and purpose

Magnetic resonance imaging (MRI)-only workflow is used in photon radiotherapy (RT) today, but not yet for protons. To bring MRI-only proton RT into clinical use, proton dose calculation on MRI-derived synthetic CT (sCT) must be validated. We evaluated proton dose calculation accuracy of prostate cancer proton plans using a commercially available sCT generator already validated for photon planning.

Materials and methods

The retrospective planning study included 10 prostate cancer patients who underwent MRI and planning CT (pCT) before RT. sCT were generated from the MRI with MRI Planner v2.3, and compared to pCT using structural mean absolute error (MAE). The pCT was used to create one-arc volumetric modulated arc therapy (VMAT) photon plan and two-field intensity modulated proton therapy (IMPT) proton plan. Each plan was recalculated on the sCT and compared to pCT doses. Dose volume histogram parameters, gamma analyses and range differences were evaluated.

Results

Median MAE for the body contour was 71 HU. Dose differences between pCT and sCT were small and similar for VMAT and IMPT plans. Median (range) gamma pass rates were lower for IMPT plans with 95.8 (89.3-98.7) % compared to VMAT plans with 99.4 (91.2-99.6) %. The proton range difference was 1.0 (interquartile range -0.1 - 0.2) mm deeper for sCT compared to the reference.

Conclusion

MRI-only IMPT planning for prostate cancer seems feasible in a clinical setting for the evaluated beam arrangement and sCT generator. More patients and evaluation of other beam arrangements are needed for a more general conclusion.

3D Unsupervised deep learning method for magnetic resonance imaging-to-computed tomography synthesis in prostate radiotherapy

Background and purpose

Magnetic resonance imaging (MRI)-to-computed tomography (CT) synthesis is essential in MRI-only radiotherapy workflows, particularly through deep learning techniques known for their accuracy. However, current supervised methods are limited to specific center's learnings and depend on registration precision. The aim of this study was to evaluate the accuracy of unsupervised and supervised approaches in the context of prostate MRI-to-CT generation for radiotherapy dose calculation.

Methods

CT/MRI image pairs from 99 prostate cancer patients across three different centers were used. A comparison between supervised and unsupervised conditional Generative Adversarial Networks (cGAN) was conducted. Unsupervised training incorporates a style transfer method with. Content and Style Representation for Enhanced Perceptual synthesis (CREPs) loss. For dose evaluation, the photon prescription dose was 60 Gy delivered in volumetric modulated arc therapy (VMAT). Imaging endpoint for sCT evaluation was Mean Absolute Error (MAE). Dosimetric endpoints included absolute dose differences and gamma analysis between CT and sCT dose calculations.

Results

The unsupervised paired network exhibited the highest accuracy for the body with a MAE at 33.6 HU, the highest MAE was 45.5 HU obtained with unsupervised unpaired learning. All architectures provided clinically acceptable results for dose calculation with gamma pass rates above 94 % (1 % 1 mm 10 %).

Conclusions

This study shows that multicenter data can produce accurate sCTs via unsupervised learning, eliminating CT-MRI registration. The sCTs not only matched HU values but also enabled precise dose calculations, suggesting their potential for wider use in MRI-only radiotherapy workflows.

Linac- and CyberKnife-based MRI-only treatment planning of prostate SBRT using an optimized synthetic CT calibration curve

PURPOSE

CT Hounsfield Units (HUs) are converted to electron density using a calibration curve obtained from physical measurements of an electron density phantom. HU values assigned to an MRI-derived synthetic computed tomography (sCT) may present a different relationship with electron density compared to CT HU. Correct assignment of sCT HU values is critical for accurate dose calculation and delivery. The goals of this work were to develop a sCT calibration curve using patient data acquired on a clinically commissioned CT scanner and assess for CyberKnife- and volumetric modulated arc therapy (VMAT)-based MR-only treatment planning of prostate SBRT.

METHODS

Same-day CT and MRI simulation in the treatment position were performed on 10 patients treated with SBRT to the prostate. Dixon in-phase and out-of-phase MRIs were acquired on a 3T scanner using a 3D T1-weighted gradient-echo sequence to generate sCTs using a commercial sCT algorithm. CT and sCT datasets were co-registered and HU values compared using mean absolute error (MAE). An optimized HU-to-density calibration curve was created based on average HU values across an institutional patient database for each of the four sCT tissue types. Clinical CyberKnife and VMAT treatment plans were generated on each patient CT and recomputed onto corresponding sCTs. Dose distributions computed using CT and sCT were compared using gamma criteria and dose-volume-histograms.

RESULTS

For the optimized calibration curve, HU values were -96, 37, 204, and 1170 and relative electron densities were 0.95, 1.04, 1.1, and 1.7 for adipose, soft tissue, inner bone, and outer bone, respectively. The proposed sCT protocol produced total MAE of 94 ± 20HU. Gamma values mean ± std (min-max) were 98.9% ± 0.9% (97.1%-100%) and 97.7% ± 1.3% (95.3%-99.3%) for VMAT and CyberKnife plans, respectively.

CONCLUSION

MRI-derived sCT using the proposed approach shows excellent dosimetric agreement with conventional CT simulation, demonstrating the feasibility of MRI-derived sCT for prostate SBRT treatment planning.

Magnetic resonance imaging-based bone imaging of the lower limb: Strategies for generating high-resolution synthetic computed tomography

This study aims at assessing approaches for generating high-resolution magnetic resonance imaging- (MRI-) based synthetic computed tomography (sCT) images suitable for orthopedic care using a deep learning model trained on low-resolution computed tomography (CT) data. To that end, paired MRI and CT data of three anatomical regions were used: high-resolution knee and ankle data, and low-resolution hip data. Four experiments were conducted to investigate the impact of low-resolution training CT data on sCT generation and to find ways to train models on low-resolution data while providing high-resolution sCT images. Experiments included resampling of the training data or augmentation of the low-resolution data with high-resolution data. Training sCT generation models using low-resolution CT data resulted in blurry sCT images. By resampling the MRI/CT pairs before the training, models generated sharper images, presumably through an increase in the MRI/CT mutual information. Alternatively, augmenting the low-resolution with high-resolution data improved sCT in terms of mean absolute error proportionally to the amount of high-resolution data. Overall, the morphological accuracy was satisfactory as assessed by an average intermodal distance between joint centers ranging from 0.7 to 1.2 mm and by an average intermodal root-mean-squared distances between bone surfaces under 0.7 mm. Average dice scores ranged from 79.8% to 87.3% for bony structures. To conclude, this paper proposed approaches to generate high-resolution sCT suitable for orthopedic care using low-resolution data. This can generalize the use of sCT for imaging the musculoskeletal system, paving the way for an MR-only imaging with simplified logistics and no ionizing radiation.

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.

Advances in Imaging (Intraop Cone-Beam Computed Tomography, Synthetic Computed Tomography, Bone Scan, Low-Dose Protocols)

Spine surgery has seen a rapid advance in the refinement and development of 3-dimensional and nuclear imaging modalities in recent years. Cone-beam CT has proven to be a valuable tool for improving the accuracy of pedicle screw placement. The use of synthetic CT and low-dose CT have also emerged as modalities which allow for little to no radiation while streamlining imaging workflows. Bone scans also serve to provide functional information about bone metabolism in both the preoperative and postoperative monitoring phases.

SC-GAN: Structure-completion generative adversarial network for synthetic CT generation from MR images with truncated anatomy

Creating synthetic CT (sCT) from magnetic resonance (MR) images enables MR-based treatment planning in radiation therapy. However, the MR images used for MR-guided adaptive planning are often truncated in the boundary regions due to the limited field of view and the need for sequence optimization. Consequently, the sCT generated from these truncated MR images lacks complete anatomic information, leading to dose calculation error for MR-based adaptive planning. We propose a novel structure-completion generative adversarial network (SC-GAN) to generate sCT with full anatomic details from the truncated MR images. To enable anatomy compensation, we expand input channels of the CT generator by including a body mask and introduce a truncation loss between sCT and real CT. The body mask for each patient was automatically created from the simulation CT scans and transformed to daily MR images by rigid registration as another input for our SC-GAN in addition to the MR images. The truncation loss was constructed by implementing either an auto-segmentor or an edge detector to penalize the difference in body outlines between sCT and real CT. The experimental results show that our SC-GAN achieved much improved accuracy of sCT generation in both truncated and untruncated regions compared to the original cycleGAN and conditional GAN methods.

