Magnetic Particle Imaging (MPI), an emerging imaging modality characterized by its high sensitivity and spatiotemporal resolution, holds immense promise for advanced biomedical applications. However, its clinical translation is significantly hindered by the laborious and time-intensive physical calibration process requisite for acquiring the system matrix (SM). To address this limitation, this paper introduces the Adaptive Chebyshev Tensor Decomposition (ACTD) method. By incorporating an online-optimizable adaptive affine basis into a compact tensor-product model, the proposed method achieves an accurate representation of the intrinsic structure of the SM. We further establish a unified, SM-aware reconstruction framework wherein the iterative solving process is entirely transformed into the low-rank Chebyshev spectral domain, thus circumventing the need for explicit storage and manipulation of the complete SM. Experimental evaluations on the public OpenMPIData demonstrate that ACTD facilitates high-fidelity reconstruction from merely 1.5625% sparse sampling, significantly outperforming existing state-of-the-art methods with respect to key metrics such as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM). Moreover, the proposed framework reduces the memory footprint for reconstruction to only 6.7% of that required by conventional methods and inherently supports arbitrary-scale super-resolution imaging. Consequently, ACTD provides an efficient and robust approach for rapid and flexible MPI reconstruction, independent of supervised data-driven paradigms, thereby offering strong potential for practical applications. The code for this work has been open-sourced at: https://github.com/GALIFeng/ACTD/.

Deep learning-based methods have become the dominant approach for low-dose CT (LDCT) denoising. However, their performance often degrades on cross-domain datasets due to domain gaps, highlighting the need for effective domain adaptation techniques. While domain adaptation methods based on the pretraining and fine-tuning paradigm show great potential, they typically require additional labeled data from the target domain, which limits their practicality. Therefore, this work aims to develop a self-supervised fine-tuning method for LDCT denoising. In our work, we propose to fine-tune pretrained models using self-supervised loss based on pixel shuffle image preprocessing. Additionally, we design a two-stage fine-tuning strategy to mitigate the input misalignment between the pretraining and fine-tuning stages. Furthermore, to effectively capture prior knowledge from the source domain, we design a dual-scale SwinIR model as the pretrained backbone. We evaluate our method on two public datasets, and the results demonstrate that it bridges the domain gap without requiring target-domain labels, achieving effective denoising performance and strong cross-domain generalization. Code and model for our proposed approach are publicly available at https://github.com/Wasserdawn/TSFDAN.

Objective To address the scarcity of annotated data for pediatric abdominal CT imaging and the insufficient generalization capability of existing models,we constructed a self-supervised pretraining archi-tecture tailored for pediatric CT domain adaptation based on the visual foundation model DINOv3,and validated its performance in the task of pediatric abdominal multi-organ segmentation.Methods We built a general-pur-pose radiological visual representation using the large-scale adult CT dataset CT-3M,and introduced a Gram-anchoring mechanism that employs a frozen adult pretrained model as a structural teacher to guide domain align-ment of local topological structures on unlabeled pediatric CT data.Combined with a multi-scale feature aggrega-tion strategy and a lightweight Primus decoder,downstream segmentation tasks were evaluated on a public pedi-atric CT dataset.Based on case-wise paired results,we compared the mean Dice similarity coefficient(DSC)and mean intersection over union(IoU)between our model and the baseline nnU-Net using the Wilcoxon signed-rank test,and computed the relative performance improvements.Results A total of 867 abdominal CT imaging cases were collected,constituting a pretraining dataset comprising 367 588 two-dimensional CT slices.On the public Pediatric-CT-SEG dataset(359 cases),our model achieved a mean DSC of(71.38±1.08)%and a mean IoU of(63.73±1.01)%,representing improvements of 3.22%and 3.59%over the baseline nnU-Net,respectively,with statistically significant differences(P＜0.05).Stable improvements in mean DSC were also observed for small-volume or boundary-ambiguous organs,including the duodenum(5.78%),pancreas(4.69%),left adrenal gland(2.89%),right adrenal gland(1.12%),and gallblad-der(1.86%).Ablation experiments demonstrated that DSC improved by 1.44%,3.66%,5.78%,and 6.45%following adult pretraining,pediatric domain adaptation,high-resolution adaptation,and multi-scale feature aggregation,respectively.Conclusions The self-supervised pretraining framework proposed in this study effectively alleviates the domain shift between adult and pediatric abdominal CT images,significantly enhances segmentation accuracy for pediatric abdominal multi-organs-particularly small organs and structures with complex boundaries-and provides a reliable technical solution for intelligent pediatric imaging analysis in scenarios with limited annotated data.

