Safety analysis of Oseltamivir and Baloxavir Marboxil after market approval: a pharmacovigilance study based on the FDA adverse event reporting system.

BACKGROUND AND OBJECTIVES: Amidst limited influenza treatment options, evaluating the safety of Oseltamivir and Baloxavir Marboxil is crucial, particularly given their comparable efficacy. This study investigates post-market safety profiles, exploring adverse events (AEs) and their drug associations to provide essential clinical references. METHODS: A meticulous analysis of FDA Adverse Event Reporting System (FAERS) data spanning the first quarter of 2004 to the fourth quarter of 2022 was conducted. Using data mining techniques like reporting odds ratio (ROR), proportional reporting ratio, Bayesian Confidence Propagation Neural Network, and Multiple Gamma Poisson Shrinkage, AEs related to Oseltamivir and Baloxavir Marboxil were examined. Venn analysis compared and selected specific AEs associated with each drug. RESULTS: Incorporating 15,104 Oseltamivir cases and 1,594 Baloxavir Marboxil cases, Wain analysis unveiled 21 common AEs across neurological, psychiatric, gastrointestinal, dermatological, respiratory, and infectious domains. Oseltamivir exhibited 221 significantly specific AEs, including appendicolith [ROR (95% CI), 459.53 (340.88 ∼ 619.47)], acne infantile [ROR (95% CI, 368.65 (118.89 ∼ 1143.09)], acute macular neuroretinopathy [ROR (95% CI), 294.92 (97.88 ∼ 888.64)], proctitis [ROR (95% CI), 245.74 (101.47 ∼ 595.31)], and Purpura senile [ROR (95% CI), 154.02 (81.96 ∼ 289.43)]. designated adverse events (DMEs) associated with Oseltamivir included fulminant hepatitis [ROR (95% CI), 12.12 (8.30-17.72), n=27], ventricular fibrillation [ROR (95% CI), 7.68 (6.01-9.83), n=64], toxic epidermal necrolysis [ROR (95% CI), 7.21 (5.74-9.05), n=75]. Baloxavir Marboxil exhibited 34 specific AEs, including Melaena [ROR (95% CI), 21.34 (14.15-32.18), n = 23], cystitis haemorrhagic [ROR (95% CI), 20.22 (7.57-54.00), n = 4], ileus paralytic [ROR (95% CI), 18.57 (5.98-57.71), n = 3], and haemorrhagic diathesis [ROR (95% CI), 16.86 (5.43-52.40)), n = 3]. DMEs associated with Baloxavir Marboxil included rhabdomyolysis [ROR (95% CI), 15.50 (10.53 ∼ 22.80), n = 26]. CONCLUSION: Monitoring fulminant hepatitis during Oseltamivir treatment, especially in patients with liver-related diseases, is crucial. Oseltamivir's potential to induce abnormal behavior, especially in adolescents, necessitates special attention. Baloxavir Marboxil, with lower hepatic toxicity, emerges as a potential alternative for patients with liver diseases. During Baloxavir Marboxil treatment, focused attention on the occurrence of rhabdomyolysis is advised, necessitating timely monitoring of relevant indicators for those with clinical manifestations. The comprehensive data aims to provide valuable insights for clinicians and healthcare practitioners, facilitating an understanding of the safety profiles of these influenza treatments in real-world scenarios.

Restructuring Electrolyte Solvation by a Versatile Diluent Toward Beyond 99.9% Coulombic Efficiency of Sodium Plating/Stripping at Ultralow Temperatures.