"sCT-Feasibility" - a feasibility study for deep learning-based MRI-only brain radiotherapy

BACKGROUND

Radiotherapy (RT) is an important treatment modality for patients with brain malignancies. Traditionally, computed tomography (CT) images are used for RT treatment planning whereas magnetic resonance imaging (MRI) images are used for tumor delineation. Therefore, MRI and CT need to be registered, which is an error prone process. The purpose of this clinical study is to investigate the clinical feasibility of a deep learning-based MRI-only workflow for brain radiotherapy, that eliminates the registration uncertainty through calculation of a synthetic CT (sCT) from MRI data.

METHODS

A total of 54 patients with an indication for radiation treatment of the brain and stereotactic mask immobilization will be recruited. All study patients will receive standard therapy and imaging including both CT and MRI. All patients will receive dedicated RT-MRI scans in treatment position. An sCT will be reconstructed from an acquired MRI DIXON-sequence using a commercially available deep learning solution on which subsequent radiotherapy planning will be performed. Through multiple quality assurance (QA) measures and reviews during the course of the study, the feasibility of an MRI-only workflow and comparative parameters between sCT and standard CT workflow will be investigated holistically. These QA measures include feasibility and quality of image guidance (IGRT) at the linear accelerator using sCT derived digitally reconstructed radiographs in addition to potential dosimetric deviations between the CT and sCT plan. The aim of this clinical study is to establish a brain MRI-only workflow as well as to identify risks and QA mechanisms to ensure a safe integration of deep learning-based sCT into radiotherapy planning and delivery.

DISCUSSION

Compared to CT, MRI offers a superior soft tissue contrast without additional radiation dose to the patients. However, up to now, even though the dosimetrical equivalence of CT and sCT has been shown in several retrospective studies, MRI-only workflows have still not been widely adopted. The present study aims to determine feasibility and safety of deep learning-based MRI-only radiotherapy in a holistic manner incorporating the whole radiotherapy workflow.

TRIAL REGISTRATION

NCT06106997.

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.

Contouring practices and artefact management within a synthetic CT-based radiotherapy workflow for the central nervous system

BACKGROUND

The incorporation of magnetic resonance (MR) imaging in radiotherapy (RT) workflows improves contouring precision, yet it introduces geometrical uncertainties when registered with computed tomography (CT) scans. Synthetic CT (sCT) images could minimize these uncertainties and streamline the RT workflow. This study aims to compare the contouring capabilities of sCT images with conventional CT-based/MR-assisted RT workflows, with an emphasis on managing artefacts caused by surgical fixation devices (SFDs).

METHODS

The study comprised a commissioning cohort of 100 patients with cranial tumors treated using a conventional CT-based/MR-assisted RT workflow and a validation cohort of 30 patients with grade IV glioblastomas treated using an MR-only workflow. A CE-marked artificial-intelligence-based sCT product was utilized. The delineation accuracy comparison was performed using dice similarity coefficient (DSC) and average Hausdorff distance (AHD). Artefacts within the commissioning cohort were visually inspected, classified and an estimation of thickness was derived using Hausdorff distance (HD). For the validation cohort, boolean operators were used to extract artefact volumes adjacent to the target and contrasted to the planning treatment volume.

RESULTS

The combination of high DSC (0.94) and low AHD (0.04 mm) indicates equal target delineation capacity between sCT images and conventional CT scans. However, the results for organs at risk delineation were less consistent, likely because of voxel size differences between sCT images and CT scans and absence of standardized delineation routines. Artefacts observed in sCT images appeared as enhancements of cranial bone. When close to the target, they could affect its definition. Therefore, in the validation cohort the clinical target volume (CTV) was expanded towards the bone by 3.5 mm, as estimated by HD analysis. Subsequent analysis on cone-beam CT scans showed that the CTV adjustment was enough to provide acceptable target coverage.

CONCLUSION

The tested sCT product performed on par with conventional CT in terms of contouring capability. Additionally, this study provides both the first comprehensive classification of metal artefacts on a sCT product and a novel method to assess the clinical impact of artefacts caused by SFDs on target delineation. This methodology encourages similar analysis for other sCT products.

The use of PET/MRI in radiotherapy

Positron emission tomography/magnetic resonance imaging (PET/MRI) is a hybrid imaging technique that quantitatively combines the metabolic and functional data from positron emission tomography (PET) with anatomical and physiological information from MRI. As PET/MRI technology has advanced, its applications in cancer care have expanded. Recent studies have demonstrated that PET/MRI provides unique advantages in the field of radiotherapy and has become invaluable in guiding precision radiotherapy techniques. This review discusses the rationale and clinical evidence supporting the use of PET/MRI for radiation positioning, target delineation, efficacy evaluation, and patient surveillance.Critical relevance statement This article critically assesses the transformative role of PET/MRI in advancing precision radiotherapy, providing essential insights into improved radiation positioning, target delineation, efficacy evaluation, and patient surveillance in clinical radiology practice.Key points• The emergence of PET/MRI will be a key bridge for precise radiotherapy.• PET/MRI has unique advantages in the whole process of radiotherapy.• New tracers and nanoparticle probes will broaden the use of PET/MRI in radiation.• PET/MRI will be utilized more frequently for radiotherapy.

Evaluating a radiotherapy deep learning synthetic CT algorithm for PET-MR attenuation correction in the pelvis

BACKGROUND

Positron emission tomography-magnetic resonance (PET-MR) attenuation correction is challenging because the MR signal does not represent tissue density and conventional MR sequences cannot image bone. A novel zero echo time (ZTE) MR sequence has been previously developed which generates signal from cortical bone with images acquired in 65 s. This has been combined with a deep learning model to generate a synthetic computed tomography (sCT) for MR-only radiotherapy. This study aimed to evaluate this algorithm for PET-MR attenuation correction in the pelvis.

METHODS

Ten patients being treated with ano-rectal radiotherapy received a [Formula: see text]F-FDG-PET-MR in the radiotherapy position. Attenuation maps were generated from ZTE-based sCT (sCTAC) and the standard vendor-supplied MRAC. The radiotherapy planning CT scan was rigidly registered and cropped to generate a gold standard attenuation map (CTAC). PET images were reconstructed using each attenuation map and compared for standard uptake value (SUV) measurement, automatic thresholded gross tumour volume (GTV) delineation and GTV metabolic parameter measurement. The last was assessed for clinical equivalence to CTAC using two one-sided paired t tests with a significance level corrected for multiple testing of [Formula: see text]. Equivalence margins of [Formula: see text] were used.

RESULTS

Mean whole-image SUV differences were -0.02% (sCTAC) compared to -3.0% (MRAC), with larger differences in the bone regions (-0.5% to -16.3%). There was no difference in thresholded GTVs, with Dice similarity coefficients [Formula: see text]. However, there were larger differences in GTV metabolic parameters. Mean differences to CTAC in [Formula: see text] were [Formula: see text] (± standard error, sCTAC) and [Formula: see text] (MRAC), and [Formula: see text] (sCTAC) and [Formula: see text] (MRAC) in [Formula: see text]. The sCTAC was statistically equivalent to CTAC within a [Formula: see text] equivalence margin for [Formula: see text] and [Formula: see text] ([Formula: see text] and [Formula: see text]), whereas the MRAC was not ([Formula: see text] and [Formula: see text]).

