Investigating the generalisation of an atlas-based synthetic-CT algorithm to another centre and MR scanner for prostate MR-only radiotherapy

An MRI sequence independent convolutional neural network for synthetic head CT generation in proton therapy

A magnetic resonance imaging (MRI) sequence independent deep learning technique was developed and validated to generate synthetic computed tomography (sCT) scans for MR guided proton therapy. 47 meningioma patients previously undergoing proton therapy based on pencil beam scanning were divided into training (33), validation (6), and test (8) cohorts. T₁, T₂, and contrast enhanced T₁ (T1CM) MRI sequences were used in combination with the planning CT (pCT) data to train a 3D U-Net architecture with ResNet-Blocks. A hyperparameter search was performed including two loss functions, two group sizes of normalisation, and depth of the network. Training outcome was compared between models trained for each individual MRI sequence and for all sequences combined. The performance was evaluated based on a metric and dosimetric analysis as well as spot difference maps. Furthermore, the influence of immobilisation masks that are not visible on MRIs was investigated. Based on the hyperparameter search, the final model was trained with fixed features per group for the group normalisation, six down-convolution steps, an input size of 128×192×192, and feature loss. For the test dataset for body/bone the mean absolute error (MAE) values were on average 79.8/216.3Houndsfield unit (HU) when trained using T1 images, 71.1/186.1HU for T2, and 82.9/236.4HU for T1CM. The structural similarity metric (SSIM) ranged from 0.95 to 0.98 for all sequences. The investigated dose parameters of the target structures agreed within 1% between original proton treatment plans and plans recalculated on sCTs. The spot difference maps had peaks at ±0.2cm and for 98% of all spots the difference was less than 1cm. A novel MRI sequence independent sCT generator was developed, which suggests that the training phase of neural networks can be disengaged from specific MRI acquisition protocols. In contrast to previous studies, the patient cohort consisted exclusively of actual proton therapy patients (i.e. "real-world data").

Deep learning based synthetic-CT generation in radiotherapy and PET: A review

Recently,deep learning (DL)-based methods for the generation of synthetic computed tomography (sCT) have received significant research attention as an alternative to classical ones. We present here a systematic review of these methods by grouping them into three categories, according to their clinical applications: (i) to replace computed tomography in magnetic resonance (MR) based treatment planning, (ii) facilitate cone-beam computed tomography based image-guided adaptive radiotherapy, and (iii) derive attenuation maps for the correction of positron emission tomography. Appropriate database searching was performed on journal articles published between January 2014 and December 2020. The DL methods' key characteristics were extracted from each eligible study, and a comprehensive comparison among network architectures and metrics was reported. A detailed review of each category was given, highlighting essential contributions, identifying specific challenges, and summarizing the achievements. Lastly, the statistics of all the cited works from various aspects were analyzed, revealing the popularity and future trends and the potential of DL-based sCT generation. The current status of DL-based sCT generation was evaluated, assessing the clinical readiness of the presented methods.

Are cone beam CT image matching skills transferrable from planning CT to planning MRI for MR-only prostate radiotherapy?

Objectives:

Treatment verification for MR-only planning has focused on fiducial marker matching, however, these are difficult to identify on MR. An alternative is using the MRI for soft-tissue matching with cone beam computed tomography images (MR-CBCT). However, therapeutic radiographers have limited experience of MRI. This study aimed to assess transferability of therapeutic radiographers CT-CBCT prostate image matching skills to MR-CBCT image matching.

Methods:

23 therapeutic radiographers with 3 months–5 years’ experience of online daily CT-CBCT soft-tissue matching prostate cancer patients participated. Each observer completed a baseline assessment of 10 CT-CBCT prostate soft-tissue image matches, followed by 10 MR-CBCT prostate soft-tissue image match assessment. A MRI anatomy training intervention was delivered and the 10 MR-CBCT prostate soft-tissue image match assessment was repeated. Limits of agreement were calculated as the disagreement of the observers with mean of all observers.

Results:

Limits of agreement at CT-CBCT baseline were 2.8 mm, 2.8 mm, 0.7 mm (vertical, longitudinal, lateral). MR-CBCT matches prior to training were 3.3 mm, 3.1 mm, 0.9 mm, and after training 2.6 mm, 2.4 mm, 1.1 mm (vertical, longitudinal, lateral). Results show similar limits of agreement across the assessments, and variation reduced following the training intervention.

Conclusion:

This suggests therapeutic radiographers’ prostate CBCT image matching skills are transferrable to a MRI planning scan, since MR-CBCT matching has comparable observer variation to CT-CBCT matching.

Advances in knowledge:

This is the first publication assessing interobserver MR-CBCT prostate soft tissue matching in an MR-only pathway.

A Systematic Review of the Clinical Implementation of Pelvic Magnetic Resonance Imaging-Only Planning for External Beam Radiation Therapy

The use of magnetic resonance (MR) imaging scans alone for radiation therapy treatment planning (MR-only planning) has been highlighted as one method of improving patient outcomes. Recent technologic advances have meant that introducing MR-only planning to the clinic is becoming a reality, with several specialist radiation therapy clinics using this technique for treatment. As such, substantial efforts are being made to introduce this technique into wide-spread clinical implementation. A systematic review of publications investigating the clinical implementation of pelvic MR-only radiation therapy treatment planning was undertaken following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The Medline, Embase, Scopus, Science Direct, Cumulative Index to Nursing and Allied Health Literature, and Web of Science databases were searched (timespan: all years to January 2, 2019). Twenty-six articles met the inclusion criteria. The studies were grouped into the following categories: (1) MR acquisition and synthetic computed tomography generation verification, (2) MR distortion quantification and phantom development, (3) clinical validation of patient treatment positioning in an MR-only workflow, and (4) MR-only commissioning processes. Key conclusions from this review are (1) MR-only planning has been implemented clinically for prostate cancer treatments; (2) a substantial amount of work remains to translate MR-only planning into widespread clinical implementation for all pelvic sites; (3) MR scanner distortions are no longer a barrier to MR-only planning, but they must be managed appropriately; (4) MR-only-based patient positioning verification shows promise, but limited evidence is reported in the literature and further investigation is required; and (5) a number of MR-only commissioning processes have been reported, which can aid centers as they undertake local commissioning; however, this needs to be formalized in guidance from national bodies.

MRI-only treatment planning: benefits and challenges

Over the past decade, the application of magnetic resonance imaging (MRI) has increased, and there is growing evidence to suggest that improvements in the accuracy of target delineation in MRI-guided radiation therapy may improve clinical outcomes in a variety of cancer types. However, some considerations should be recognized including patient motion during image acquisition and geometric accuracy of images. Moreover, MR-compatible immobilization devices need to be used when acquiring images in the treatment position while minimizing patient motion during the scan time. Finally, synthetic CT images (i.e. electron density maps) and digitally reconstructed radiograph images should be generated from MRI images for dose calculation and image guidance prior to treatment. A short review of the concepts and techniques that have been developed for implementation of MRI-only workflows in radiation therapy is provided in this document.