20 January 2025 in IEEE Transactions on Computational Imaging
Dual-energy computed tomography (DECT) offers quantitative insights and facilitates material decomposition, aiding in precise diagnosis and treatment planning. However, existing methods for material decomposition, often tailored to specific material types, need more generalizability and increase computational load with each additional material. We propose a CLIP-Driven Universal Model for adaptive Multi-Material Decomposition (MMD) to tackle this challenge. This model utilizes the semantic capabilities of text embeddings from Contrastive Language-Image Pre-training (CLIP), allowing a single network to manage structured feature embedding for multiple materials. A novel Siamese encoder and differential map fusion technique have also been integrated to enhance the decomposition accuracy while maintaining robustness across various conditions. Experiments on the simulated and physical patient studies have evidenced our model's superiority over traditional methods. Notably, it has significantly improved the Dice Similarity Coefficient—4.1%. These results underscore the potential of our network in clinical MMD applications, suggesting a promising avenue for enhancing DECT imaging analysis.
23 August 2024 in IEEE Transactions on Radiation and Plasma Medical Sciences
Four-dimensional Cone-beam Computed Tomography (4D CBCT) has recently been recognized as a proficient technique in mitigating motion artifacts attributed to respiratory organ movement. The primary challenges in 4D CBCT reconstruction encompass the precision in projection grouping, the efficacy in reconstructing from sparsely sampled data, and the accuracy in deformation field estimation. To surmount these challenges, we propose an innovative approach that integrates meticulous respiratory curve extraction for projection grouping and utilizes a diffusion model network with motion compensation (MoCo) techniques targeted at significantly enhancing image quality. An object detection network is employed to ascertain the exact position of the diaphragm, which is then normalized to formulate the respiratory curve. Further, we employ a U-Net architecture-based diffusion model, which integrates attention mechanisms to enhance sparse-view reconstruction and reduce artifacts through Guided-Diffusion. Deviating from conventional optical flow methods, our approach introduces an unsupervised registration network for deformation vector field (DVF) in phase-enhanced images. This DVF is then utilized in a motion-compensated, ordered-subset, simultaneous algebraic reconstruction technique, culminating in the generation of 4D CBCT images. The efficacy of this method has been substantiated through validation on both simulated and clinical datasets, with the results from comparative experiments indicating promising outcomes.
13 May 2025 in IEEE Transactions on Radiation and Plasma Medical Sciences
Micro-CT provides tomographic information for small animals and plays an important role in preclinical research. Currently, the most micro-CT systems employ the single-source configuration and low-dose micro-focus X-ray tube, resulting in long scanning time and weak low-contrast discriminative ability. To address the two vital issues, we design a novel rotation gantry mounted triple-source conebeam micro-CT for the first time. Specifically, three pairs of tubes and detectors are installed in a single gantry. Compared with a single source configuration, the proposal increases the scanning efficiency by three times without aggravating any mechanical burden. The low-contrast discrimination is improved by multi-material decomposition scheme based on triple-energy CT (TECT) images. Experiments were conducted using both phantoms and live rats. In the high-speed study, the proposal has reasonable agreement in signal-to-noise ratio and MTF compared to single-source scanner, even with a threefold increase in scanning speed. In the low-contrast discrimination study, on the digital phantom, the TECT correctly discriminates the materials of 2% linear attenuation coefficient difference. On the live rat, the decomposition accuracy of TECT has been enhanced by up to 21.9% compared to single-energy CT. These results validate the promising of the proposal for high-speed and low-contrast discrimination applications.
1 January 2026 in IEEE Transactions on Image Processing
Mamba, the state space model (SSM), has attracted significant attention for its ability to model long-range dependencies with linear complexity, achieving success in medical image segmentation. However, the previous cross-scanning approach in Mamba struggles to capture both long-range and short-range dependencies simultaneously and treats the features of each path equally. This imbalance between local and global modeling capabilities can adversely impact segmentation performance. To address these challenges, we propose WAS-Mamba, a novel method specifically designed for Mamba-based medical image segmentation. WAS-Mamba introduces a cross-channel window scanning strategy (CCWScan) that enables sequences to preserve original local image features during the transformation process. Furthermore, WAS-Mamba employs a weighted state space model (WSSM) to dynamically fuse spatial and frequency domain information, improving the capture of local details and global context for accurate segmentation. We validated the superior performance of WAS-Mamba across five datasets covering different anatomical regions, which include CT and MRI images: Synapse, BTCV, ACDC, BraTS, and Decathlon-Lung. In particular, we achieved a Dice coefficient of 88.09% on the Synapse dataset, with a 33% reduction in computational complexity and inference time compared to the second-best model. The code and model will be released at https://github.com/1605066114/WAS-Mamba.
4 September 2025 in IEEE Transactions on Instrumentation and Measurement
Utilizing Cone Beam Computed Tomography (CBCT) to monitor Minimally Invasive Surgery (MIS) is essential for ensuring surgical safety, particularly in high-precision procedures such as cochlear implantation (CI). Existing CBCT systems—including whole-body units and those tailored for ear diagnostics, dentistry, or head imaging—are typically bulky and poorly integrated, limiting their utility in intraoperative environments. To address this gap, we propose a highly integrated, high-resolution CBCT system specifically designed for intraoperative cochlear MIS. The system incorporates a dual-joint robotic structure, enabling both compact integration and mechanical flexibility by allowing the imaging chain to be folded to accommodate other surgical instruments. From a measurement perspective, the system is engineered to optimize key imaging performance parameters through a balance between Field of View (FOV), spatial resolution, and mechanical stability. Geometric parameters were quantitatively optimized to minimize rotational bending moments, ensuring consistent and precise scanning. Imaging performance was systematically assessed using four types of phantoms. Quantitative measurements demonstrated a spatial resolution of 150 μm (3.3 lp/mm), a Contrast-to-Noise Ratio (CNR) of 8.63 at a 2.65% contrast level, and a 3D uniformity deviation of 1.8%. Additionally, qualitative validation was conducted using a live pig and a human cadaver head to simulate real clinical conditions. These findings highlight the system’s capability to deliver accurate, repeatable, and measurement-driven intraoperative imaging, offering enhanced safety and precision for cochlear MIS. The work contributes not only to imaging system integration but also to quantitative measurement methodology aligned with the focus of instrumentation and measurement research.