CONCLUSION

Attenuation correction using this radiotherapy ZTE-based sCT algorithm was substantially more accurate than current MRAC methods with only a 40 s increase in MR acquisition time. This did not impact tumour delineation but did significantly improve the accuracy of whole-image and tumour SUV measurements, which were clinically equivalent to CTAC. This suggests PET images reconstructed with sCTAC would enable accurate quantitative PET images to be acquired on a PET-MR scanner.

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.

CNN-based multi-modal radiomics analysis of pseudo-CT utilization in MRI-only brain stereotactic radiotherapy: a feasibility study

BACKGROUND

Pseudo-computed tomography (pCT) quality is a crucial issue in magnetic resonance image (MRI)-only brain stereotactic radiotherapy (SRT), so this study systematically evaluated it from the multi-modal radiomics perspective.

METHODS

34 cases (< 30 cm³) were retrospectively included (2021.9-2022.10). For each case, both CT and MRI scans were performed at simulation, and pCT was generated by a convolutional neural network (CNN) from planning MRI. Conformal arc or volumetric modulated arc technique was used to optimize the dose distribution. The SRT dose was compared between pCT and planning CT with dose volume histogram (DVH) metrics and gamma index. Wilcoxon test and Spearman analysis were used to identify key factors associated with dose deviations. Additionally, original image features were extracted for radiomic analysis. Tumor control probability (TCP) and normal tissue complication probability (NTCP) were employed for efficacy evaluation.

RESULTS

There was no significant difference between pCT and planning CT except for radiomics. The mean value of Hounsfield unit of the planning CT was slightly higher than that of pCT. The Gadolinium-based agents in planning MRI could increase DVH metrics deviation slightly. The median local gamma passing rates (1%/1 mm) between planning CTs and pCTs (non-contrast) was 92.6% (range 63.5-99.6%). Also, differences were observed in more than 85% of original radiomic features. The mean absolute deviation in TCP was 0.03%, and the NTCP difference was below 0.02%, except for the normal brain, which had a 0.16% difference. In addition, the number of SRT fractions and lesions, and lesion morphology could influence dose deviation.

CONCLUSIONS

This is the first multi-modal radiomics analysis of CNN-based pCT from planning MRI for SRT of small brain lesions, covering dosiomics and radiomics. The findings suggest the potential of pCT in SRT plan design and efficacy prediction, but caution needs to be taken for radiomic analysis.

The contest between internal and external-beam dosimetry: The Zeno's paradox of Achilles and the tortoise

Radionuclide therapy, also called molecular radiotherapy (MRT), has come of age, with several novel radiopharmaceuticals being approved for clinical use or under development in the last decade. External beam radiotherapy (EBRT) is a well-established treatment modality, with about half of all oncologic patients expected to receive at least one external radiation treatment over their disease course. The efficacy and the toxicity of both types of treatment rely on the interaction of radiation with biological tissues. Dosimetry played a fundamental role in the scientific and technological evolution of EBRT, and absorbed doses to the target and to the organs at risk are calculated on a routine basis. In contrast, in MRT the usefulness of internal dosimetry has long been questioned, and a structured path to include absorbed dose calculation is missing. However, following a similar route of development as EBRT, MRT treatments could probably be optimized in a significant proportion of patients, likely based on dosimetry and radiobiology. In the present paper we describe the differences and the similarities between internal and external-beam dosimetry in the context of radiation treatments, and we retrace the main stages of their development over the last decades.

Clinical feasibility of MRI-based synthetic CT imaging in the diagnosis of lumbar disc herniation: a comparative study

BACKGROUND

Computed tomography (CT) and magnetic resonance imaging (MRI) are indicated for use in preoperative planning and may complicate diagnosis and place a burden on patients with lumbar disc herniation.

PURPOSE

To investigate the diagnostic potential of MRI-based synthetic CT with conventional CT in the diagnosis of lumbar disc herniation.

MATERIAL AND METHODS

After obtaining prior institutional review board approval, 19 patients who underwent conventional and synthetic CT imaging were enrolled in this prospective study. Synthetic CT images were generated from the MRI data using U-net. The two sets of images were compared and analyzed qualitatively by two musculoskeletal radiologists. The images were rated on a 4-point scale to determine their subjective quality. The agreement between the conventional and synthetic images for a diagnosis of lumbar disc herniation was determined independently using the kappa statistic. The diagnostic performances of conventional and synthetic CT images were evaluated for sensitivity, specificity, and accuracy, and the consensual results based on T2-weighted imaging were employed as the reference standard.

RESULTS

The inter-reader and intra-reader agreement were almost moderate for all evaluated modalities (κ = 0.57-0.79 and 0.47-0.75, respectively). The sensitivity, specificity, and accuracy for detecting lumbar disc herniation were similar for synthetic and conventional CT images (synthetic vs. conventional, reader 1: sensitivity = 91% vs. 81%, specificity = 83% vs. 100%, accuracy = 87% vs. 91%; P < 0.001; reader 2: sensitivity = 84% vs. 81%, specificity = 85% vs. 98%, accuracy = 84% vs. 90%; P < 0.001).

CONCLUSION

Synthetic CT images can be used in the diagnostics of lumbar disc herniation.

A multi-centric evaluation of self-learning GAN based pseudo-CT generation software for low field pelvic magnetic resonance imaging

Purpose/objectives

An artificial intelligence-based pseudo-CT from low-field MR images is proposed and clinically evaluated to unlock the full potential of MRI-guided adaptive radiotherapy for pelvic cancer care.

Materials and method

In collaboration with TheraPanacea (TheraPanacea, Paris, France) a pseudo-CT AI-model was generated using end-to-end ensembled self-supervised GANs endowed with cycle consistency using data from 350 pairs of weakly aligned data of pelvis planning CTs and TrueFisp-(0.35T)MRIs. The image accuracy of the generated pCT were evaluated using a retrospective cohort involving 20 test cases coming from eight different institutions (US: 2, EU: 5, AS: 1) and different CT vendors. Reconstruction performance was assessed using the organs at risk used for treatment. Concerning the dosimetric evaluation, twenty-nine prostate cancer patients treated on the low field MR-Linac (ViewRay) at Montpellier Cancer Institute were selected. Planning CTs were non-rigidly registered to the MRIs for each patient. Treatment plans were optimized on the planning CT with a clinical TPS fulfilling all clinical criteria and recalculated on the warped CT (wCT) and the pCT. Three different algorithms were used: AAA, AcurosXB and MonteCarlo. Dose distributions were compared using the global gamma passing rates and dose metrics.

Results

The observed average scaled (between maximum and minimum HU values of the CT) difference between the pCT and the planning CT was 33.20 with significant discrepancies across organs. Femoral heads were the most reliably reconstructed (4.51 and 4.77) while anal canal and rectum were the less precise ones (63.08 and 53.13). Mean gamma passing rates for 1%1mm, 2%/2mm, and 3%/3mm tolerance criteria and 10% threshold were greater than 96%, 99% and 99%, respectively, regardless the algorithm used. Dose metrics analysis showed a good agreement between the pCT and the wCT. The mean relative difference were within 1% for the target volumes (CTV and PTV) and 2% for the OARs.

Conclusion

This study demonstrated the feasibility of generating clinically acceptable an artificial intelligence-based pseudo CT for low field MR in pelvis with consistent image accuracy and dosimetric results.

Comprehensive evaluation of similarity between synthetic and real CT images for nasopharyngeal carcinoma

BACKGROUND

Although magnetic resonance imaging (MRI)-to-computed tomography (CT) synthesis studies based on deep learning have significantly progressed, the similarity between synthetic CT (sCT) and real CT (rCT) has only been evaluated in image quality metrics (IQMs). To evaluate the similarity between synthetic CT (sCT) and real CT (rCT) comprehensively, we comprehensively evaluated IQMs and radiomic features for the first time.

METHODS

This study enrolled 127 patients with nasopharyngeal carcinoma who underwent CT and MRI scans. Supervised-learning (Unet) and unsupervised-learning (CycleGAN) methods were applied to build MRI-to-CT synthesis models. The regions of interest (ROIs) included nasopharynx gross tumor volume (GTVnx), brainstem, parotid glands, and temporal lobes. The peak signal-to-noise ratio (PSNR), mean absolute error (MAE), root mean square error (RMSE), and structural similarity (SSIM) were used to evaluate image quality. Additionally, 837 radiomic features were extracted for each ROI, and the correlation was evaluated using the concordance correlation coefficient (CCC).