The reversible and durable operation of sodium metal batteries at low temperatures (LT) is essential for cold-climate applications but is plagued by dendritic Na plating and unstable solid-electrolyte interphase (SEI). Current Coulombic efficiencies of sodium plating/stripping at LT fall far below 99.9%, representing a significant performance gap yet to be filled. Here, the solvation structure of the conventional 1 m NaPF6 in diglyme electrolyte by facile cyclic ether (1,3-dioxolane, DOL) dilution is efficiently reconfigured. DOL diluents help shield the Na+-PF6 - Coulombic interaction and intermolecular forces of diglyme, leading to anomalously high Na+-ion conductivity. Besides, DOL participates in the solvation sheath and weakens the chelation of Na+ by diglyme for facilitated desolvation. More importantly, it promotes concentrated electron cloud distribution around PF6 - in the solvates and promotes their preferential decomposition. A desired inorganic-rich SEI is generated with compositional uniformity, high ionic conductivity, and high Young's modulus. Consequently, a record-high Coulombic efficiency over 99.9% is achieved at an ultralow temperature of -55 °C, and a 1 Ah capacity pouch cell of initial anode-free sodium metal battery retains 95% of the first discharge capacity over 100 cycles at -25 °C. This study thus provides new insights for formulating electrolytes toward increased Na reversibility at LT.

Cycle-Refined Multidecision Joint Alignment Network for Unsupervised Domain Adaptive Hyperspectral Change Detection.

Hyperspectral change detection, which provides abundant information on land cover changes in the Earth's surface, has become one of the most crucial tasks in remote sensing. Recently, deep-learning-based change detection methods have shown remarkable performance, but the acquirement of labeled data is extremely expensive and time-consuming. It is intuitive to learn changes from the scene with sufficient labeled data and adapting them into an unlabeled new scene. However, the nonnegligible domain shift between different scenes leads to inevitable performance degradation. In this article, a cycle-refined multidecision joint alignment network (CMJAN) is proposed for unsupervised domain adaptive hyperspectral change detection, which realizes progressive alignment of the data distributions between the source and target domains with cycle-refined high-confidence labeled samples. There are two key characteristics: 1) progressively mitigate the distribution discrepancy to learn domain-invariant difference feature representation and 2) update the high-confidence training samples of the target domain in a cycle manner. The benefit is that the domain shift between the source and target domains is progressively alleviated to promote change detection performance on the target domain in an unsupervised manner. Experimental results on different datasets demonstrate that the proposed method can achieve better performance than the state-of-the-art change detection methods.

Research on the Influence of Heterogeneity and Viscosity on the Fluid Intrusion Mechanism of the Water Flooding Process Based on the Microscopic Visualization Experiment.

The flow law of immiscible fluids in porous media plays an important role in the development of oil and gas fields. In the process of water flooding reservoir development, when the water phase displaces the oil phase, a fluid with higher viscosity, as a fluid with lower viscosity, the oil-water interface will always be unstable, resulting in different fingering effects. After water flooding, the distribution law of oil and water in the reservoir is mainly affected by the fluid intrusion mechanism. Due to the difference of capillary force, viscous force, and other microscopic forces, the fluid intrusion mechanism is mainly divided into two types: viscous fingering and capillary fingering. At the same time, due to the influence of reservoir heterogeneity, the fingering effect in the process of water displacement in porous media will be influenced to a certain extent. Based on the two-dimensional microscopic visualization experiment, this paper extracted the variance of the static parameter G in the capillary number calculation method of the two-dimensional microscopic model to represent the heterogeneity and conducted displacement experiments with different viscosities and flow rates to study the influence of the flow rate, viscosity, and heterogeneity on the results of water flooding. The experiments found that as for the influence of flow velocity, with the increase of flow velocity, that is, with the increase of capillary number, the recovery degree decreases first and then increases. As for the influence of viscosity, from a numerical point of view, the displacement efficiency and conformance coefficient of the low-viscosity group are higher than those of the high-viscosity group. From the trend, with the increase of the capillary number, the displacement efficiency of both the low-viscosity and high-viscosity groups increases, while the conformance coefficient decreases first and then increases, indicating that capillary fingering and viscous fingering can occur in different viscosity reservoirs. As for the influence of heterogeneity, the conformance coefficient of the water flooding decreases with the increase of heterogeneity, and the viscous pointing trend caused by heterogeneity is stronger, resulting in an uneven water injection sweep and higher oil displacement efficiency within the swept area. It can be seen from the fluid intrusion mechanism diagram that with the increase of heterogeneity, the viscous fingering trend becomes more obvious; with the increase of viscosity, the fluid intrusion mechanism boundary moves down and the viscous fingering trend becomes more obvious.

