There is a lack of consistency in conducting and reporting analyses from IOV studies. We suggest a framework to use for future studies evaluating IOV.

Summary Research in artificial intelligence for radiology and radiotherapy has recently become increasingly reliant on the use of deep learning‐based algorithms. While the performance of the models which these algorithms produce can significantly outperform more traditional machine learning methods, they do rely on larger datasets being available for training. To address this issue, data augmentation has become a popular method for increasing the size of a training dataset, particularly in fields where large datasets aren’t typically available, which is often the case when working with medical images. Data augmentation aims to generate additional data which is used to train the model and has been shown to improve performance when validated on a separate unseen dataset. This approach has become commonplace so to help understand the types of data augmentation techniques used in state‐of‐the‐art deep learning models, we conducted a systematic review of the literature where data augmentation was utilised on medical images (limited to CT and MRI) to train a deep learning model. Articles were categorised into basic, deformable, deep learning or other data augmentation techniques. As artificial intelligence models trained using augmented data make their way into the clinic, this review aims to give an insight to these techniques and confidence in the validity of the models produced.

The exquisite soft-tissue contrast of magnetic resonance imaging (MRI) has meant that the technique is having an increasing role in contouring the gross tumor volume (GTV) and organs at risk (OAR) in radiation therapy treatment planning systems (TPS). MRI-planning scans from diagnostic MRI scanners are currently incorporated into the planning process by being registered to CT data. The soft-tissue data from the MRI provides target outline guidance and the CT provides a solid geometric and electron density map for accurate dose calculation on the TPS computer. There is increasing interest in MRI machine placement in radiotherapy clinics as an adjunct to CT simulators. Most vendors now offer 70 cm bores with flat couch inserts and specialised RF coil designs. We would refer to these devices as MR-simulators. There is also research into the future application of MR-simulators independent of CT and as in-room image-guidance devices. It is within the background of this increased interest in the utility of MRI in radiotherapy treatment planning that this paper is couched. The paper outlines publications that deal with standard MRI sequences used in current clinical practice. It then discusses the potential for using processed functional diffusion maps (fDM) derived from diffusion weighted image sequences in tracking tumor activity and tumor recurrence. Next, this paper reviews publications that describe the use of MRI in patient-management applications that may, in turn, be relevant to radiotherapy treatment planning. The review briefly discusses the concepts behind functional techniques such as dynamic contrast enhanced (DCE), diffusion-weighted (DW) MRI sequences and magnetic resonance spectroscopic imaging (MRSI). Significant applications of MR are discussed in terms of the following treatment sites: brain, head and neck, breast, lung, prostate and cervix. While not yet routine, the use of apparent diffusion coefficient (ADC) map analysis indicates an exciting future application for functional MRI. Although DW-MRI has not yet been routinely used in boost adaptive techniques, it is being assessed in cohort studies for sub-volume boosting in prostate tumors.

Inter-observer variability in volume delineation can be reduced with the use of guidelines, provision of autocontours and teaching. The use of multimodality imaging is useful in certain tumour sites.

Distortion in magnetic resonance images needs to be taken into account for the purposes of radiotherapy treatment planning (RTP). A commercial MRI grid phantom was scanned on four different MRI scanners with multiple sequences to assess variations in the geometric distortion. The distortions present across the field of view were then determined. The effect of varying bandwidth on image distortion and signal to noise was also investigated. Distortion maps were created and these were compared to the location of patient anatomy within the scanner bore to estimate the magnitude and distribution of distortions located within specific clinical regions. Distortion magnitude and patterns varied between MRI sequence protocols and scanners. The magnitude of the distortions increased with increasing distance from the isocentre of the scanner within a 2D imaging plane. Average distortion across the phantom generally remained below 2.0 mm, although towards the edge of the phantom for a turbo spin echo sequence, the distortion increased to a maximum value of 4.1 mm. Application of correction algorithms supplied by each vendor reduced but did not completely remove distortions. Increasing the bandwidth of the acquisition sequence decreased the amount of distortion at the expense of a reduction in signal-to-noise ratio of 13.5 across measured bandwidths. Imaging protocol parameters including bandwidth, slice thickness and phase encoding direction, should be noted for distortion investigations in RTP since each can influence the distortion. The magnitude of distortion varies across different clinical sites.