RESULTS

The MAE, RMSE, SSIM, and PSNR of the body were 91.99, 187.12, 0.97, and 51.15 for Unet and 108.30, 211.63, 0.96, and 49.84 for CycleGAN. For the metrics, Unet was superior to CycleGAN (P < 0.05). For the radiomic features, the percentage of four levels (i.e., excellent, good, moderate, and poor, respectively) were as follows: GTVnx, 8.5%, 14.6%, 26.5%, and 50.4% for Unet and 12.3%, 25%, 38.4%, and 24.4% for CycleGAN; other ROIs, 5.44% ± 3.27%, 5.56% ± 2.92%, 21.38% ± 6.91%, and 67.58% ± 8.96% for Unet and 5.16% ± 1.69%, 3.5% ± 1.52%, 12.68% ± 7.51%, and 78.62% ± 8.57% for CycleGAN.

CONCLUSIONS

Unet-sCT was superior to CycleGAN-sCT for the IQMs. However, neither exhibited absolute superiority in radiomic features, and both were far less similar to rCT. Therefore, further work is required to improve the radiomic similarity for MRI-to-CT synthesis.

TRIAL REGISTRATION

This study was a retrospective study, so it was free from registration.

MRI-based synthetic CT in the detection of knee osteoarthritis: Comparison with CT

Magnetic resonance Imaging is the gold standard for assessment of soft tissues; however, X-ray-based techniques are required for evaluating bone-related pathologies. This study evaluated the performance of synthetic computed tomography (sCT), a novel MRI-based bone visualization technique, compared with CT, for the scoring of knee osteoarthritis. sCT images were generated from the 3T T1-weighted gradient-echo MR images using a trained machine learning algorithm. Two readers scored the severity of osteoarthritis in tibiofemoral and patellofemoral joints according to OACT, which enables the evaluation of osteoarthritis, from its characteristics of joint space narrowing, osteophytes, cysts and sclerosis in CT (and sCT) images. Cohen's κ was used to assess the interreader agreement for each modality, and intermodality agreement of CT- and sCT-based scores for each reader. We also compared the confidence level of readers for grading CT and sCT images using confidence scores collected during grading. Inter-reader agreement for tibiofemoral and patellofemoral joints were almost-perfect for both modalities (κ = 0.83-0.88). The intermodality agreement of osteoarthritis scores between CT and sCT was substantial to almost-perfect for tibiofemoral (κ = 0.63 and 0.84 for the two readers) and patellofemoral joints (κ = 0.78 and 0.81 for the two readers). The analysis of diagnosis confidence scores showed comparable visual quality of the two modalities, where both are showing acceptable confidence levels for scoring OA. In conclusion, in this single-center study, sCT and CT were comparable for the scoring of knee OA.

Development and implementation of optimized endogenous contrast sequences for delineation in adaptive radiotherapy on a 1.5T MR-linear-accelerator: a prospective R-IDEAL stage 0-2a quantitative/qualitative evaluation of in vivo site-specific quality-assurance using a 3D T2 fat-suppressed platform for head and neck cancer

Purpose

To improve segmentation accuracy in head and neck cancer (HNC) radiotherapy treatment planning for the 1.5T hybrid magnetic resonance imaging/linear accelerator (MR-Linac), three-dimensional (3D), T2-weighted, fat-suppressed magnetic resonance imaging sequences were developed and optimized.

Approach

After initial testing, spectral attenuated inversion recovery (SPAIR) was chosen as the fat suppression technique. Five candidate SPAIR sequences and a nonsuppressed, T2-weighted sequence were acquired for five HNC patients using a 1.5T MR-Linac. MR physicists identified persistent artifacts in two of the SPAIR sequences, so the remaining three SPAIR sequences were further analyzed. The gross primary tumor volume, metastatic lymph nodes, parotid glands, and pterygoid muscles were delineated using five segmentors. A robust image quality analysis platform was developed to objectively score the SPAIR sequences on the basis of qualitative and quantitative metrics.

Results

Sequences were analyzed for the signal-to-noise ratio and the contrast-to-noise ratio and compared with fat and muscle, conspicuity, pairwise distance metrics, and segmentor assessments. In this analysis, the nonsuppressed sequence was inferior to each of the SPAIR sequences for the primary tumor, lymph nodes, and parotid glands, but it was superior for the pterygoid muscles. The SPAIR sequence that received the highest combined score among the analysis categories was recommended to Unity MR-Linac users for HNC radiotherapy treatment planning.

Conclusions

Our study led to two developments: an optimized, 3D, T2-weighted, fat-suppressed sequence that can be disseminated to Unity MR-Linac users and a robust image quality analysis pathway that can be used to objectively score SPAIR sequences and can be customized and generalized to any image quality optimization protocol. Improved segmentation accuracy with the proposed SPAIR sequence will potentially lead to improved treatment outcomes and reduced toxicity for patients by maximizing the target coverage and minimizing the radiation exposure of organs at risk.

Geometric distortions in clinical MRI sequences for radiotherapy: insights gained from a multicenter investigation

BACKGROUND

As magnetic resonance imaging (MRI) becomes increasingly integrated into radiotherapy (RT) for enhanced treatment planning and adaptation, the inherent geometric distortion in acquired MR images pose a potential challenge to treatment accuracy. This study aimed to evaluate the geometric distortion levels in the clinical MRI protocols used across Danish RT centers and discuss influence of specific sequence parameters. Based on the variety in geometric performance across centers, we assess if harmonization of MRI sequences is a relevant measure.

MATERIALS AND METHODS

Nine centers participated with 12 MRI scanners and MRI-Linacs (MRL). Using a travelling phantom approach, a reference MRI sequence was used to assess variation in baseline distortion level between scanners. The phantom was also scanned with local clinical MRI sequences for brain, head/neck (H/N), abdomen, and pelvis. The influence of echo time, receiver bandwidth, image weighting, and 2D/3D acquisition was investigated.

RESULTS

We found a large variation in geometric accuracy across 93 clinical sequences examined, exceeding the baseline variation found between MRI scanners (σ = 0.22 mm), except for abdominal sequences where the variation was lower. Brain and abdominal sequences showed lowest distortion levels ([0.22, 2.26] mm), and a large variation in performance was found for H/N and pelvic sequences ([0.19, 4.07] mm). Post hoc analyses revealed that distortion levels decreased with increasing bandwidth and a less clear increase in distortion levels with increasing echo time. 3D MRI sequences had lower distortion levels than 2D (median of 1.10 and 2.10 mm, respectively), and in DWI sequences, the echo-planar imaging read-out resulted in highest distortion levels.

CONCLUSION

There is a large variation in the geometric distortion levels of clinical MRI sequences across Danish RT centers, and between anatomical sites. The large variation observed makes harmonization of MRI sequences across institutions and adoption of practices from well-performing anatomical sites, a relevant measure within RT.

