Innovative Joint Research Articles

A Joint Estimation Method of Distribution Network Topology and Line Parameters Based on Power Flow Graph Convolutional Networks

Accurate identification of network topology and line parameters is essential for effective management of distribution systems. An innovative joint estimation method for distribution network topology and line parameters is presented, utilizing a power flow graph convolutional network (PFGCN). This approach addresses the limitations of traditional methods that rely on costly voltage phase angle measurements. The node correlation principle is applied to construct a node correlation matrix, and a minimum distance iteration algorithm is proposed to generate candidate topologies, which serve as graph inputs for the parameter estimation model. Based on the topological dependencies and convolutional properties of AC power flow equations, a PFGCN model is designed for line parameter estimation. Parameter refinement is achieved through an alternating iterative process of pseudo-trend calculation and neural network training. Training convergence and loss function values are used as feedback to filter and validate candidate topologies, enabling precise joint estimation of both topologies and parameters. The proposed method’s accuracy, transferability, and robustness are demonstrated through experiments on the IEEE-33 and modified IEEE-69 distribution systems. Multiple metrics, including MAPE, IAE, MAE, and R2, highlight the proposed method’s advantages over Adaptive Ridge Regression (ARR). In the C33 scenario, the proposed method achieves MAPEs of 4.6% for g and 5.7% for b, outperforming the ARR method with MAPEs of 7.1% and 7.9%, respectively. Similarly, in the IC69 scenario, the proposed method records MAPEs of 3.0% for g and 5.9% for b, surpassing the ARR method’s 5.1% and 8.3%.

Experimental study on the working mechanism and seismic performance of novel energy-dissipation boxes with different configurations

This paper presents a novel composite energy-dissipation beam-column joint that incorporates friction side plates (FSPs) and composite energy-dissipation boxes (EDBs). As replaceable components of this innovative joint, the composite EDBs significantly influence its overall seismic performance. To conduct a comprehensive study of the composite EDBs, eight specimens were fabricated, with parameters including filling material, construction gaps, and secondary loading. This investigation primarily examined the failure modes, hysteretic performance, strength degradation, stiffness degradation, and energy-dissipation capacity of the composite EDBs. The results indicated that increasing the construction gap can delay the buckling of dog-bone energy-dissipation plates (DB-EDPs), while secondary loading enhanced both the strength and stiffness of the specimens. Additionally, the filling material effectively restrained the out-of-plane buckling of the DB-EDPs, thereby improving the load-bearing capacity of the composite EDBs. The effectiveness of the filling materials was ranked as follows: wood > steel box > rubber. Finally, a theoretical load-bearing capacity formula for the specimens was derived, and the applicability was validated by comparing the experimental values with the theoretical predictions.

Antibody-SGM, a Score-Based Generative Model for Antibody Heavy-Chain Design.

Traditional computational methods for antibody design involved random mutagenesis followed by energy function assessment for candidate selection. Recently, diffusion models have garnered considerable attention as cutting-edge generative models, lauded for their remarkable performance. However, these methods often focus solely on the backbone or sequence, resulting in the incomplete depiction of the overall structure and necessitating additional techniques to predict the missing component. This study presents Antibody-SGM, an innovative joint structure-sequence diffusion model that addresses the limitations of existing protein backbone generation models. Unlike previous models, Antibody-SGM successfully integrates sequence-specific attributes and functional properties into the generation process. Our methodology generates full-atom native-like antibody heavy chains by refining the generation to create valid pairs of sequences and structures, starting with random sequences and structural properties. The versatility of our method is demonstrated through various applications, including the design of full-atom antibodies, antigen-specific CDR design, antibody heavy chains optimization, validation with Alphafold3, and the identification of crucial antibody sequences and structural features. Antibody-SGM also optimizes protein function through active inpainting learning, allowing simultaneous sequence and structure optimization. These improvements demonstrate the promise of our strategy for protein engineering and significantly increase the power of protein design models.