Although Ta4HfC5 ceramic has a highest melting point reported, it ablates rapidly in oxidative environment because Ta2O5 has a low melting point of 1872 degrees C. In this work, ultrafine Ta-Hf-C solid-solution powders (49-70 nm) with tunable Ta/Hf ratio were synthesized via a solvothermal route combined with carbothermal reduction using a liquid-phase precursor. Ta-Hf-C solid-solution bulks with relative densities of 98-99% were subsequently obtained by an ultra-fast pressure sintering technique. The effects of Ta/Hf ratio on densification behavior, mechanical properties, and oxidation-ablation resistance were systematically evaluated. Among all compositions, the 1TaC-4HfC ceramic exhibited the lowest mass and linear ablation rates of 0.422 mg/s and 14.237 mu m/s after 120 s exposure to laser irradiation at a heat flux of 30 MW/m2, with a surface temperature up to 3800 degrees C. During ablation, a dense protective oxide layer formed through the reaction between HfO2 and Ta2O5, generating Hf6Ta2O17 coatings on the HfO2 framework, while molten Ta2O5 acted as a liquid bonding phase to wet and integrate the oxide phases with the residual carbide skeleton. This synergistic HfO2/Hf6Ta2O17/ Ta2O5 composite structure effectively suppressed oxygen permeation and enhanced high-temperature stability above 3000 degrees C.

Sparse-view computed tomography (SVCT) offers the advantages of accelerated scanning and reduced X-ray radiation dose in different clinical applications. However, it faces a challenge due to incomplete data acquisition, resulting in streak artifacts in the analytically reconstructed CT images. Utilizing self-supervised learning, implicit neural representation (INR) recently has shown great promise in addressing inverse problems such as SVCT reconstruction. Nonetheless, given that the input of original INR only contains coordinate information, it is limited to represent one SVCT instance at a time, and its performance significantly declines when performing cross-instance reconstruction. In this study, we propose a novel self-supervised framework named VFMINR, which leverages generalizable representations extracted from the visual foundation models (VFMs) to tackle the cross-instance reconstruction issue of INR. Specifically, VFMINR first utilizes VFMs to effectively capture the spatial and frequency domain representations of sinograms, and then a fusion module is applied to fuse two domain features into complementary representations. This combination maximizes the utilization of local detail information from the spatial domain and the global structural information from the frequency domain. Subsequently, an adaptive cell decoding strategy is designed to map representations into variable resolution hybrid feature grids, which are integrated into the learning of the INR to enhance its generalizability for different SVCT instances. The VFMs and VFMINR are trained by using only SV sinogram data, and extensive results confirm that the proposed method can effectively handle the generalization problem of INR, while achieving superior performance in image fidelity and artifact suppression. The code is available at: https://github.com/nightastars/VFMINR-main.

Current auto-regressive models can generate high-quality, topologically precise meshes; however, they necessitate thousands-or even tens of thousands-of next-token predictions during inference, resulting in substantial latency. We introduce XSpecMesh, a quality-preserving acceleration method for auto-regressive mesh generation models. XSpecMesh employs a lightweight, multi-head speculative decoding scheme to predict multiple tokens in parallel within a single forward pass, thereby accelerating inference. We further propose a verification and resampling strategy: the backbone model verifies each predicted token and resamples any tokens that do not meet the quality criteria. In addition, we propose a distillation strategy that trains the lightweight decoding heads by distilling from the backbone model, encouraging their prediction distributions to align and improving the success rate of speculative predictions. Extensive experiments demonstrate that our method achieves a 1.7x speedup without sacrificing generation quality. Our code will be released.