Region of interest focused MRI to synthetic CT translation using regression and segmentation multi-task network

Objective. In MR-only clinical workflow, replacing CT with MR image is of advantage for workflow efficiency and reduces radiation to the patient. An important step required to eliminate CT scan from the workflow is to generate the information provided by CT via an MR image. In this work, we aim to demonstrate a method to generate accurate synthetic CT (sCT) from an MR image to suit the radiation therapy (RT) treatment planning workflow. We show the feasibility of the method and make way for a broader clinical evaluation.Approach. We present a machine learning method for sCT generation from zero-echo-time (ZTE) MRI aimed at structural and quantitative accuracies of the image, with a particular focus on the accurate bone density value prediction. The misestimation of bone density in the radiation path could lead to unintended dose delivery to the target volume and results in suboptimal treatment outcome. We propose a loss function that favors a spatially sparse bone region in the image. We harness the ability of the multi-task network to produce correlated outputs as a framework to enable localization of region of interest (RoI) via segmentation, emphasize regression of values within RoI and still retain the overall accuracy via global regression. The network is optimized by a composite loss function that combines a dedicated loss from each task.Main results. We have included 54 brain patient images in this study and tested the sCT images against reference CT on a subset of 20 cases. A pilot dose evaluation was performed on 9 of the 20 test cases to demonstrate the viability of the generated sCT in RT planning. The average quantitative metrics produced by the proposed method over the test set were-(a) mean absolute error (MAE) of 70 ± 8.6 HU; (b) peak signal-to-noise ratio (PSNR) of 29.4 ± 2.8 dB; structural similarity metric (SSIM) of 0.95 ± 0.02; and (d) Dice coefficient of the body region of 0.984 ± 0.Significance. We demonstrate that the proposed method generates sCT images that resemble visual characteristics of a real CT image and has a quantitative accuracy that suits RT dose planning application. We compare the dose calculation from the proposed sCT and the real CT in a radiation therapy treatment planning setup and show that sCT based planning falls within 0.5% target dose error. The method presented here with an initial dose evaluation makes an encouraging precursor to a broader clinical evaluation of sCT based RT planning on different anatomical regions.

Comparison of four synthetic CT generators for brain and prostate MR-only workflow in radiotherapy

BACKGROUND

The interest in MR-only workflows is growing with the introduction of artificial intelligence in the synthetic CT generators converting MR images into CT images. The aim of this study was to evaluate several commercially available sCT generators for two anatomical localizations.

METHODS

Four sCT generators were evaluated: one based on the bulk density method and three based on deep learning methods. The comparison was performed on large patient cohorts (brain: 42 patients and pelvis: 52 patients). It included geometric accuracy with the evaluation of Hounsfield Units (HU) mean error (ME) for several structures like the body, bones and soft tissues. Dose evaluation included metrics like the Dmean ME for bone structures (skull or femoral heads), PTV and soft tissues (brain or bladder or rectum). A 1%/1 mm gamma analysis was also performed.

RESULTS

HU ME in the body were similar to those reported in the literature. Dmean ME were smaller than 2% for all structures. Mean gamma pass rate down to 78% were observed for the bulk density method in the brain. Performances of the bulk density generator were generally worse than the artificial intelligence generators for the brain but similar for the pelvis. None of the generators performed best in all the metrics studied.

CONCLUSIONS

All four generators can be used in clinical practice to implement a MR-only workflow but the bulk density method clearly performed worst in the brain.

Clinical feasibility of a commercially available MRI-only method for radiotherapy treatment planning of the brain

BACKGROUND

Advancements in deep-learning based synthetic computed tomography (sCT) image conversion methods have enabled the development of magnetic resonance imaging (MRI)-only based radiotherapy treatment planning (RTP) of the brain.

PURPOSE

This study evaluates the clinical feasibility of a commercial, deep-learning based MRI-only RTP method with respect to dose calculation and patient positioning verification performance in RTP of the brain.

METHODS

Clinical validation of dose calculation accuracy was performed by a retrospective evaluation for 25 glioma and 25 brain metastasis patients. Dosimetric and image quality of the studied MRI-only RTP method was evaluated by a direct comparison of the sCT-based and computed tomography (CT)-based external beam radiation therapy (EBRT) images and treatment plans. Patient positioning verification accuracy of sCT images was evaluated retrospectively for 10 glioma and 10 brain metastasis patients based on clinical cone-beam computed tomography (CBCT) imaging.

RESULTS

An average mean dose difference of Dmean = 0.1% for planning target volume (PTV) and 0.6% for normal tissue (NT) structures were obtained for glioma patients. Respective results for brain metastasis patients were Dmean = 0.5% for PTVs and Dmean =1.0% for NTs. Global three-dimensional (3D) gamma pass rates using 2%/2 mm dose difference and distance-to-agreement (DTA) criterion were 98.0% for the glioma subgroup, and 95.2% for the brain metastasis subgroup using 1%/1 mm criterion. Mean distance differences of <1.0 mm were observed in all Cartesian directions between CT-based and sCT-based CBCT patient positioning in both subgroups.

CONCLUSIONS

In terms of dose calculation and patient positioning accuracy, the studied MRI-only method demonstrated its clinical feasibility for RTP of the brain. The results encourage the use of the studied method as part of a routine clinical workflow.

Clinical validation of MR imaging time reduction for substitute/synthetic CT generation for prostate MRI-only treatment planning

Radiotherapy treatment planning based only on magnetic resonance imaging (MRI) has become clinically achievable. Though computed tomography (CT) is the gold standard for radiotherapy imaging, directly providing the electron density values needed for planning calculations, MRI has superior soft tissue visualisation to guide treatment planning decisions and optimisation. MRI-only planning removes the need for the CT scan, but requires generation of a substitute/synthetic/pseudo CT (sCT) for electron density information. Shortening the MRI imaging time would improve patient comfort and reduce the likelihood of motion artefacts. A volunteer study was previously carried out to investigate and optimise faster MRI sequences for a hybrid atlas-voxel conversion to sCT for prostate treatment planning. The aim of this follow-on study was to clinically validate the performance of the new optimised sequence for sCT generation in a treated MRI-only prostate patient cohort. 10 patients undergoing MRI-only treatment were scanned on a Siemens Skyra 3T MRI as part of the MRI-only sub-study of the NINJA clinical trial (ACTRN12618001806257). Two sequences were used, the standard 3D T2-weighted SPACE sequence used for sCT conversion which has been previously validated against CT, and a modified fast SPACE sequence, selected based on the volunteer study. Both were used to generate sCT scans. These were then compared to evaluate the fast sequence conversion for anatomical and dosimetric accuracy against the clinically approved treatment plans. The average Mean Absolute Error (MAE) for the body was 14.98 ± 2.35 HU, and for bone was 40.77 ± 5.51 HU. The external volume contour comparison produced a Dice Similarity Coefficient (DSC) of at least 0.976, and an average of 0.985 ± 0.004, and the bony anatomy contour comparison a DSC of at least 0.907, and an average of 0.950 ± 0.018. The fast SPACE sCT agreed with the gold standard sCT within an isocentre dose of -0.28% ± 0.16% and an average gamma pass rate of 99.66% ± 0.41% for a 1%/1 mm gamma tolerance. In this clinical validation study, the fast sequence, which reduced the required imaging time by approximately a factor of 4, produced an sCT with similar clinical dosimetric results compared to the standard sCT, demonstrating its potential for clinical use for treatment planning.

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.

Exploring contrast generalisation in deep learning-based brain MRI-to-CT synthesis

BACKGROUND

Synthetic computed tomography (sCT) has been proposed and increasingly clinically adopted to enable magnetic resonance imaging (MRI)-based radiotherapy. Deep learning (DL) has recently demonstrated the ability to generate accurate sCT from fixed MRI acquisitions. However, MRI protocols may change over time or differ between centres resulting in low-quality sCT due to poor model generalisation.

PURPOSE

investigating domain randomisation (DR) to increase the generalisation of a DL model for brain sCT generation.

METHODS

CT and corresponding T1-weighted MRI with/without contrast, T2-weighted, and FLAIR MRI from 95 patients undergoing RT were collected, considering FLAIR the unseen sequence where to investigate generalisation. A "Baseline" generative adversarial network was trained with/without the FLAIR sequence to test how a model performs without DR. Image similarity and accuracy of sCT-based dose plans were assessed against CT to select the best-performing DR approach against the Baseline.

RESULTS

The Baseline model had the poorest performance on FLAIR, with mean absolute error (MAE) = 106 ± 20.7 HU (mean ±σ). Performance on FLAIR significantly improved for the DR model with MAE = 99.0 ± 14.9 HU, but still inferior to the performance of the Baseline+FLAIR model (MAE = 72.6 ± 10.1 HU). Similarly, an improvement in γ-pass rate was obtained for DR vs Baseline.