A deep learning-based approach for predicting coking duration in a commercial coke oven

ABSTRACT The prediction of coking duration is critical during operation of commercial coke oven batteries; however, it is still far from accurately predict since it has to rely on experience-based estimation. To address this problem, we proposed a machine learning-based approach, and presented a general practice and theoretical framework for predicting coking duration accurately and efficiently. Three types of neutral networks, including the multi-layer perceptron (MLP) model, the radial basis function (RBF) model, and the long short-term memory (LSTM) model, were tested and evaluated in terms of predicting accuracy, training and predicting time. On this basis, a novel two-stream joint MLP-RBF neural network was proposed. All the networks were trained and tested with onsite-collected datasets. In the modeling, the potential relationships between variables are deeply learnt automatically. The results show that the innovative joint network inherits advantages from both sides and exhibits the lowest prediction error, 2.79% (MAPE) or 4.26% (RMSPE), and is recommended as the most appropriate for practical application. This work not only presents an efficient approach for predicting coking duration, but also examines the extent to which the machine learning technology can be employed during the digital transformation of traditional industries.

Deformation ability of steel inner sleeve T-joint in modular gymnasia based on 3D-DIC method

Substantial loads the roofs bear exert stringent performance standards on connecting joints in movable modular gymnasiums. Previously, we introduced a novel joint design, utilizing internal sleeves and bolts to integrate the upper and lower columns of modular units. To explore the actual deformation behavior and force distribution of this joint, this study employed a three-dimensional digital image correlation (3D-DIC) measurement method to visualize its full-field displacement and strain distribution. The reliability of the 3D-DIC results was confirmed through comparison with traditional measurement methods. Building upon this solid foundation, this study proceeds to assess the static performance of this innovative joint in two distinct directions, successfully revealing micro-deformation and local buckling phenomena undetectable to the naked eye. Notably, the quality of the welds connecting the beams and columns plays a pivotal role in ensuring the overall structural integrity of the joint. Under lateral loading conditions, the destructive behavior of the joints is primarily governed by material strength. Moment-rotation curves were plotted to delve further into the mechanical characteristics of the joint, revealing that the inner-sleeve T-joint exhibits the attributes of a rigid joint. Moreover, the primary damage modes observed in the specimens were fractures at the beam root or buckling of the column, with the joint area remaining intact. Consequently, this joint qualifies as a full-strength connection. These findings will deepen knowledge and understanding of this novel joint for designers.

Validating a finite element model for rigid aircraft pavement load transfer against full scale testing

ABSTRACT Rigid aircraft pavements are generally comprised of unreinforced concrete slabs constructed on a bound or granular sub-base layer, over the natural or constructed subgrade or fill. Important to any rigid aircraft pavement design and construction are the joints between the concrete slabs. The joints control shrinkage cracks during curing, allow for thermal expansion and contraction during temperature cycles, they isolate concrete slabs from structural penetrations and importantly, provide load transfer between adjacent slabs. Load transfer is commonly modelled using finite element methods; however, any new model should be validated against observed physical events. Recently, full-scale testing was performed at the Federal Aviation Authority National Airport Pavement Testing Facility as part of Construction Cycle 8 (CC8), which investigated load transfer characteristics for a range of joint types. The results from CC8 tests provide up-to-date stress and strain values to validate any new finite element model. This paper presents the development of a finite element model to predict load transfer of a doweled construction joint, and its validation against CC8 test results. The model showed good agreement with CC8 test results, and now that it is validated, will be used to investigate innovative joint solutions for rigid aircraft pavements.