SCFormer: Spectral Coordinate Transformer for Cross-Domain Few-Shot Hyperspectral Image Classification.

Cross-domain (CD) hyperspectral image classification (HSIC) has been significantly boosted by methods employing Few-Shot Learning (FSL) based on CNNs or GCNs. Nevertheless, the majority of current approaches disregard the prior information of spectral coordinates with limited interpretability, leading to inadequate robustness and knowledge transfer. In this paper, we propose an asymmetric encoder-decoder architecture, Spectral Coordinate Transformer (SCFormer), for the CDFSL HSIC task. Several dense Spectral Coordinate blocks (SC blocks) are embedded in the backbone of the encoder to establish feature representation with better generalization, which integrates spectral coordinates via Rotary Position Embedding (RoPE) to minimize spectral position disturbance caused by the convolution operation. Due to a large amount of hyperspectral image data and the high demand for model generalization ability in cross-domain scenarios, we design two mask patterns (Random Mask and Sequential Mask) built on unexploited spectral coordinates within the SC blocks, which are unified with the asymmetric structure to learn high-capacity models efficiently and effectively with satisfactory generalization. Besides, from the perspective of the loss function, we devise an intra-domain loss function founded on the Orthogonal Complement Space Projection (OCSP) theory to facilitate the aggregation of samples in the metric space, which promotes intra-domain consistency and increases interpretability. Finally, the strengthened class expression capacity of the intra-domain loss function contributes to the inter-domain loss function constructed by Wasserstein Distance (WD) for realizing domain alignment. Experimental results on four benchmark data sets demonstrate the superiority of the SCFormer.

MDEformer: Mixed Difference Equation Inspired Transformer for Compressed Video Quality Enhancement.

Deep learning methods have achieved impressive performance in compressed video quality enhancement tasks. However, these methods rely excessively on practical experience by manually designing the network structure and do not fully exploit the potential of the feature information contained in the video sequences, i.e., not taking full advantage of the multiscale similarity of the compressed artifact information and not seriously considering the impact of the partition boundaries in the compressed video on the overall video quality. In this article, we propose a novel Mixed Difference Equation inspired Transformer (MDEformer) for compressed video quality enhancement, which provides a relatively reliable principle to guide the network design and yields a new insight into the interpretable transformer. Specifically, drawing on the graphical concept of the mixed difference equation (MDE), we utilize multiple cross-layer cross-attention aggregation (CCA) modules to establish long-range dependencies between encoders and decoders of the transformer, where partition boundary smoothing (PBS) modules are inserted as feedforward networks. The CCA module can make full use of the multiscale similarity of compression artifacts to effectively remove compression artifacts, and recover the texture and detail information of the frame. The PBS module leverages the sensitivity of smoothing convolution to partition boundaries to eliminate the impact of partition boundaries on the quality of compressed video and improve its overall quality, while not having too much impacts on non-boundary pixels. Extensive experiments on the MFQE 2.0 dataset demonstrate that the proposed MDEformer can eliminate compression artifacts for improving the quality of the compressed video, and surpasses the state-of-the-arts (SOTAs) in terms of both objective metrics and visual quality.

Traumatic Iliac Arteriovenous Fistula Treated With the Amplatzer Vascular Plug II: A Case Report and Literature Review.

Traumatic iliac arteriovenous fistula is a rare complication of vascular injury. Open surgical repair has an incidence of postoperative complications. In recent years, endovascular treatment has shown better efficacy. We report a 62-year-old female AVF patient with a stab injury history of more than 16 years. Computed tomography angiography (CTA) revealed a large arteriovenous fistula between the right internal iliac artery and the common iliac vein. After considering the patient's relevant conditions, an endovascular approach was satisfactorily performed with the implantation of an Amplatzer Vascular Plug II to interrupt the abnormal vascular communication and maintain arterial and venous patency. The final control images showed closure of the arteriovenous communication.