Vanadium films were grown on MgO (100) substrates using the magnetron sputtering method by adjusting the depositing time and substrate temperature. The superconducting transition Ta changing from 4.43 K to 2.96 K in two-dimensional (2D) V films can be explained by proximity effect, and the Ta decreases from 3.22 K to 2.65 K as the disorder degree described by Ioffe-Regel constant kFl increases in three-dimensional (3D) V films. Superconductor-metal transition (SMT) occurs in 3D V films that were deposited at different substrate temperatures. In 3D V film deposited at 723 K, the resistivity first decreases abruptly and then becomes nearly independent of temperature due to dissipation effect. While in V film deposited at 623 K, the resistivity exhibits three-step development of superconductivity because of the non-uniformity of the Josephson couplings between islands, the slippages of thermally activation (TA) and macroscopic quantum tunneling (MQT) in the film. The 2D V film is the crossover regime, while the 3D V film is dirty limit.

The pure phase and ultrafine WB2 powders were successfully synthesized using a liquid phase precursor synergized with boro/carbothermal reduction method. Subsequently, dense WB2-SiC composites (99.5 %) were fabricated through hot pressing sintering. The effects of La2O3 additive on the densification, mechanical properties, and ablation resistance of WB2-SiC composites were systematically investigated. The ablation behavior the composites was evaluated using a plasma flame. The results indicate that the mass ablation rate of WB2-SiC- La2O3 is 0.463 mg/s, and the linear ablation rate is 0.311 mu m/s when exposed to a 2273 K plasma flame for 60 The excellent ablation resistance is attributed to the reaction between La2O3 and SiO2 at high temperature, which generates La2Si2O7, effectively pinned in the glassy phase to inhibit the volatilization of B2O3. Furthermore, La2O3 also reacts with B2O3 and SiO2 to generate a highly viscous B-Si-O-La protective layer, which blocks ingress of oxygen.

Comprehensive performance of ultra-high temperature ceramic composites is predominantly governed by purity, morphology and crystallinity of HfB2 powders. Nevertheless, achieving controllable synthesis of high-quality HfB2 ceramic powders remains challenging due to insufficient mechanistic insights into the evolution and growth of powder morphology. In this work, we systematically investigate the effects of diverse carbon sources on the morphology and crystallinity of HfB2 powders during carbothermal reduction. The results demonstrate that the graphitization degree of carbon sources directly governs whether HfB2 powder exists in particle or rodlike morphology. Microstructural evolution analysis indicates that the morphological differentiation in HfB2 powders is predominantly determined by the graphitization degree of the carbon source, which directly controls the oriented attachment of HfO2 particles in precursors and modulates the reaction temperature during subsequent carbothermal reduction processes. The elevated graphitization degree promotes a more orderly arrangement of HfO2 nanograins. The HfB2 powders maintain the anisotropic rod-like morphology inherent in their HfO2 precursors throughout the growth process, resulting in high-crystallinity, rod-shaped ceramic powders. These crystalline, rod-shaped powders display remarkable antioxidant properties; their mass fraction decreases by only 2.25 % following non-isothermal oxidation at 1000 degrees C. This research holds significant implications for developing controllable preparation methods for ceramic powders of different morphologies and provides a theoretical foundation for enhancing the performance of ultra-high temperature ceramic materials.

High-quality N-doped VO2 was produced by thermal oxidation of Vanadium nitride (VN). The content of nitrogen in VO2 during the thermal treatment was analyzed via photoelectron spectroscopy, revealing a gradual decrease in nitrogen content with increasing the annealing temperature. Even after the disappearance of VN peaks in XRD, a residual amount of V-N and V-N-O bonds persist in the annealed compounds, indicating the formation of Ndoped VO2. The obtained N-doped VO2 has a high enthalpy confirmed by differential scanning calorimeter, and its phase transition temperatures can be changed in the range from 59.64 C to 70.03 C. The impact of varying annealing temperatures on the phase transition and thermochromic property of N-doped VO2 was systematically studied. N-doped VO2 obtained at 500 C exhibits optimal thermochromic performance, with an average Tlum of 43.38 % and Delta TIR of 17.53 %. Additionally, when using a quartz substrate as a background, and placing it on the heating table, the temperature of the quartz is 101.8 C. Under these conditions, the infrared thermography of the compressed N-doped VO2 sheet demonstrates a maximum temperature reduction of 23.7 % compared to the quartz background, indicating its strong potential for adaptive infrared camouflage applications. In all, this study introduces a novel approach for synthesizing VO2, highlighting VN as an effective precursor for N-doped VO2 mass production.