CONCLUSION

DR improved image similarity and dose accuracy on the unseen sequence compared to training only on acquired MRI. DR makes the model more robust, reducing the need for re-training when applying a model on sequences unseen and unavailable for retraining.

Synthetic computed tomography based dose calculation in prostate cancer patients with hip prostheses for magnetic resonance imaging-only radiotherapy

Background and purpose

Metallic hip prostheses cause substantial artefacts in both computed tomography (CT) and magnetic resonance (MR) images used in radiotherapy treatment planning (RTP) for prostate cancer patients. The aim of this study was to evaluate the dose calculation accuracy of a synthetic CT (sCT) generation workflow and the improvement in implant visibility using metal artefact reduction sequences.

Materials and methods

The study included 23 patients with prostate cancer who had hip prostheses, of which 10 patients had bilateral hip implants. An in-house protocol was applied to create sCT images for dose calculation comparison. The study compared prostheses volumes and resulting avoidance sectors against planning target volume (PTV) dose uniformity and organs at risk (OAR) sparing.

Results

Median PTV dose difference between sCT and CT-based dose calculation among all patients was 0.1 % (-0.4 to 0.4%) (median(range)). Bladder and rectum differences (V_50Gy) were 0.2 % (-0.3 to 1.1%) and 0.1 % (-0.9 to 0.5%). The median 3D local gamma pass rate for partial arc cases using a Dixon MR sequence was Γ_20%^2mm/2% = 99.9%. For the bilateral full arc cases, using a metal artefact reconstruction sequence, the pass rate was Γ_20%^2mm/2% = 99.0%.

Conclusions

An in-house protocol for generating sCT images for dose calculation provided clinically feasible dose calculation accuracy for prostate cancer patients with hip implants. PTV median dose difference for uni- and bilateral patients with avoidance sectors remained <0.4%. The Outphase images enhanced implant visibility resulting in smaller avoidance sectors, better OAR sparing, and improved PTV uniformity.

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.

Synthetic computed tomography for low-field magnetic resonance-only radiotherapy in head-and-neck cancer using residual vision transformers

Background and purpose

Synthetic computed tomography (sCT) scans are necessary for dose calculation in magnetic resonance (MR)-only radiotherapy. While deep learning (DL) has shown remarkable performance in generating sCT scans from MR images, research has predominantly focused on high-field MR images. This study presents the first implementation of a DL model for sCT generation in head-and-neck (HN) cancer using low-field MR images. Specifically, the use of vision transformers (ViTs) was explored.

Materials and methods

The dataset consisted of 31 patients, resulting in 196 pairs of deformably-registered computed tomography (dCT) and MR scans. The latter were obtained using a balanced steady-state precession sequence on a 0.35T scanner. Residual ViTs were trained on 2D axial, sagittal, and coronal slices, respectively, and the final sCTs were generated by averaging the models' outputs. Different image similarity metrics, dose volume histogram (DVH) deviations, and gamma analyses were computed on the test set (n = 6). The overlap between auto-contours on sCT scans and manual contours on MR images was evaluated for different organs-at-risk using the Dice score.

Results

The median [range] value of the test mean absolute error was 57 [37-74] HU. DVH deviations were below 1% for all structures. The median gamma passing rates exceeded 94% in the 2%/2mm analysis (threshold = 90%). The median Dice scores were above 0.7 for all organs-at-risk.

Conclusions

The clinical applicability of DL-based sCT generation from low-field MR images in HN cancer was proved. High sCT-dCT similarity and dose metric accuracy were achieved, and sCT suitability for organs-at-risk auto-delineation was shown.

Association between Body Composition Contents and Hepatic Fibrosis in Sarcopenic Obesity

It is well established that sarcopenic obesity (SO) is linked to many diseases such as metabolic and non-alcoholic fatty liver diseases, but there is little known about the relationship between SO and hepatic fibrosis progression in chronic liver disease. This study compared body composition contents in patients with non-obesity (NOb) and SO using abdominal magnetic resonance imaging and investigated the relationship between hepatic fibrosis and SO factors. This retrospective study enrolled 60 patients (28 NOb; 32 SO) from June 2014 to December 2020. Patients underwent histopathologic investigation where they classified fibrosis stages based on the Meta-analysis of Histological Data in Viral Hepatitis fibrosis scoring system. Muscle and fat areas at the third lumber vertebra level were assessed. The variation in the areas of muscle (MA), subcutaneous adipose tissue (SAT), and visceral adipose tissue (VAT) among fibrosis stages, and associations between hepatic fibrosis and SO factors, were analyzed. There were significant differences in SAT and VAT (p < 0.001), whereas there was no difference in MA (p = 0.064). There were significant differences in MA/SAT (p = 0.009), MA/VAT (p < 0.001), and MA/(SAT+VAT) (p < 0.001). In all the patients, hepatic fibrosis positively correlated with serum aspartate aminotransferase level (AST, R = 0.324; p = 0.025). Especially in SO patients, hepatic fibrosis closely correlated with body mass index (BMI, R = 0.443; p = 0.011), AST (R = 0.415; p = 0.044), VAT (R = 0.653; p < 0.001), MA/VAT (R = -0.605; p < 0.001), and MA/(SAT+VAT) (R = -0.416; p = 0.018). However, there was no association in NOb patients. This study demonstrated that SO patients had larger SAT and VAT than NOb patients. Hepatic fibrosis in SO positively correlated with body visceral fat composition in combination with BMI and AST level. These findings will be useful for understanding the relationship between the hepatic manifestation of fibrosis and body fat composition in sarcopenia and SO.

Automated deep learning auto-segmentation of air volumes for MRI-guided online adaptive radiation therapy of abdominal tumors

Objective. In the current MR-Linac online adaptive workflow, air regions on the MR images need to be manually delineated for abdominal targets, and then overridden by air density for dose calculation. Auto-delineation of these regions is desirable for speed purposes, but poses a challenge, since unlike computed tomography, they do not occupy all dark regions on the image. The purpose of this study is to develop an automated method to segment the air regions on MRI-guided adaptive radiation therapy (MRgART) of abdominal tumors.Approach. A modified ResUNet3D deep learning (DL)-based auto air delineation model was trained using 102 patients' MR images. The MR images were acquired by a dedicated in-house sequence named 'Air-Scan', which is designed to generate air regions that are especially dark and accentuated. The air volumes generated by the newly developed DL model were compared with the manual air contours using geometric similarity (Dice Similarity Coefficient (DSC)), and dosimetric equivalence using Gamma index and dose-volume parameters.Main results. The average DSC agreement between the DL generated and manual air contours is 99% ± 1%. The gamma index between the dose calculations with overriding the DL versus manual air volumes with density of 0.01 is 97% ± 2% for a local gamma calculation with a tolerance of 2% and 2 mm. The dosimetric parameters from planning target volume-PTV and organs at risk-OARs were all within 1% between when DL versus manual contours were overridden by air density. The model runs in less than five seconds on a PC with 28 Core processor and NVIDIA Quadro®P2000 GPU.Significance: a DL based automated segmentation method was developed to generate air volumes on specialized abdominal MR images and generate results that are practically equivalent to the manual contouring of air volumes.

Comprehensive dose evaluation of a Deep Learning based synthetic Computed Tomography algorithm for pelvic Magnetic Resonance-only radiotherapy

BACKGROUND AND PURPOSE

Magnetic Resonance (MR)-only radiotherapy enables the use of MR without the uncertainty of MR-Computed Tomography (CT) registration. This requires a synthetic CT (sCT) for dose calculations, which can be facilitated by a novel Zero Echo Time (ZTE) sequence where bones are visible and images are acquired in 65 seconds. This study evaluated the dose calculation accuracy for pelvic sites of a ZTE-based Deep Learning sCT algorithm developed by GE Healthcare.