Removing infrarenal inferior vena cava filters (IVCFs) with thrombus under protection of suprarenal IVCFs: A retrospective study in a single-center institution.

To determine feasibility of removing inferior vena cava filters (IVCFs) with massive thrombus (>1*1cm) under protection of suprarenal IVCFs, and evaluate the filter thrombus detachment due to removal. The patients who had massive infrarenal IVCFs thrombus and received retrieval under protection of suprarenal IVCFs were retrospectively reviewed from July 2018 to December 2021. Medical data of them including demographics, filter types, dwell time, management, thrombus detachment was collected, and analyzed. There were 33 patients having massive infrarenal IVCFs thrombus and receiving retrieval under protection of suprarenal IVCFs including 23 males and 10 females with a mean age of 55.30 ± 11.97 (range, 30-85 years). All Infrarenal IVCFs were removed successfully and 29 cases (87.88%) were confirmed detachment of thrombus by cavography including 7 small-size thrombus (<1*1cm) and 22 large-size thrombus (>1*1cm). Twenty-two suprarenal IVCFs trapped large-size thrombus were treated with additional anticoagulation and 21 of them had successful retrievals with additional anticoagulation period of 66.18 ± 43.38 days (range, 9-154 days). The large-size IVCFs thrombus may be break off during retrieval, and IVCFs with large-size thrombus could be removed safely with suprarenal IVCFs protection. The thrombus trapped in filters could be reduced with an additional period of anticoagulation.

Progressive Spatial Information-Guided Deep Aggregation Convolutional Network for Hyperspectral Spectral Super-Resolution.

Fusion-based spectral super-resolution aims to yield a high-resolution hyperspectral image (HR-HSI) by integrating the available high-resolution multispectral image (HR-MSI) with the corresponding low-resolution hyperspectral image (LR-HSI). With the prosperity of deep convolutional neural networks, plentiful fusion methods have made breakthroughs in reconstruction performance promotions. Nevertheless, due to inadequate and improper utilization of cross-modality information, the most current state-of-the-art (SOTA) fusion-based methods cannot produce very satisfactory recovery quality and only yield desired results with a small upsampling scale, thus affecting the practical applications. In this article, we propose a novel progressive spatial information-guided deep aggregation convolutional neural network (SIGnet) for enhancing the performance of hyperspectral image (HSI) spectral super-resolution (SSR), which is decorated through several dense residual channel affinity learning (DRCA) blocks cooperating with a spatial-guided propagation (SGP) module as the backbone. Specifically, the DRCA block consists of an encoding part and a decoding part connected by a channel affinity propagation (CAP) module and several cross-layer skip connections. In detail, the CAP module is customized by exploiting the channel affinity matrix to model correlations among channels of the feature maps for aggregating the channel-wise interdependencies of the middle layers, thereby further boosting the reconstruction accuracy. Additionally, to efficiently utilize the two cross-modality information, we developed an innovative SGP module equipped with a simulation of the degradation part and a deformable adaptive fusion part, which is capable of refining the coarse HSI feature maps at pixel-level progressively. Extensive experimental results demonstrate the superiority of our proposed SIGnet over several SOTA fusion-based algorithms.

Feature Pyramid Fusion Network for Hyperspectral Pansharpening.

Hyperspectral (HS) pansharpening aims at fusing an observed HS image with a panchromatic (PAN) image, to produce an image with the high spectral resolution of the former and the high spatial resolution of the latter. Most of the existing convolutional neural networks (CNNs)-based pansharpening methods reconstruct the desired high-resolution image from the encoded low-resolution (LR) representation. However, the encoded LR representation captures semantic information of the image and is inadequate in reconstructing fine details. How to effectively extract high-resolution and LR representations for high-resolution image reconstruction is the main objective of this article. In this article, we propose a feature pyramid fusion network (FPFNet) for pansharpening, which permits the network to extract multiresolution representations from PAN and HS images in two branches. The PAN branch starts from the high-resolution stream that maintains the spatial resolution of the PAN image and gradually adds LR streams in parallel. The structure of the HS branch remains highly consistent with that of the PAN branch, but starts with the LR stream and gradually adds high-resolution streams. The representations with corresponding resolutions of PAN and HS branches are fused and gradually upsampled in a coarse to fine manner to reconstruct the high-resolution HS image. Experimental results on three datasets demonstrate the significant superiority of the proposed FPFNet over the state-of-the-art methods in terms of both qualitative and quantitative comparisons.