Various hollow structured HfB2 powders were synthesized via a co-precipitation assisted boron/carbon thermal reduction method, and the influence mechanism of carbon source on morphology was systematically investigated. When glucose or inositol was employed as the carbon source, the hexagonal ring structure formed in the precursor during complexation directed the growth of HfB2 nanoparticles that were anchored on hollow spherical carbon templates, resulting in a unique hollow core-shell architecture. Additionally, the morphology could be precisely adjusted from hollow spherical to hollow rod-like structures by simply controlling the stirring temperature during complexation. The growth process of HfB2 with various morphologies was elucidated, and the mechanism of this unique morphological evolution was revealed to originate from the structural differences of precursor complexes. The HfB2 powders with hollow spherical structure derived from glucose exhibits excellent electromagnetic absorption ability due to the synergistic effect of multiple scattering and interface polarization within the unique hollow structure, with a maximum reflection loss (RL) of -57.8dB. Quantitative analysis of the loss mechanisms further demonstrates that the outstanding absorption performance is predominantly governed by polarization loss rather than conduction loss.

High-temperature thermal-insulation aerogels with high mechanical strength and well oxidation resistance are critical for the thermal protect system (TPS) of aerospace vehicles. However, it is still challenging to achieve large-scale preparation of aerogels that integrate the above functions through a simple method. In this work, carbon fiber (CF) felts are introduced as a reinforcement and template to prepare silicon carbide nanowire (SiCnw) aerogel composites through a simple in-situ growth strategy. The high-strength CF as the skeleton endows high mechanical strength, and the in-situ grown SiCnw aerogels lead to good thermal insulation of the composites. More importantly, a dense SiC shell was also formed and tightly coated on the surface of CF (CF@SiC) during the in-situ growth process, which effectively protects the CF from oxidation, thus enhancing the high temperature oxidation resistance of the composites. As a result, the optimal CF@SiC/SiCnw aerogel composites possess a low thermal conductivity of only 0.056 W m- 1 K- 1 at room temperature. The back temperature of the 9 mm thick composites is only 151.9 degrees C after exposure to a butane flame at 1220 degrees C for 300 s, and it still maintains considerable thermal insulation and mechanical properties after the ablation. In addition, the machinability of the CF felts makes it easy to prepare the composites in various shapes and larger sizes. This work provides a simple and scalable strategy to synthesize reliable SiC-based aerogel composites for high-temperature thermal insulation applied in extreme environments.

Low-dose computed tomography (LDCT) denoising methods based on supervised learning with labeled simulation data have made significant progress. However, these methods usually struggle to directly process unlabeled LDCT images due to inherent biases. While unsupervised methods have been explored to utilize unlabeled LDCT images, they typically involve complex network structures with limited denoising performance. To address these issues, we propose a self-adaptive weight embedded lightweight semi-supervised network (SWELNet) for unlabeled LDCT image denoising, which integrates supervised and unsupervised learning in a lightweight architecture. Unlike other semi-supervised algorithms that only consider the correlations between labeled simulation data and unlabeled data, the proposed SWELNet not only takes into account correlations but also the differences between data. There are three modules in the proposed network respectively for feature extraction, refinement and self-adaptive weight. Specially, the multi-scale convolution feature extraction module (MCFEM) and recursive module (RECM) extract and refine common representations from labeled simulation and unlabeled data with the well-designed. After that, the softmax feature fusion module (SFFM) with self-adaptive weighted learning for forming different feature spaces for two types of data. Extensive experiments using one simulation and two unlabeled datasets demonstrate that the proposed SWELNet outperforms several state-of-the-art baseline network methods in terms of robustness and generalization, as well as inference efficiency. The code is available at https://github.com/nightastars/SWELNet-main.git.

Incorporation of high thermal conductive materials with carbon fiber (CF) reinforced carbon matrix (C/C) composites is an effective strategy for enhancing their overall thermal conductivity. However, this approach often disrupts the orientation of graphite within the carbon matrix. In this study, graphene-modified C/C composites with preferred orientation were constructed by decorating CF with graphene oxide (GO), followed by liquid phase impregnation of mesophase pitch, carbonization, and graphitization. The orientation of mesophase pitch derived-graphite carbon matrix was preserved after the aligned decoration of GO on the CF, due to the chemical bonding between GO and polydopamine (PDA) on the CF surface. Importantly, the moderate GO decoration amount increased the overall graphitization degree of the C/C composites, while simultaneously strengthening their microstructural anisotropy. As a result, the in-plane thermal conductivity of the C/C composites increased from 183.09 to 210.97 W/(m center dot K) when 2.4 wt% GO was decorated on the CF surfaces. The finite element analysis results also indicate that the in-plane thermal conductivity was enhanced with GO decoration. In addition, the tensile strength of the C/C composites increased from 12.76 to 16.52 MPa. This work provides valuable insights into the design and fabrication of C/C composites with enhanced directional thermal conductivity and mechanical properties.