Geographical variation in dementia prevalence across China: a geospatial analysis

Dementia poses great health and social challenges in China. Dementia prevalence may vary across geographic areas, while comparable estimations on provincial level is lacking. This study aims to estimate dementia prevalence by provinces across China, taking into account risk factors of individual level and potential spatial correlation of provinces. In this study, 17,176 adults aged 50 years or older were included from the fourth wave of the ChinaHealth and Retirement Longitudinal Study (CHARLS 2018), covering 28 provinces, autonomous regions and municipalities. To improve provincial representativeness, we constructed provincial survey weights based on China 7th census (2020). The prevalence of dementia and 95% Bayesian credible intervals (BCIs) were estimated using a Bayesian conditional autoregressive (CAR) model with spatially varying coefficients of covariates. The weighted prevalence of dementia at provincial level in China in 2018 ranged from 2.62% (95%BCI: 1.70%, 3.91%) to 13.53% (95%BCI: 8.82%, 20.93%). High dementia prevalence was concentrated in North China, with a prominent high-high cluster, while provinces of low prevalence were concentrated on East and South China, characterized by a low-low cluster. Ordered by the median estimation of prevalence, the top 10% of provinces, include Xinjiang, Jilin, and Beijing. Meanwhile, Fujian, Zhejiang, and Guangdong rank among the last. The association between dementia prevalence and drinking, smoking, social isolation, physical inactivity, hearing impairment, hypertension, and diabetes exhibits provincial variation. Our study identifies a geospatial disparity in dementia prevalence and risk factor effects across China's provinces, with high-high and low-low clusters in some northern and southern provinces, respectively. The findings emphasize the need for targeted strategies, such as addressing hypertension and hearing impairment, in specific regions for more effective dementia prevention and treatment. National Science Foundation of China/the Economic and Social Research Council, UK Research and Innovation joint call: Understanding and Addressing Health and Social Challenges for Ageing in the UK and China. UK-China Health And Social Challenges Ageing Project (UKCHASCAP): present and future burden of dementia, and policy responses (grant number 72061137003, ES/T014377/1).

Calculation of the energy dependence of fission fragments yields and kinetic energy distributions for neutron-induced 235U fission* *Supported by the National Natural Science Foundation of China (12075105), the NSFC-Nuclear Technology Innovation Joint Fund (U2167203), the Major Science and Technology Projects of Gansu Province (22ZD6GB020), the Fundamental Research Funds for the Central Universities of China (lzujbky-2022-ey14, lzujbky-2022- kb07)

Fission fragments yields and average total kinetic energy are fundamental nuclear data for nuclear energy applications and the study of nuclear devices. Certain fission products, such as 95Zr, 99Mo, 140Ba, 144Ce, and 147Nd, serve as burnup monitors, assessing the number of fissions induced by neutrons on 235U. However, current experimental data for these fission products worldwide are inconsistent, introducing significant uncertainty into related scientific research. In this study, we employed the Potential-driving Model to calculate the independent yields of 235U and evaluate its advantages in such calculations. Additionally, we investigated the energy dependence of independent yields to select important products. Furthermore, we calculated the cumulative yields of 95Zr, 99Mo, 140Ba, 144Ce, and 147Nd, and compared them with existing literature data to explore the energy dependence of fission products for 235U. Given the lack of fission product yield data above 14.8 MeV, we extended our calculated incident neutron energy to 20 MeV, aiming to support future scientific research. The Geant4 physical model does not consider the influence of incident neutron energy on the average total kinetic energy of fission fragments; thus, we introduced the excitation function of the total kinetic energy of fission fragments recommended by Madland et al., which effectively describes the experimental data of the average total kinetic energy of fragments formed in 235U fission. In this paper, we comprehensively discuss the energy dependence of fission product yields and average total kinetic energy.

Seismic performance of the WCFTC-beams joint with high-strength double-side plates

This research presents a novel joint design for wall-type concrete filled steel tube (WCFTC) beams, integrated with high-strength double-side plates, resolving the problem of traditional side plate extending from the wall and interfering with wall panel installation. The study synthesizes a reliable finite element (FE) model corroborated by previous experimental analyses, enabling a detailed parametric study on the dimensions of the side plates. Comprehensive analysis assessed the impact of steel plate thickness and material strength on the joint's seismic properties, uncovering that the innovative joint design reduces steel usage while maintaining load capacity. Subsequent full-scale testing of prototypes under cyclic loading revealed the failure modes and hysteresis responses, which exhibited notable energy dissipation reflected in ductility coefficients exceeding 3.0. The study also identifies the plastic hinge location to be at a distance ranging from 1.0 to 1.5 times the beam's height from the column's edge, validating the efficacy of the joint design categorized as full-strength or rigid. Furthermore, comparative assessments of Q690D and Q460C steel plates indicated similar performance, suggesting a flexibility in high-strength material choice without loss of joint integrity. The culmination of the study is the proposal of a clear and actionable design method, predicated on the synthesized findings, offering a pragmatic reference for designers and engineers in the field. This method aims to streamline the adoption of this advanced joint construction for enhanced seismic resilience in structural applications.