MPCNet: Compressed multi-view video restoration via motion-parallax complementation network.

The performance in restoring compressed multi-view video (MVV) of the existing learning-based methods is limited because they only utilize information of temporally adjacent frames or parallax neighboring views. However, the compression artifacts caused by multi-view coding (MVC) may be related to the reference errors of intra-frame, inter-frame, and inter-view. In this paper, with delicately utilizing the stereo information from both temporal and parallax domains, a motion-parallax complementation network (MPCNet) is proposed to restore the quality of compressed MVV more efficiently. First, we introduce a motion-parallax complementation strategy consisting of a coarse stage and a fine stage. By mutually compensating the feature extracted from multiple domains, useful multi-frame information can be efficiently preserved and aggregated step by step. Second, an attention-based feature filtering and modulation module (AFFM) is proposed, which provides an efficient fusion method for two features by suppressing misleading information. By deploying it in most submodules of the proposed approach, the representational ability of MPCNet can be improved, resulting in a more substantial restoration performance. Experimental results prove the effectiveness of MPCNet by an average increase of 1.978 dB in PSNR, and 0.0282 in MS-SSIM. The BD-rate reduction can reach 47.342% on average. The subjective quality is greatly improved and lots of compression distortions are eliminated. Meanwhile, this work also benefits the accuracy improvement for high-level vision tasks, e.g., mIoU of semantic segmentation and mAP of object detection achieve 0.352 and 51.71, respectively. Quantitative and qualitative analyses demonstrate that MPCNet outperforms state-of-the-art approaches.

Block-Wise Partner Learning for Model Compression.

Despite the great potential of convolutional neural networks (CNNs) in various tasks, the resource-hungry nature greatly hinders their wide deployment in cost-sensitive and low-powered scenarios, especially applications in remote sensing. Existing model pruning approaches, implemented by a "subtraction" operation, impose a performance ceiling on the slimmed model. Self-knowledge distillation (Self-KD) resorts to auxiliary networks that are only active in the training phase for performance improvement. However, the knowledge is holistic and crude, and the learning-based knowledge transfer is mediate and lossy. Here, we propose a novel model-compression method, termed block-wise partner learning (BPL), which comprises "extension" and "fusion" operations and liberates the compressed model from the bondage of baseline. Different from the Self-KD, the proposed BPL creates a partner for each block for performance enhancement in training. For the model to absorb more diverse information, a diversity loss (DL) is designed to evaluate the difference between the original block and the partner. Besides, the partner is fused equivalently instead of being discarded directly. After training, we can simply adopt the fused compressed model that contains the enhancement information of partners but with fewer parameters and less inference cost. As validated using the UC Merced land-use, NWPU-RESISC45, and RSD46-WHU datasets, the BPL demonstrates superiority over other compared model-compression approaches. For example, it attains a substantial floating-point operations (FLOPs) reduction of 73.97% with only 0.24 accuracy (ACC.) loss for ResNet-50 on the UC Merced land-use dataset. The code is available at https://github.com/zhangxin-xd/BPL.

HTD-TS 3 : Weakly Supervised Hyperspectral Target Detection Based on Transformer via Spectral-Spatial Similarity.