This topical review provides an up-to-date overview of the theoretical and practical aspects of therapeutic kilovoltage x-ray beam dosimetry. Kilovoltage x-ray beams have the property that the maximum dose occurs very close to the surface and thus, they are predominantly used in the treatment of skin cancers but also have applications for the treatment of other cancers. In addition, kilovoltage x-ray beams are used in intra operative units, within animal irradiators and in on-board imagers on linear accelerators and kilovoltage dosimetry is important in these applications as well. This review covers both reference and relative dosimetry of kilovoltage x-ray beams and provides recommendations for clinical measurements based on the literature to date. In particular, practical aspects for the selection of dosimeter and phantom material are reviewed to provide suitable advice for medical physicists. An overview is also presented of dosimeters other than ionization chambers which can be used for both relative and in vivo dosimetry. Finally, issues related to the treatment planning and the use of Monte Carlo codes for solving radiation transport problems in kilovoltage x-ray beams are presented.

Inter-observer variability in anatomical contouring is the biggest contributor to uncertainty in radiation treatment planning. Contouring studies are frequently performed to investigate the differences between multiple contours on common datasets. There is, however, no widely accepted method for contour comparisons. The purpose of this study is to review the literature on contouring studies in the context of radiation oncology, with particular consideration of the contouring comparison methods they employ. A literature search, not limited by date, was conducted using Medline and Google Scholar with key words: contour, variation, delineation, inter/intra observer, uncertainty and trial dummy-run. This review includes a description of the contouring processes and contour comparison metrics used. The use of different processes and metrics according to tumour site and other factors were also investigated with limitations described. A total of 69 relevant studies were identified. The most common tumour sites were prostate (26), lung (10), head and neck cancers (8) and breast (7).The most common metric of comparison was volume used 59 times, followed by dimension and shape used 36 times, and centre of volume used 19 times. Of all 69 publications, 67 used a combination of metrics and two used only one metric for comparison. No clear relationships between tumour site or any other factors that may influence the contouring process and the metrics used to compare contours were observed from the literature. Further studies are needed to assess the advantages and disadvantages of each metric in various situations.

(2015). Continuous table acquisition MRI for radiotherapy treatment planning: Distortion assessment with a new extended 3D volumetric phantom. Medical Physics, 42 (4), 1982Physics, 42 (4), -1991 Continuous table acquisition MRI for radiotherapy treatment planning: distortion assessment with a new extended 3D volumetric phantom AbstractPurpose: Accurate geometry is required for radiotherapy treatment planning (RTP). When considering the use of magnetic resonance imaging (MRI) for RTP, geometric distortions observed in the acquired images should be considered. While scanner technology and vendor supplied correction algorithms provide some correction, large distortions are still present in images, even when considering considerably smaller scan lengths than those typically acquired with CT in conventional RTP. This study investigates MRI acquisition with a moving table compared with static scans for potential geometric benefits for RTP. Methods: A full field of view (FOV) phantom (diameter 500 mm; length 513 mm) was developed for measuring geometric distortions in MR images over volumes pertinent to RTP. The phantom consisted of layers of refined plastic within which vitamin E capsules were inserted. The phantom was scanned on CT to provide the geometric gold standard and on MRI, with differences in capsule location determining the distortion. MRI images were acquired with two techniques. For the first method, standard static table acquisitions were considered. Both 2D and 3D acquisition techniques were investigated. With the second technique, images were acquired with a moving table. The same sequence was acquired with a static table and then with table speeds of 1.1 mm/s and 2 mm/s. All of the MR images acquired were registered to the CT dataset using a deformable B-spline registration with the resulting deformation fields providing the distortion information for each acquisition. Results: MR images acquired with the moving table enabled imaging of the whole phantom length while images acquired with a static table were only able to image 50%-70% of the phantom length of 513 mm. Maximum distortion values were reduced across a larger volume when imaging with a moving table. Increased table speed resulted in a larger contribution of distortion from gradient nonlinearities in the through-plane direction and an increased blurring of capsule images, resulting in an apparent capsule volume increase by up to 170% in extreme axial FOV regions. Blurring increased with table speed and in the central regions of the phantom, geometric distortion was less for static table acquisitions compared to a table speed of 2 mm/s over the same volume. Overall, the best geometric accuracy was achieved with a table speed of 1.1 mm/s. Conclusions: The phantom designed enables full FOV imaging for distortion assessment for the purposes of RTP. MRI acquisition with a moving table extends the imaging volume in the z direction with reduced distortions which could be useful particularly if considering MR-only planning. If utilizing MR image...