Joint planning for PV-SESS-MESS in distribution network towards 100% self-consumption of PV via consensus-based ADMM

The integration of distributed photovoltaic (PV) has become a prominent characteristic of distribution network (DN). However, achieving 100% PV self-consumption poses challenges due to intermittent generation and varying load demands. To address these challenges, this paper proposes an innovative joint planning method for PV, stationary energy storage system (SESS), and mobile energy storage system (MESS) using the consensus-based alternating directional multiplier method (ADMM). First, to account for multiple scenarios with real-world variations, a two-stage clustering method with clustering by fast search and finding of density peaks and k-means is employed. Second, a joint planning model for PV, SESS, and MESS is developed to minimize the total cost of DN, including installation, operation, and maintenance costs. The model optimizes the PV capacity, SESS capacity, and MESS capacity, as well as their locations. Third, a distributed planning framework is proposed using consensus-based ADMM, which decomposes the problem into a PV planning subproblem under the worst-case scenario and energy storage system planning subproblems under multiple scenarios. Finally, the proposed method is demonstrated on an IEEE-33 node system, and the results show the method effectively enhances the access capacity of PV by 11.80% and leads to a reduction in the total cost.

Rehabilitation with brain-computer interface and upper limb motor function in ischemic stroke: A randomized controlled trial

Upper limb motor dysfunction is a major problem in the rehabilitation of patients with stroke. Brain-computer interface (BCI) is a kind of communication system that converts the "ideas" in the brain into instructions and has been used in stroke rehabilitation. This study aimed to investigate the efficacy and safety of BCI in rehabilitation training on upper limb motor function among patients with ischemic stroke. This was an investigator-initiated, multicenter, randomized, open-label, blank-controlled clinical trial with blinded outcome assessment conducted at 17 centers in China. Patients were assigned in a 1:1 ratio to the BCI group or the control group based on traditional rehabilitation training. The primary efficacy outcome is the difference in improvement of the Fugl-Meyer Assessment upper extremity (FMA-UE) score between two groups at month 1 after randomization. The safety outcomes were any adverse events within 3months. A total of 296 patients with ischemic stroke were enrolled and randomly allocated to the BCI group (n= 150) and the control group (n= 146). The primary efficacy outcomes of FMA-UE score change from baseline to 1month were 13.17 (95% confidence interval [CI], 11.56-14.79) in the BCI group and 9.83 (95% CI, 8.19-11.47) in the control group (mean difference between groups was 3.35; 95% CI, 1.05-5.65; p= 0.0045). Adverse events occurred in 33 patients (22.00%) in the BCI group and in 31 patients (21.23%) in the control group. BCI rehabilitation training can further improve upper limb motor function based on traditional rehabilitation training in patients with ischemic stroke. This study was registered at ClinicalTrials.gov: NCT04387474. This work was supported by the National Key R&D Program of China (2018YFC1312903), the National Key Research and Development Program of China (2022YFC3600600), the Training Fund for Open Projects at Clinical Institutes and Departments of Capital Medical University (CCMU2022ZKYXZ009), the Beijing Natural Science Foundation Haidian original innovation joint fund (L222123), the Fund for Young Talents of Beijing Medical Management Center (QML20230505), and the high-level public health talents (xuekegugan-02-47).