As an advanced technique in remote sensing, hyperspectral target detection (HTD) is widely concerned in civilian and military applications. However, the limitation of prior and heterogeneous backgrounds makes HTD models sensitive to data corruption under various interference from the environment. In this article, a novel united HTD framework based on the concept of transformer is proposed to extract HTD based on transformer via spectral-spatial similarity (HTD-TS 3 ) under weak supervision, which opens up more flexible ways to study HTD. For the first time, the transformer mechanism is introduced into the HTD task to extract spectral and spatial features in a unified optimization procedure. By modeling long-range dependence among spectra, it realizes spectral-spatial joint inference based on long-range context, which addresses the issues of insufficient utilization of spatial information. To provide samples for weakly supervised learning (WSL), the coarse sample selection and spectral sequence construction in an efficient way are proposed, which makes full use of limited prior information. Finally, an exponential constrained nonlinear function is adopted to acquire pixel-level prediction via combining discriminative spectral-spatial features and coarse spatial information. Experiments on real hyperspectral images (HSIs) captured by different sensors at various scenes verify the effectiveness and efficiency of HTD-TS 3 .

REAF: Remembering Enhancement and Entropy-Based Asymptotic Forgetting for Filter Pruning.

Neurologically, filter pruning is a procedure of forgetting and remembering recovering. Prevailing methods directly forget less important information from an unrobust baseline at first and expect to minimize the performance sacrifice. However, unsaturated base remembering imposes a ceiling on the slimmed model leading to suboptimal performance. And significantly forgetting at first would cause unrecoverable information loss. Here, we design a novel filter pruning paradigm termed Remembering Enhancement and Entropy-based Asymptotic Forgetting (REAF). Inspired by robustness theory, we first enhance remembering by over-parameterizing baseline with fusible compensatory convolutions which liberates pruned model from the bondage of baseline at no inference cost. Then the collateral implication between original and compensatory filters necessitates a bilateral-collaborated pruning criterion. Specifically, only when the filter has the largest intra-branch distance and its compensatory counterpart has the strongest remembering enhancement power, they are preserved. Further, Ebbinghaus curve-based asymptotic forgetting is proposed to protect the pruned model from unstable learning. The number of pruned filters is increasing asymptotically in the training procedure, which enables the remembering of pretrained weights gradually to be concentrated in the remaining filters. Extensive experiments demonstrate the superiority of REAF over many state-of-the-art (SOTA) methods. For example, REAF removes 47.55% FLOPs and 42.98% parameters of ResNet-50 only with 0.98% TOP-1 accuracy loss on ImageNet. The code is available at https://github.com/zhangxin-xd/REAF.

Noise Prior Knowledge Informed Bayesian Inference Network for Hyperspectral Super-Resolution.

Well-known deep learning (DL) is widely used in fusion based hyperspectral image super-resolution (HS-SR). However, DL-based HS-SR models have been designed mostly using off-the-shelf components from current deep learning toolkits, which lead to two inherent challenges: i) they have largely ignored the prior information contained in the observed images, which may cause the output of the network to deviate from the general prior configuration; ii) they are not specifically designed for HS-SR, making it hard to intuitively understand its implementation mechanism and therefore uninterpretable. In this paper, we propose a noise prior knowledge informed Bayesian inference network for HS-SR. Instead of designing a "black-box" deep model, our proposed network, termed as BayeSR, reasonably embeds the Bayesian inference with the Gaussian noise prior assumption to the deep neural network. In particular, we first construct a Bayesian inference model with the Gaussian noise prior assumption that can be solved iteratively by the proximal gradient algorithm, and then convert each operator involved in the iterative algorithm into a specific form of network connection to construct an unfolding network. In the process of network unfolding, based on the characteristics of the noise matrix, we ingeniously convert the diagonal noise matrix operation which represents the noise variance of each band into the channel attention. As a result, the proposed BayeSR explicitly encodes the prior knowledge possessed by the observed images and considers the intrinsic generation mechanism of HS-SR through the whole network flow. Qualitative and quantitative experimental results demonstrate the superiority of the proposed BayeSR against some state-of-the-art methods.

Development and validation of a predictive nomogram for lower extremity deep vein thrombosis dislodgement in orthopedic patients.

