Cone-beam computed tomography (CBCT) is widely used for dento-maxillofacial diagnostics and treatment planning, and comprehensive multi-structure segmentation remains time-consuming, limiting large-scale, reproducible research. In this article, we present ToothFairy2, a MICCAI 2024 challenge on multi-structure segmentation in maxillofacial CBCT. The accompanying dataset comprises 530 CBCT volumes (480 public training, 50 hidden test) with expert 3D annotations of 42 classes, including maxilla, mandible, crowns, bridges, implants, inferior alveolar canals, maxillary sinuses, pharynx, and teeth labeled according to the International Tooth Numbering System (FDI). 26 international teams participated in ToothFairy2, and their methods were run and evaluated for voxel-wise multi-class segmentation using a standardized protocol. This report extends the evaluation of teeth to also investigate the current capabilities of tooth detection and FDI numbering. Furthermore, ranking stability was analyzed to assess the robustness of the final challenge outcome. Overall, challenge participants achieved consistently high performance for large, high-contrast structures such as jawbones, pharynx, and most teeth, while maxillary sinuses, dental restorations, and fine structures remain challenging due to class imbalance and metal artifacts. Analysis of tooth-related metrics further revealed that assigning correct FDI numbers was more challenging than delineating individual teeth. By releasing CBCT data, 3D annotations, baseline models, and evaluation code, ToothFairy2 establishes a long-term benchmark to drive the development of automated methods for robust, clinically meaningful multi-structure segmentation in maxillofacial CBCT.

The segmentation of pelvic fracture fragments in CT and X-ray images is crucial for trauma diagnosis, surgical planning, and intraoperative guidance. However, accurately and efficiently delineating the bone fragments remains a significant challenge due to complex anatomy and imaging limitations. The PENGWIN challenge, organized as a MICCAI 2024 satellite event, aimed to advance automated fracture segmentation by benchmarking state-of-the-art algorithms on these complex tasks. A diverse dataset of 150 CT scans was collected from multiple clinical centers, and a large set of simulated X-ray images was generated using the DeepDRR method. Final submissions from 16 teams worldwide were evaluated under a rigorous multi-metric testing scheme. The top-performing CT algorithm achieved an average fragment-wise intersection over union (IoU) of 0.930, demonstrating satisfactory accuracy. However, in the X-ray task, the best algorithm achieved an IoU of 0.774, which is promising but not yet sufficient for intra-operative decision-making, reflecting the inherent challenges of fragment overlap in projection imaging. Beyond the quantitative evaluation, the challenge revealed methodological diversity in algorithm design. Variations in instance representation, such as primary-secondary classification versus boundary-core separation, led to differing segmentation strategies. Despite promising results, the challenge also exposed inherent uncertainties in fragment definition, particularly in cases of incomplete fractures. These findings suggest that interactive segmentation approaches, integrating human decision-making with task-relevant information, may be essential for improving model reliability and clinical applicability.

Limited-angle cone-beam computed tomography (LA-CBCT) enables rapid imaging and reduced radiation exposure, but its severely incomplete projection data lead to ill-posed reconstructions with prominent artifacts, limiting clinical applicability. Recent advances in 3D Gaussian Splatting (3D-GS) have shown promise for efficient tomographic reconstruction, yet its performance remains highly sensitive to initialization. In this work, we present SPARK (Structurally-Informed Projection-Accelerated Reconstruction), a two-stage framework that introduces a generative, structurally informed initialization for 3D-GS. In the first stage, a geometry-conditioned network directly predicts complete 3D Gaussian parameters from a sparse subset of projections, embedding learned anatomical priors to mitigate artifact propagation. In the second stage, the generated scene is refined through physics-based 3D-GS optimization, yielding high-fidelity reconstructions consistent with measured projections. Extensive experiments on public datasets demonstrate that SPARK substantially improves both image quality and convergence speed, achieving superior PSNR/SSIM in severely limited-angle scenarios compared with analytical, iterative, and deep learning baselines. Moreover, SPARK reconstructions provide enhanced inputs for downstream post-processing networks, further boosting image fidelity. These results suggest that SPARK is a promising prior-informed 3D-GS framework for simulated LA-CBCT reconstruction under limited angular coverage, providing an effective bridge between data-driven anatomical priors and physics-based projection-domain refinement.