Hybrid joints consisting of pre-tensioned bolts and a bonded connection—The influence of adhesives on the load-bearing capacity

This study explores the intricate mechanics of hybrid joints, which combine prestressed bolts and adhesive bonding, with a focus on increasing the joint capacity of steel structures. Despite their potential, the fundamental behaviour of these joints is not fully understood. The research focuses on unravelling the load-bearing capacity of these hybrid joints, emphasising the influence of the adhesive properties. The unique manufacturing process, involving the pre-tensioning of bolts while the adhesive is uncured, introduces additional dependencies not present in conventional adhesive joints. Analysis of the mechanical and physical properties of five adhesives provides key insights into the mechanics of load transfer and failure modes in hybrid joints, providing a comprehensive understanding. The study shows that load transfer occurs through micron-thin adhesive layers, highlighting the significant influence of adhesive properties on joint performance. Epoxies emerge as the top performers, with 2.5 times the average friction joint and more than five times the average bonded joint capacity, indicating their superiority in hybrid joint applications. Hybrid joints show significantly less variation than adhesive and friction joints, underlining their consistency and reliability in structural applications. When comparing hybrid and bonded joints with the same adhesive (SW7240), the variance was suppressed to one seventh by the application of preloaded bolts, with practical implications for design values based on quantiles. The study establishes correlations between joint strength and adhesive properties, emphasising the critical role of Young's modulus and tensile strength. It also finds that the elongation of hybrid joints at break is almost entirely dependent on load capacity, with viscosity having a lesser effect. Fracture surface investigations provide insight into the failure modes of hybrid joints, highlighting differences from those observed in frictional or purely bonded joints. This comprehensive study, combining experimental campaigns and in-depth analysis, provides valuable insights into the intricate relationships between joint capacity and adhesive properties in hybrid joints. The findings guide the selection of appropriate adhesives and provide the basis for the development of accurate load prediction methods, unlocking the full potential of hybrid joints and overcoming the limitations of conventional joining techniques in steel structures. Further research is encouraged to investigate the influence of adhesive class, post-tensioning stability and the nuanced mechanics of load transfer in these innovative joints.

Innovative Joint CFO and Channel Estimation for RIS-Aided OFDM Systems

Innovative Joint CFO and Channel Estimation for RIS-Aided OFDM Systems

A novel joint extraction model based on cross-attention mechanism and global pointer using context shield window

Relational triple extraction is a critical step in knowledge graph construction. Compared to pipeline-based extraction, joint extraction is gaining more attention because it can better utilize entity and relation information without causing error propagation issues. Yet, the challenge with joint extraction lies in handling overlapping triples. Existing approaches adopt sequential steps or multiple modules, which often accumulate errors and interfere with redundant data. In this study, we propose an innovative joint extraction model with cross-attention mechanism and global pointers with context shield window. Specifically, our methodology begins by inputting text data into a pre-trained RoBERTa model to generate word vector representations. Subsequently, these embeddings are passed through a modified cross-attention layer along with entity type embeddings to address missing entity type information. Next, we employ the global pointer to transform the extraction problem into a quintuple extraction problem, which skillfully solves the issue of overlapping triples. It is worth mentioning that we design a context shield window on the global pointer, which facilitates the identification of correct entities within a limited range during the entity extraction process. Finally, the capability of our model against malicious samples is improved by adding adversarial training during the training process. Demonstrating superiority over mainstream models, our approach achieves impressive results on three publicly available datasets.

Seismic performance and design of an innovative dual vertical diaphragms joint between flat CFST column and H-beam

Steel-concrete composite structure is one of the research hotspots in the structural engineering field in recent years. Among the various types of composite structures, flat concrete-filled steel tube (CFST) column structures are preferred for residential buildings owing to the smooth elevation to coordinate with the indoor decoration. The seismic performance of beam-to-column joints is momentous as a key part of flat CFST column structures. This study provided a new type of joint between flat CFST column and H-beam with dual vertical diaphragms through the column. The design procedure for the vertical diaphragms was proposed to design the test specimens. An experimental study on the seismic behavior of the innovative joint under cyclic load was performed on three full-scale specimens. The seismic performance of the joint, including the failure mode, hysteretic and skeleton curves, ductility and energy dissipation capacity, was observed and discussed in detail. Subsequently, nonlinear finite element models (FEMs), based on the specimens, were developed using the ABAQUS/Standard program and validated by comparing the numerical results with the test ones. The experimental and numerical results demonstrated that the proposed joint exhibited satisfactory seismic performance and all the specimens exhibited a ductile failure mode. The failure process and damage distribution of the joint under earthquake were revealed based on the validated FEM. Furthermore, a mechanical model was established to predict the load-bearing capacity of the shear force and bending moment of the proposed joint. The mechanical model was validated to be accurate and reliable by comparing the numerical simulation results with the theoretical ones. This study provided a practical joint between flat CFST column and H-beam as well as its design method, which was beneficial for the promotion of the flat CFST column structures.