MATERIALS AND METHODS

ZTE and CT images were acquired in 56 pelvic radiotherapy patients in the radiotherapy position. A 2D U-net convolutional neural network was trained using pairs of deformably registered CT and ZTE images from 36 patients. In the remaining 20 patients the dosimetric accuracy of the sCT was assessed using cylindrical dummy Planning Target Volumes (PTVs) positioned at four different central axial locations, as well as the clinical treatment plans (for prostate (n=10), rectum (n=4) and anus (n=6) cancers). The sCT was rigidly and deformably registered, the plan recalculated and the doses compared using mean differences and gamma analysis.

RESULTS

Mean dose differences to the PTV D98% were ≤ 0.5% for all dummy PTVs and clinical plans (rigid registration). Mean gamma pass rates at 1%/1 mm were 98.0 ± 0.4% (rigid) and 100.0 ± 0.0% (deformable), 96.5 ± 0.8% and 99.8 ± 0.1%, and 95.4 ± 0.6% and 99.4 ± 0.4% for the clinical prostate, rectum and anus plans respectively.

CONCLUSIONS

A ZTE-based sCT algorithm with high dose accuracy throughout the pelvis has been developed. This suggests the algorithm is sufficiently accurate for MR-only radiotherapy for all pelvic sites.

Quality assurance and temporal stability of a 1.5 T MRI scanner for MR-guided Photon and Particle Therapy

PURPOSE

To describe performance measurements, adaptations and time stability over 20 months of a diagnostic MR scanner for integration into MR-guided photon and particle radiotherapy.

MATERIAL AND METHODS

For realization of MR-guided photon and particle therapy (MRgRT/MRgPT), a 1.5 T MR scanner was installed at the Heidelberg Ion Beam Therapy Center. To integrate MRI into the treatment process, a flat tabletop and dedicated coil holders for flex coils were used, which prevent deformation of the patient external contour and allow for the use of immobilization tools for reproducible positioning. The signal-to-noise ratio (SNR) was compared for the diagnostic and therapy-specific setup using the flat couch top and flexible coils for the a) head & neck and b) abdominal region as well as for different bandwidths and clinical pulse sequences. Additionally, a quality assurance (QA) protocol with monthly measurements of the ACR phantom and measurement of geometric distortions for a large field-of-view (FOV) was implemented to assess the imaging quality parameters of the device over the course of 20 months.

RESULTS

The SNR measurements showed a decreased SNR for the RT-specific as compared to the diagnostic setup of (a) 26% to 34% and (b) 11% to 33%. No significant bandwidth dependency for this ratio was found. The longitudinal assessment of the image quality parameters with the ACR and distortion phantom confirmed the long-term stability of the MRI device.

CONCLUSION

A diagnostic MRI was commissioned for use in MR-guided particle therapy. Using a radiotherapy specific setup, a high geometric accuracy and signal homogeneity was obtained after some adaptions and the measured parameters were shown to be stable over a period of 20 months.

CT synthesis from multi-sequence MRI using adaptive fusion network

OBJECTIVE

To investigate a method using multi-sequence magnetic resonance imaging (MRI) to synthesize computed tomography (CT) for MRI-only radiation therapy.

APPROACH

We proposed an adaptive multi-sequence fusion network (AMSF-Net) to exploit both voxel- and context-wise cross-sequence correlations from multiple MRI sequences to synthesize CT using element- and patch-wise fusions, respectively. The element- and patch-wise fusion feature spaces were combined, and the most representative features were selected for modeling. Finally, a densely connected convolutional decoder was applied to utilize the selected features to produce synthetic CT images.

MAIN RESULTS

This study includes a total number of 90 patients' T1-weighted MRI, T2-weighted MRI and CT data. The AMSF-Net reduced the average mean absolute error (MAE) from 52.88-57.23 to 49.15 HU, increased the peak signal-to-noise ratio (PSNR) from 24.82-25.32 to 25.63 dB, increased the structural similarity index measure (SSIM) from 0.857-0.869 to 0.878, and increased the dice coefficient of bone from 0.886-0.896 to 0.903 compared to the other three existing multi-sequence learning models. The improvements were statistically significant according to two-tailed paired t-test. In addition, AMSF-Net reduced the intensity difference with real CT in five organs at risk, four types of normal tissue and tumor compared with the baseline models. The MAE decreases in parotid and spinal cord were over 8% and 16% with reference to the mean intensity value of the corresponding organ, respectively. Further, the qualitative evaluations confirmed that AMSF-Net exhibited superior structural image quality of synthesized bone and small organs such as the eye lens.

SIGNIFICANCE

The proposed method can improve the intensity and structural image quality of synthetic CT and has potential for use in clinical applications.

Evaluation of MRI-only based online adaptive radiotherapy of abdominal region on MR-linac

PURPOSE

A hybrid magnetic resonance linear accelerator (MRL) can perform magnetic resonance imaging (MRI) with high soft-tissue contrast to be used for online adaptive radiotherapy (oART). To obtain electron densities needed for the oART dose calculation, a computed tomography (CT) is often deformably registered to MRI. Our aim was to evaluate an MRI-only based synthetic CT (sCT) generation as an alternative to the deformed CT (dCT)-based oART in the abdominal region.

METHODS

The study data consisted of 57 patients who were treated on a 0.35 T MRL system mainly for abdominal tumors. Simulation MRI-CT pairs of 43 patients were used for training and validation of a prototype convolutional neural network sCT-generation algorithm, based on HighRes3DNet, for the abdominal region. For remaining test patients, sCT images were produced from simulation MRIs and daily MRIs. The dCT-based plans were re-calculated on sCT with identical calculation parameters. The sCT and dCT were compared in terms of geometric agreement and calculated dose.

RESULTS

The mean and one standard deviation of the geometric agreement metrics over dCT-sCT-pairs were: mean error of 8 ± 10 HU, mean absolute error of 49 ± 10 HU, and Dice similarity coefficient of 55 ± 12%, 60 ± 5%, and 82 ± 15% for bone, fat, and lung tissues, respectively. The dose differences between the sCT and dCT-based dose for planning target volumes were 0.5 ± 0.9%, 0.6 ± 0.8%, and 0.5 ± 0.8% at D_2% , D_50% , and D_98% in physical dose and 0.8 ± 1.4%, 0.8 ± 1.2%, and 0.6 ± 1.1% in biologically effective dose (BED). For organs-at-risk, the dose differences of all evaluated dose-volume histogram points were within [-4.5%, 7.8%] and [-1.1 Gy, 3.5 Gy] in both physical dose and BED.

CONCLUSIONS

The geometric agreement metrics were within typically reported values and most average relative dose differences were within 1%. Thus, an MRI-only sCT-based approach is a promising alternative to the current clinical practice of the abdominal oART on MRL.

Rapid Morphological Measurement Method of Aortic Dissection Stent Based on Spatial Observation Point Set

OBJECTIVES

Post-operative stent morphology of aortic dissection patients is important for performing clinical diagnosis and prognostic assessment. However, stent morphologies still need to be manually measured, which is a process prone to errors, high time consumption and difficulty in exploiting inter-data associations. Herein, we propose a method based on the stepwise combination of basic, non-divisible data sets to quickly obtain morphological parameters with high accuracy.

METHODS

We performed the 3D reconstruction of 109 post-operative follow-up CT image data from 26 patients using mimics software. By extracting the spatial locations of the basic morphological observation points on the stent, we defined a basic and non-reducible set of observation points. Further, we implemented a fully automatic stent segmentation and an observation point extraction algorithm. We analyzed the stability and accuracy of the algorithms on a test set containing 8 cases and 408 points. Based on this dataset, we calculated three morphological parameters of different complexity for the different spatial structural features exhibited by the stent. Finally, we compared the two measurement schemes in four aspects: data variability, data stability, statistical process complexity and algorithmic error.