During the process of computed tomography (CT), metallic implants often cause disruptive artifacts in the reconstructed images, impeding accurate diagnosis. Many supervised deep learning-based approaches have been proposed for metal artifact reduction (MAR). However, these methods heavily rely on training with paired simulated data, which are challenging to acquire. This limitation can lead to decreased performance when applying these methods in clinical practice. Existing unsupervised MAR methods, whether based on learning or not, typically work within a single domain, either in the image domain or the sinogram domain. In this paper, we propose an unsupervised MAR method based on the diffusion model, a generative model with a high capacity to represent data distributions. Specifically, we first train a diffusion model using CT images without metal artifacts. Subsequently, we iteratively introduce the diffusion priors in both the sinogram domain and image domain to restore the degraded portions caused by metal artifacts. Besides, we design temporally dynamic weight masks for the image-domian fusion. The dual-domain processing empowers our approach to outperform existing unsupervised MAR methods, including another MAR method based on diffusion model. The effectiveness has been qualitatively and quantitatively validated on synthetic datasets. Moreover, our method demonstrates superior visual results among both supervised and unsupervised methods on clinical datasets. Codes are available in github.com/DeepXuan/DuDoDp-MAR.

A novel method is proposed to obtain four-dimensional (4D) cone-beam computed tomography (CBCT) images from a routine scan in patients with upper abdominal cancer. The projections are sorted according to the location of the lung diaphragm before being reconstructed to phase-sorted data. A multiscale-discriminator generative adversarial network (MSD-GAN) is proposed to alleviate the severe streaking artifacts in the original images. The MSD-GAN is trained using simulated CBCT datasets from patient planning CT images. The enhanced images are further used to estimate the deformable vector field (DVF) among breathing phases using a deformable image registration method. The estimated DVF is then applied in the motion-compensated ordered-subset simultaneous algebraic reconstruction approach to generate 4D CBCT images. The proposed MSD-GAN is compared with U-Net on the performance of image enhancement. Results show that the proposed method significantly outperforms the total variation regularization-based iterative reconstruction approach and the method using only MSD-GAN to enhance original phase-sorted images in simulation and patient studies on 4D reconstruction quality. The MSD-GAN also shows higher accuracy than the U-Net. The proposed method enables a practical way for 4D-CBCT imaging from a single routine scan in upper abdominal cancer treatment including liver and pancreatic tumors.

Since the tumor moves with the patient's breathing movement in clinical surgery, the real-time prediction of respiratory movement is required to improve the efficacy of radiotherapy. Some RNN-based respiratory management methods have been proposed for this purpose. However, these existing RNN-based methods often suffer from the degradation of generalization performance for a long-term window (such as 600 ms) because of the structural consistency constraints. In this paper, we propose an innovative Long Short-term Transformer (LSTformer) for long-term real-time accurate respiratory prediction. Specifically, a novel Long-term Information Enhancement module (LIE) is proposed to solve the performance degradation under a long window by increasing the long-term memory of latent variables. A lightweight Transformer Encoder (LTE) is proposed to satisfy the real-time requirement via simplifying the architecture and limiting the number of layers. In addition, we propose an application-oriented data augmentation strategy to generalize our LSTformer to practical application scenarios, especially robotic radiotherapy. Extensive experiments on our augmented dataset and publicly available dataset demonstrate the state-of-the-art performance of our method on the premise of satisfying the real-time demand.

SUMMARY: Nowadays, it is feasible to collect massive features for quantitative representation and precision medicine, and thus, automatic ranking to figure out the most informative and discriminative ones becomes increasingly important. To address this issue, 42 feature ranking (FR) methods are integrated to form a MATLAB toolbox (matFR). The methods apply mutual information, statistical analysis, structure clustering and other principles to estimate the relative importance of features in specific measure spaces. Specifically, these methods are summarized, and an example shows how to apply a FR method to sort mammographic breast lesion features. The toolbox is easy to use and flexible to integrate additional methods. Importantly, it provides a tool to compare, investigate and interpret the features selected for various applications. AVAILABILITY AND IMPLEMENTATION: The toolbox is freely available at http://github.com/NicoYuCN/matFR. A tutorial and an example with a dataset are provided.

Sparse-view computed tomography (CT) holds great promise for speeding up data acquisition and reducing radiation dose in CT scans. Recent advances in reconstruction algorithms for sparse-view CT, such as iterative reconstruction algorithms, obtained high-quality image while requiring advanced computing power. Lately, deep learning (DL) has been widely used in various applications and has obtained many remarkable outcomes. In this paper, we propose a new method for sparse-view CT reconstruction based on the DL approach. The method can be divided into two steps. First, filter backprojection (FBP) was used to reconstruct the CT image from sparsely sampled sinogram. Then, the FBP results were fed to a DL neural network, which is a DenseNet and deconvolution-based network (DD-Net). The DD-Net combines the advantages of DenseNet and deconvolution and applies shortcut connections to concatenate DenseNet and deconvolution to accelerate the training speed of the network; all of those operations can greatly increase the depth of network while enhancing the expression ability of the network. After the training, the proposed DD-Net achieved a competitive performance relative to the state-of-the-art methods in terms of streaking artifacts removal and structure preservation. Compared with the other state-of-the-art reconstruction methods, the DD-Net method can increase the structure similarity by up to 18% and reduce the root mean square error by up to 42%. These results indicate that DD-Net has great potential for sparse-view CT image reconstruction.