This paper proposes a convolutional Long Short-Term Memory (ConvLSTM) network integrated with multi-channel features dedicated to ionospheric total electron content (TEC) forecasting. To improve generalization, solar, and geomagnetic activity indices are added as auxiliary channel inputs. The model is built upon an Encoder–Decoder (ED) architecture enhanced with residual connections and convolutional channel projection, which collectively improve the synergy among its core components. Based on this framework, we developed ED-ConvLSTM-Res, a multi-channel feature-based global ionospheric TEC prediction model. Comprehensive accuracy evaluation and comparative tests were carried out using datasets from the solar minimum year of 2019 and the current solar maximum year of 2024. The results indicate that the proposed model consistently achieves strong predictive performance compared with other models, along with a significantly enhanced feature representation capability. Specifically, the Root Mean Square Errors (RMSE) of the ED-ConvLSTM-Res model’s predictions in 2019 and 2024 are 1.28 TECU and 5.28 TECU, respectively, while the corresponding Mean Absolute Errors (MAE) are 0.87 and 3.87, and the coefficients of determination (R2) are 0.95 and 0.94. In the current high solar activity year 2024, the proposed model achieves error reductions of 13.6% in MAE and 11.6% in RMSE compared with the Center for Orbit Determination in Europe (CODE)’s one-day-ahead forecast product, c1pg. These results confirm that the proposed model not only outperforms the ConvLSTM model without additional indices and c1pg but also exhibits strong generalization capability, maintaining stable performance with low errors under both high and low solar activity conditions.

Magnetic Particle Imaging (MPI) offers unique advantages, including high sensitivity, real-time imaging, and absence of ionizing radiation. However, the prevailing system matrix (SM)-based reconstruction in MPI faces critical limitations: time-consuming calibration, noise vulnerability, and reliance on high-resolution training data. To overcome these challenges, we propose an imaging physics driven neural field framework for efficient SM calibration and robust reconstruction. Key innovations include: (1) First-order derivative constraints to suppress spiky noise, (2) An M-order separable representation to enforce smoothness and reduce fluctuation artifacts, and (3) Chebyshev polynomial integration to enhance encoding efficiency and embed imaging physics priors. The method adapts to variable resolution requirements, reduces dependency on high-resolution data, and demonstrates robustness to noise across diverse SNR conditions. Experiments on the OpenMPI dataset show remarkable performance, achieving 1.55% nRMSE at 25% sparsity and minimal 0.21% degradation at 6.25% sparsity. Furthermore, upsampling sparse internal MPI system via the proposed method successfully reconstructs phantom geometries with high fidelity. These results validate the method's potential to advance MPI toward broader research applications.

Zirconium boride (ZrB2) whiskers hold significant potential for toughening ultra-high temperature ceramics (UHTCs). Yet, their fabrication remains challenging due to limited knowledge of the microstructure evolution. Herein, high-purity ZrB2 whiskers were successfully prepared by electrospinning combined with carbothermal reduction. Specifically, a precursor solution containing Zr-O-C-B networks stabilized through hydrogen bonds with polyvinyl pyrrolidone was synthesized as a spinnable ZrB2 precursor. High-purity ZrB2 whiskers with a diameter of ca. 200 nm and a length of 5-10 mu m were obtained by precise heat treatment of ZrB2 precursor fibers. In this process, ZrO2 grains formed initially and transformed into ZrB2 grains through carbothermal reduction, which further converted into ZrB2 whiskers through a solid-liquid-solid mechanism. To testify to the toughening effect for UHTCs, the ZrB2 whiskers were added into ZrB2 powders and followed by sintering. This strategic addition of 5 wt% ZrB2 whiskers led to a 19% enhancement in the fracture toughness of ZrB2 bulk. This study illuminated the preparation strategy and formation mechanism for ZrB2 whiskers and their application for toughening UHTCs.