Objective: To analyze the risk factors of lower extremity deep venous thrombosis (DVT) detachment in orthopedic patients, and to establish a risk nomogram prediction model. Methods: The clinical data of 334 patients with orthopedic DVT admitted to the Third Hospital of Hebei Medical University from January 2020 to July 2021 were retrospectively analyzed. General statistics included gender, age, BMI, thrombus detachment, inferior vena cava filter window type, filter implantation time, medical history, trauma history, operation, use of tourniquet, thrombectomy, anesthesia mode, anesthesia grade, operative position, blood loss during operation, blood transfusion, immobilization, use of anticoagulants, thrombus side, thrombus range, D-dimer content before filter implantation and during removal of inferior vena cava filter. Logistic regression was used to perform univariate and multivariate analysis on the possible factors of thrombosis detachment, screen out independent risk factors, establish a risk nomogram prediction model by variables, and internally verify the predictability and accuracy of the model. Results: Binary logistic regression analysis showed that Short time window filter (OR = 5.401, 95% CI = 2.338-12.478), lower extremity operation (OR = 3.565, 95% CI = 1.553-8.184), use of tourniquet (OR = 3.871, 95% CI = 1.733-8.651), non-strict immobilization (OR = 3.207, 95% CI = 1.387-7.413), non-standardized anticoagulation (OR = 4.406, 95% CI = 1.868-10.390), distal deep vein thrombosis (OR = 2.212, 95% CI = 1.047-4.671) were independent risk factors for lower extremity DVT detachment in orthopedic patients (P < 0.05). Based on these six factors, a prediction model for the risk of lower extremity DVT detachment in orthopedic patients was established, and the risk prediction ability of the model was verified. The C-index of the nomogram model was 0.870 (95% CI: 0.822-0.919). The results indicate that the risk nomogram model has good accuracy in predicting the loss of deep venous thrombosis in orthopedic patients. Conclusion: The nomogram risk prediction model based on six clinical factors, including filter window type, operation condition, tourniquet use, braking condition, anticoagulation condition, and thrombosis range, has good predictive performance.

An Interpretable Unsupervised Unrolling Network for Hyperspectral Pansharpening.

Existing deep convolutional neural networks (CNNs) have recently achieved great success in pansharpening. However, most deep CNN-based pansharpening models are based on "black-box" architecture and require supervision, making these methods rely heavily on the ground-truth data and lose their interpretability for specific problems during network training. This study proposes a novel interpretable unsupervised end-to-end pansharpening network, called as IU2PNet, which explicitly encodes the well-studied pansharpening observation model into an unsupervised unrolling iterative adversarial network. Specifically, we first design a pansharpening model, whose iterative process can be computed by the half-quadratic splitting algorithm. Then, the iterative steps are unfolded into a deep interpretable iterative generative dual adversarial network (iGDANet). Generator in iGDANet is interwoven by multiple deep feature pyramid denoising modules and deep interpretable convolutional reconstruction modules. In each iteration, the generator establishes an adversarial game with the spatial and spectral discriminators to update both spectral and spatial information without ground-truth images. Extensive experiments show that, compared with the state-of-the-art methods, our proposed IU2PNet exhibits very competitive performance in terms of quantitative evaluation metrics and qualitative visual effects.

Interstitial Hydrogen Atom to Boost Intrinsic Catalytic Activity of Tungsten Oxide for Hydrogen Evolution Reaction.

