This white paper examines the potential of pioneering technologies and artificial intelligence-driven solutions in advancing clinical trials involving radiation therapy. As the field of radiation therapy evolves, the integration of cutting-edge approaches such as radiopharmaceutical dosimetry, FLASH radiation therapy, image guided radiation therapy, and artificial intelligence promises to improve treatment planning, patient care, and outcomes. Additionally, recent advancements in quantum science, linear energy transfer/relative biological effect, and the combination of radiation therapy and immunotherapy create new avenues for innovation in clinical trials. The paper aims to provide an overview of these emerging technologies and discuss their challenges and opportunities in shaping the future of radiation oncology clinical trials. By synthesizing knowledge from experts across various disciplines, this white paper aims to present a foundation for the successful integration of these innovations into radiation therapy research and practice, ultimately enhancing patient outcomes and revolutionizing cancer care.

Radiotherapy treatment for prostate cancer relies on computed tomography (CT) and/or magnetic resonance imaging (MRI) for segmentation of target volumes and organs at risk (OARs). Manual segmentation of these volumes is regarded as the gold standard for ground truth in machine learning applications, but to acquire such data is tedious and time-consuming. A publicly available clinical dataset is presented, comprising MRI- and synthetic CT (sCT) images, target and OARs segmentations, and radiotherapy dose distributions for 432 prostate cancer patients treated with MRI-guided radiotherapy. An extended dataset with 35 patients is also included, with the addition of deep learning (DL)-generated segmentations, DL segmentation uncertainty maps, and DL segmentations manually adjusted by four radiation oncologists. The publication of these resources aims to aid research in automated radiotherapy treatment planning, segmentation, inter-observer analyses, and DL model uncertainty investigation. The dataset is hosted on the AIDA Data Hub and offers a free-to-use resource for the scientific community, valuable for the advancement of medical imaging and prostate cancer radiotherapy research.

The use of artificial intelligence (AI) in radiotherapy (RT) is expanding rapidly. However, there exists a notable lack of clinician trust in AI models, underscoring the need for effective uncertainty quantification (UQ) methods. The purpose of this study was to scope existing literature related to UQ in RT, identify areas of improvement, and determine future directions.

We followed the PRISMA-ScR scoping review reporting guidelines. We utilized the population (human cancer patients), concept (utilization of AI UQ), context (radiotherapy applications) framework to structure our search and screening process. We conducted a systematic search spanning seven databases, supplemented by manual curation, up to January 2024. Our search yielded a total of 8980 articles for initial review. Manuscript screening and data extraction was performed in Covidence. Data extraction categories included general study characteristics, RT characteristics, AI characteristics, and UQ characteristics.

We identified 56 articles published from 2015 to 2024. 10 domains of RT applications were represented; most studies evaluated auto-contouring (50 %), followed by image-synthesis (13 %), and multiple applications simultaneously (11 %). 12 disease sites were represented, with head and neck cancer being the most common disease site independent of application space (32 %). Imaging data was used in 91 % of studies, while only 13 % incorporated RT dose information. Most studies focused on failure detection as the main application of UQ (60 %), with Monte Carlo dropout being the most commonly implemented UQ method (32 %) followed by ensembling (16 %). 55 % of studies did not share code or datasets.

Our review revealed a lack of diversity in UQ for RT applications beyond auto-contouring. Moreover, we identified a clear need to study additional UQ methods, such as conformal prediction. Our results may incentivize the development of guidelines for reporting and implementation of UQ in RT.

Magnetic resonance imaging (MRI) is increasingly used in treatment preparation of ocular proton therapy, but its spatial accuracy might be limited by geometric distortions due to susceptibility artefacts. A correct geometry of the MR images is paramount since it defines where the dose will be delivered. In this study, we assessed the geometrical accuracy of ocular MRI.

A dedicated ocular 3 T MRI protocol, with localized shimming and increased gradients, was compared to computed tomography (CT) and X-ray images in a phantom and in 15 uveal melanoma patients. The MRI protocol contained three-dimensional T2-weighted and T1-weighted sequences with an isotropic reconstruction resolution of 0.3-0.4 mm. Tantalum clips were identified by three observers and clip-clip distances were compared between T2-weighted and T1-weighted MRI, CT and X-ray images for the phantom and between MRI and X-ray images for the patients.

Interobserver variability was below 0.35 mm for the phantom and 0.30(T1)/0.61(T2) mm in patients. Mean absolute differences between MRI and reference were below 0.27 ± 0.16 mm and 0.32 ± 0.23 mm for the phantom and in patients, respectively. In patients, clip-clip distances were slightly larger on MRI than on X-ray images (mean difference T1: 0.11 ± 0.38 mm, T2: 0.10 ± 0.44 mm). Differences did not increase at larger distances and did not correlate to interobserver variability.