Benign and malignant classification of clustered microcalcifications (MCs) in digital breast tomosynthesis (DBT) is an essential task in computer-aided diagnosis. However, due to the anisotropic resolution of DBT, three-dimensional (3-D) convolutional neural network (CNN)-based methods cannot extract hierarchical features efficiently. Moreover, the sparse distribution of MC points in the cluster makes it difficult for the CNN to extract discriminative structural information for classification. To comprehensively address these challenges, we propose a novel structure-aware hierarchical network (SAH-Net) for benign and malignant classification of clustered MC in a DBT volume. Specifically, the two-dimensional (2-D) group convolution is used to extract intraslice features. The one-to-one correspondence between group convolutions and slices ensures the independence of hierarchical feature extraction. Then, a partial deformable Transformer-based 3-D structural feature learning module is proposed to capture the long-range dependency between MC points in the cluster. We evaluate the proposed method on an in-house dataset with 495 clustered MCs collected from 462 DBT images. Experimental results confirm the validity of our proposed modules. The results also show that the proposed SAH-Net outperforms several other representative methods on this topic, and achieves the best classification result, with an area under the receiver operation curve (AUC) of 86.87%. The implementation of the proposed model is available at https://github.com/sunhaotian130911/SAHNet .

X-ray computed tomography (CT) plays a crucial role in modern medical imaging for its non-invasive acquisition of anatomical information about the human body. However, its inherent carcinogenic risk remains an unavoidable topic. Low-dose CT (LDCT) reduces ionizing radiation exposure to minimize harm to the human body, but it introduces noise and artifacts of varying intensity, which can affect diagnosis and analysis. In recent years, deep learning has demonstrated competitive performance in medical image denoising. However, most current deep learning-based methods primarily focus on specific noise intensities, leading to degradation when faced with noise of different intensities. In this paper, we propose an Adaptive Noise-Aware Denoising Generative Adversarial Networks (ANAD-GAN) to address the aforementioned issues. The proposed noise-aware memory module can remember and decode high-level hidden features of noise of different intensities adaptively through self-updating feature vectors. To explore a more accurate perception of noise distribution, an intricately designed block named BLOGS is incorporated in our framework. We also integrate a discriminator-based detail-preserving loss in our framework to further enhance the visual quality of the denoised images. In summary, our innovative network dynamically adjusts to varying noise intensities, yielding remarkable denoising results. It attains a PSNR of 48.36 and SSIM of 0.9848 across 5936 images in the AAPM dataset, achieves a PSNR of 48.98 and SSIM of 0.9882 on the Philips dataset with 5000 images, and demonstrates outstanding performance with a PSNR of 50.74 and SSIM of 0.9906 on the private Siemens dataset comprising 4000 images. Through extensive experiments on three datasets, our method surpasses various widely adopted techniques in effectively addressing noise of different intensities.

Low-dose computed tomography (LDCT) denoising tasks face significant challenges in practical imaging scenarios. Supervised methods encounter difficulties in real-world scenarios as there are no paired data for training. Moreover, when applied to datasets with varying noise patterns, these methods may experience decreased performance owing to the domain gap. Conversely, unsupervised methods do not require paired data and can be directly trained on real-world data. However, they often exhibit inferior performance compared to supervised methods. To address this issue, it is necessary to leverage the strengths of these supervised and unsupervised methods. In this paper, we propose a novel domain adaptive noise reduction framework (DANRF), which integrates both knowledge transfer and style generalization learning to effectively tackle the domain gap problem. Specifically, an iterative knowledge transfer method with knowledge distillation is selected to train the target model using unlabeled target data and a pre-trained source model trained with paired simulation data. Meanwhile, we introduce the mean teacher mechanism to update the source model, enabling it to adapt to the target domain. Furthermore, an iterative style generalization learning process is also designed to enrich the style diversity of the training dataset. We evaluate the performance of our approach through experiments conducted on multi-source datasets. The results demonstrate the feasibility and effectiveness of our proposed DANRF model in multi-source LDCT image processing tasks. Given its hybrid nature, which combines the advantages of supervised and unsupervised learning, and its ability to bridge domain gaps, our approach is well-suited for improving practical low-dose CT imaging in clinical settings. Code for our proposed approach is publicly available at https://github.com/tyfeiii/DANRF.