Tungsten oxide (WO3 ) is an appealing electrocatalyst for the hydrogen evolution reaction (HER) owing to its cost-effectiveness and structural adjustability. However, the WO3 electrocatalyst displays undesirable intrinsic activity for the HER, which originates from the strong hydrogen adsorption energy. Herein, for effective defect engineering, a hydrogen atom inserted into the interstitial lattice site of tungsten oxide (H0.23 WO3 ) is proposed to enhance the catalytic activity by adjusting the surface electronic structure and weakening the hydrogen adsorption energy. Experimentally, the H0.23 WO3 electrocatalyst is successfully prepared on reduced graphene oxide. It exhibits significantly improved electrocatalytic activity for HER, with a low overpotential of 33 mV to drive a current density of 10 mA cm-2 and ultra-long catalytic stability at high-throughput hydrogen output (200 000 s, 90 mA cm-2 ) in acidic media. Theoretically, density functional theory calculations indicate that strong interactions between interstitial hydrogen and lattice oxygen lower the electron density distributions of the d-orbitals of the active tungsten (W) centers to weaken the adsorption of hydrogen intermediates on W-sites, thereby sufficiently promoting fast desorption from the catalyst surface. This work enriches defect engineering to modulate the electron structure and provides a new pathway for the rational design of efficient catalysts for HER.

HPRN: Holistic Prior-Embedded Relation Network for Spectral Super-Resolution.

Spectral super-resolution (SSR) refers to the hyperspectral image (HSI) recovery from an RGB counterpart. Due to the one-to-many nature of the SSR problem, a single RGB image can be reprojected to many HSIs. The key to tackle this ill-posed problem is to plug into multisource prior information such as the natural spatial context prior of RGB images, deep feature prior, or inherent statistical prior of HSIs so as to effectively alleviate the degree of ill-posedness. However, most current approaches only consider the general and limited priors in their customized convolutional neural networks (CNNs), which leads to the inability to guarantee the confidence and fidelity of reconstructed spectra. In this article, we propose a novel holistic prior-embedded relation network (HPRN) to integrate comprehensive priors to regularize and optimize the solution space of SSR. Basically, the core framework is delicately assembled by several multiresidual relation blocks (MRBs) that fully facilitate the transmission and utilization of the low-frequency content prior of RGBs. Innovatively, the semantic prior of RGB inputs is introduced to mark category attributes, and a semantic-driven spatial relation module (SSRM) is invented to perform the feature aggregation of clustered similar ranges for refining recovered characteristics. In addition, we develop a transformer-based channel relation module (TCRM), which breaks the habit of employing scalars as the descriptors of channelwise relations in the previous deep feature prior and replaces them with certain vectors to make the mapping function more robust and smoother. In order to maintain the mathematical correlation and spectral consistency between hyperspectral bands, the second-order prior constraints (SOPCs) are incorporated into the loss function to guide the HSI reconstruction. Finally, extensive experimental results on four benchmarks demonstrate that our HPRN can reach the state-of-the-art performance for SSR quantitatively and qualitatively. Furthermore, the effectiveness and usefulness of the reconstructed spectra are verified by the classification results on the remote sensing dataset. Codes are available at https://github.com/Deep-imagelab/HPRN.

Filter Pruning via Learned Representation Median in the Frequency Domain.

In this article, we propose a novel filter pruning method for deep learning networks by calculating the learned representation median (RM) in frequency domain (LRMF). In contrast to the existing filter pruning methods that remove relatively unimportant filters in the spatial domain, our newly proposed approach emphasizes the removal of absolutely unimportant filters in the frequency domain. Through extensive experiments, we observed that the criterion for "relative unimportance" cannot be generalized well and that the discrete cosine transform (DCT) domain can eliminate redundancy and emphasize low-frequency representation, which is consistent with the human visual system. Based on these important observations, our LRMF calculates the learned RM in the frequency domain and removes its corresponding filter, since it is absolutely unimportant at each layer. Thanks to this, the time-consuming fine-tuning process is not required in LRMF. The results show that LRMF outperforms state-of-the-art pruning methods. For example, with ResNet110 on CIFAR-10, it achieves a 52.3% FLOPs reduction with an improvement of 0.04% in Top-1 accuracy. With VGG16 on CIFAR-100, it reduces FLOPs by 35.9% while increasing accuracy by 0.5%. On ImageNet, ResNet18 and ResNet50 are accelerated by 53.3% and 52.7% with only 1.76% and 0.8% accuracy loss, respectively. The code is based on PyTorch and is available at https://github.com/zhangxin-xd/LRMF.