RESULTS

The statistical results of the two methods on two low-complexity morphological parameters (spatial position of stent end and vascular stent end-slip volume) show good agreement (n = 26, P₁, P₂ < 0.001, r₁ = 0.992, r₂ = 0.988). The statistics of the proposed method for the morphological parameters of medium complexity (proximal support ring feature diameter and distal support ring feature diameter) avoid the errors caused by manual extraction, and the magnitude of this correction to the traditional method does not exceed 4 mm with an average correction of 1.38 mm. Meanwhile, our proposed automatic observation point extraction method has only 2.2% error rate on the test set, and the average spatial distance from the manually marked observation points is 0.73 mm. Thus, the proposed method is able to rapidly and accurately measure the stent circumferential deflection angle, which is highly complex and cannot be measured using traditional methods.

CONCLUSIONS

The proposed method can significantly reduce the statistical observation time and information processing cost compared to the traditional morphological observation methods. Moreover, when new morphological parameters are required, one can quickly and accurately obtain the target parameters by new "combinatorial functions." Iterative modification of the data set itself is avoided.

Investigation of a water equivalent depth method for dosimetric accuracy evaluation of synthetic CT

PURPOSE

To provide a metric that reflects the dosimetric utility of the synthetic CT (sCT) and can be rapidly determined.

METHODS

Retrospective CT and atlas-based sCT of 62 (53 IMRT and 9 VMAT) prostate cancer patients were used. For image similarity measurements, the sCT and reference CT (rCT) were aligned using clinical registration parameters. Conventional image similarity metrics including the mean absolute error (MAE) and mean error (ME) were calculated. The water equivalent depth (WED) was automatically determined for each patient on the rCT and sCT as the distance from the skin surface to the treatment plan isocentre at 36 equidistant gantry angles, and the mean WED difference (ΔWED¯) between the two scans was calculated. Doses were calculated on each scan pair for the clinical plan in the treatment planning system. The image similarity measurements and ΔWED¯ were then compared to the isocentre dose difference (ΔD_iso) between the two scans.

RESULTS

While no particular relationship to dose was observed for the other image similarity metrics, the ME results showed a linear trend against ΔD_iso with R² = 0.6, and the 95 % prediction interval for ΔD_iso between -1.2 and 1 %. The ΔWED¯ results showed an improved linear trend (R² = 0.8) with a narrower 95 % prediction interval from -0.8 % to 0.8 %.

CONCLUSION

ΔWED¯ highly correlates with ΔD_iso for the reference and synthetic CT scans. This is easy to calculate automatically and does not require time-consuming dose calculations. Therefore, it can facilitate the process of developing and evaluating new sCT generation algorithms.

Limitations of phase-sorting based pencil beam scanned 4D proton dose calculations under irregular motion

Objective.4D dose calculation (4DDC) for pencil beam scanned (PBS) proton therapy is typically based on phase-sorting of individual pencil beams onto phases of a single breathing cycle 4DCT. Understanding the dosimetric limitations and uncertainties of this approach is essential, especially for the realistic treatment scenario with irregular free breathing motion.Approach.For three liver and three lung cancer patient CTs, the deformable multi-cycle motion from 4DMRIs was used to generate six synthetic 4DCT(MRI)s, providing irregular motion (11/15 cycles for liver/lung; tumor amplitudes ∼4-18 mm). 4DDCs for two-field plans were performed, with the temporal resolution of the pencil beam delivery (4-200 ms) or with 8 phases per breathing cycle (500-1000 ms). For the phase-sorting approach, the tumor center motion was used to determine the phase assignment of each spot. The dose was calculated either using the full free breathing motion or individually repeating each single cycle. Additionally, the use of an irregular surrogate signal prior to 4DDC on a repeated cycle was simulated. The CTV volume with absolute dose differences >5% (V_dosediff>5%) and differences in CTVV_95%andD_5%-D_95%compared to the free breathing scenario were evaluated.Main results.Compared to 4DDC considering the full free breathing motion with finer spot-wise temporal resolution, 4DDC based on a repeated single 4DCT resulted inV_dosediff>5%of on average 34%, which resulted in an overestimation ofV_95%up to 24%. However, surrogate based phase-sorting prior to 4DDC on a single cycle 4DCT, reduced the averageV_dosediff>5%to 16% (overestimationV_95%up to 19%). The 4DDC results were greatly influenced by the choice of reference cycle (V_dosediff>5%up to 55%) and differences due to temporal resolution were much smaller (V_dosediff>5%up to 10%).Significance.It is important to properly consider motion irregularity in 4D dosimetric evaluations of PBS proton treatments, as 4DDC based on a single 4DCT can lead to an underestimation of motion effects.

Deep-learning-based generation of synthetic 6-minute MRI from 2-minute MRI for use in head and neck cancer radiotherapy

Background

Quick magnetic resonance imaging (MRI) scans with low contrast-to-noise ratio are typically acquired for daily MRI-guided radiotherapy setup. However, for patients with head and neck (HN) cancer, these images are often insufficient for discriminating target volumes and organs at risk (OARs). In this study, we investigated a deep learning (DL) approach to generate high-quality synthetic images from low-quality images.

Methods

We used 108 unique HN image sets of paired 2-minute T2-weighted scans (2mMRI) and 6-minute T2-weighted scans (6mMRI). 90 image sets (~20,000 slices) were used to train a 2-dimensional generative adversarial DL model that utilized 2mMRI as input and 6mMRI as output. Eighteen image sets were used to test model performance. Similarity metrics, including the mean squared error (MSE), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR) were calculated between normalized synthetic 6mMRI and ground-truth 6mMRI for all test cases. In addition, a previously trained OAR DL auto-segmentation model was used to segment the right parotid gland, left parotid gland, and mandible on all test case images. Dice similarity coefficients (DSC) were calculated between 2mMRI and either ground-truth 6mMRI or synthetic 6mMRI for each OAR; two one-sided t-tests were applied between the ground-truth and synthetic 6mMRI to determine equivalence. Finally, a visual Turing test using paired ground-truth and synthetic 6mMRI was performed using three clinician observers; the percentage of images that were correctly identified was compared to random chance using proportion equivalence tests.

Results

The median similarity metrics across the whole images were 0.19, 0.93, and 33.14 for MSE, SSIM, and PSNR, respectively. The median of DSCs comparing ground-truth vs. synthetic 6mMRI auto-segmented OARs were 0.86 vs. 0.85, 0.84 vs. 0.84, and 0.82 vs. 0.85 for the right parotid gland, left parotid gland, and mandible, respectively (equivalence p<0.05 for all OARs). The percent of images correctly identified was equivalent to chance (p<0.05 for all observers).

Conclusions

Using 2mMRI inputs, we demonstrate that DL-generated synthetic 6mMRI outputs have high similarity to ground-truth 6mMRI, but further improvements can be made. Our study facilitates the clinical incorporation of synthetic MRI in MRI-guided radiotherapy.

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.

Quality assurance for MRI-only radiation therapy: A voxel-wise population-based methodology for image and dose assessment of synthetic CT generation methods

The quality assurance of synthetic CT (sCT) is crucial for safe clinical transfer to an MRI-only radiotherapy planning workflow. The aim of this work is to propose a population-based process assessing local errors in the generation of sCTs and their impact on dose distribution. For the analysis to be anatomically meaningful, a customized interpatient registration method brought the population data to the same coordinate system. Then, the voxel-based process was applied on two sCT generation methods: a bulk-density method and a generative adversarial network. The CT and MRI pairs of 39 patients treated by radiotherapy for prostate cancer were used for sCT generation, and 26 of them with delineated structures were selected for analysis. Voxel-wise errors in sCT compared to CT were assessed for image intensities and dose calculation, and a population-based statistical test was applied to identify the regions where discrepancies were significant. The cumulative histograms of the mean absolute dose error per volume of tissue were computed to give a quantitative indication of the error for each generation method. Accurate interpatient registration was achieved, with mean Dice scores higher than 0.91 for all organs. The proposed method produces three-dimensional maps that precisely show the location of the major discrepancies for both sCT generation methods, highlighting the heterogeneity of image and dose errors for sCT generation methods from MRI across the pelvic anatomy. Hence, this method provides additional information that will assist with both sCT development and quality control for MRI-based planning radiotherapy.