A dedicated ocular MRI protocol can produce images of the eye with a geometrical accuracy below half the MRI acquisition voxel (<0.4 mm). Therefore, these images can be used for ocular proton therapy planning, both in the current model-based workflow and in proposed three-dimensional MR-based workflows.

The use of artificial intelligence (AI) in radiotherapy (RT) is expanding rapidly. However, there exists a notable lack of clinician trust in AI models, underscoring the need for effective uncertainty quantification (UQ) methods. The purpose of this study was to scope existing literature related to UQ in RT, identify areas of improvement, and determine future directions.

We followed the PRISMA-ScR scoping review reporting guidelines. We utilized the population (human cancer patients), concept (utilization of AI UQ), context (radiotherapy applications) framework to structure our search and screening process. We conducted a systematic search spanning seven databases, supplemented by manual curation, up to January 2024. Our search yielded a total of 8980 articles for initial review. Manuscript screening and data extraction was performed in Covidence. Data extraction categories included general study characteristics, RT characteristics, AI characteristics, and UQ characteristics.

We identified 56 articles published from 2015-2024. 10 domains of RT applications were represented; most studies evaluated auto-contouring (50%), followed by image-synthesis (13%), and multiple applications simultaneously (11%). 12 disease sites were represented, with head and neck cancer being the most common disease site independent of application space (32%). Imaging data was used in 91% of studies, while only 13% incorporated RT dose information. Most studies focused on failure detection as the main application of UQ (60%), with Monte Carlo dropout being the most commonly implemented UQ method (32%) followed by ensembling (16%). 55% of studies did not share code or datasets.

Our review revealed a lack of diversity in UQ for RT applications beyond auto-contouring. Moreover, there was a clear need to study additional UQ methods, such as conformal prediction. Our results may incentivize the development of guidelines for reporting and implementation of UQ in RT.

Recent advances in MRI-guided radiation therapy (MRgRT) and deep learning techniques encourage fully adaptive radiation therapy (ART), real-time MRI monitoring, and the MRI-only treatment planning workflow. Given the rapid growth and emergence of new state-of-the-art methods in these fields, we systematically review 197 studies written on or before December 31, 2022, and categorize the studies into the areas of image segmentation, image synthesis, radiomics, and real time MRI. Building from the underlying deep learning methods, we discuss their clinical importance and current challenges in facilitating small tumor segmentation, accurate x-ray attenuation information from MRI, tumor characterization and prognosis, and tumor motion tracking. In particular, we highlight the recent trends in deep learning such as the emergence of multi-modal, visual transformer, and diffusion models.

Fiducial markers are commonly used in navigation assisted minimally invasive spine surgery and they help transfer image coordinates into real world coordinates. In practice, these markers might be located outside the field-of-view (FOV) of C-arm cone-beam computed tomography (CBCT) systems used in intraoperative surgeries, due to the limited detector sizes. As a consequence, reconstructed markers in CBCT volumes suffer from artifacts and have distorted shapes, which sets an obstacle for navigation.

In this work, we propose two fiducial marker detection methods: direct detection from distorted markers (direct method) and detection after marker recovery (recovery method). For direct detection from distorted markers in reconstructed volumes, an efficient automatic marker detection method using two neural networks and a conventional circle detection algorithm is proposed. For marker recovery, a task-specific data preparation strategy is proposed to recover markers from severely truncated data. Afterwards, a conventional marker detection algorithm is applied for position detection. The networks in both methods are trained based on simulated data. For the direct method, 6800 images and 10000 images are generated respectively to train the U-Net and ResNet50. For the recovery method, the training set includes 1360 images for FBPConvNet and Pix2pixGAN. The simulated data set with 166 markers and 4 cadaver cases with real fiducials are used for evaluation.

The two methods are evaluated on simulated data and real cadaver data. The direct method achieves 100% detection rates within 1 mm detection error on simulated data with normal truncation and simulated data with heavier noise, but only detect 94.6% markers in extremely severe truncation case. The recovery method detects all the markers successfully in three test data sets and around 95% markers are detected within 0.5 mm error. For real cadaver data, both methods achieve 100% marker detection rates with mean registration error below 0.2 mm.

Our experiments demonstrate that the direct method is capable of detecting distorted markers accurately and the recovery method with the task-specific data preparation strategy has high robustness and generalizability on various data sets. The task-specific data preparation is able to reconstruct structures of interest outside the FOV from severely truncated data better than conventional data preparation. This article is protected by copyright. All rights reserved.