A Combination Model of Shifting Joint Angle Changes With 3D-Deep Convolutional Neural Network to Recognize Human Activity.

Research in the field of human activity recognition is very interesting due to its potential for various applications such as in the field of medical rehabilitation. The need to advance its development has become increasingly necessary to enable efficient detection and response to a wide range of movements. Current recognition methods rely on calculating changes in joint distance to classify activity patterns. Therefore, a different approach is required to identify the direction of movement to distinguish activities exhibiting similar joint distance changes but differing motion directions, such as sitting and standing. The research conducted in this study focused on determining the direction of movement using an innovative joint angle shift approach. By analyzing the joint angle shift value between specific joints and reference points in the sequence of activity frames, the research enabled the detection of variations in activity direction. The joint angle shift method was combined with a Deep Convolutional Neural Network (DCNN) model to classify 3D datasets encompassing spatial-temporal information from RGB-D video image data. Model performance was evaluated using the confusion matrix. The results show that the model successfully classified nine activities in the Florence 3D Actions dataset, including sitting and standing, obtaining an accuracy of (96.72 ± 0.83)%. In addition, to evaluate its robustness, this model was tested on the UTKinect Action3D dataset, obtaining an accuracy of 97.44%, proving that state-of-the-art performance has been achieved.

Development of a Joint Concept for Producing Dissimilar Joints Using a 3D Printing-Supported Technique

The joining of metals and polymer-based materials has a very high interest for many industrial sectors, as it allows to achieve components combining the specific characteristics of each material class. Additive manufacturing technologies could boost the production of these joints, allowing the controlled deposition of a polymeric material over the metal substrate. The present research is aimed to study the feasibility of a joint concept that can be used to produce aluminium/polymer-based material joints through a 3D printing-supported technique. The innovative joint concept, which is based on an interlocking mechanism promoted by a deposited pin, was compared to two conventional concepts. The innovative joint concept allows the production of samples with good mechanical behaviour, in which the failure occurs outside the material overlapping zone. This design is very suitable to be tested for the production of dissimilar material joints.

Effects of remnant cholesterol on the efficacy of genotype-guided dual antiplatelet in CYP2C19 loss-of -function carriers with minor stroke or transient ischaemic attack: a post-hoc analysis of the CHANCE-2 trial

The atherogenicity of remnant cholesterol (RC), a contributor to residual risk of cardiovascular events, has been underlined by recent guidelines. We aimed to evaluate the relationship between RC levels and the efficacy and safety of genotype-guided dual antiplatelet therapy in the CHANCE-2 trial. This post-hoc study used data from the CHANCE-2 trial, which was a randomised, double-blind, placebo-controlled trial of 6412 patients (aged >40 years) enrolled from 202 hospitals in China, between Sept 23, 2019, and March 22, 2021, who carried CYP2C19 loss-of-function alleles and had either an acute minor stroke or high-risk transient ischaemic attack (TIA), and could start treatment within 24h of symptom onset. Participants received either (1:1) ticagrelor plus aspirin (control) or clopidogrel plus aspirin (intervention) and the effect of reducing the 3-month risk of any new stroke was assessed (ischemic or haemorrhagic, primary outcome). From the CHANCE-2 study population, we enrolled 5890 patients with complete data on RC. The cutoff point of RC for distinguishing patients with greater benefit from ticagrelor-aspirin versus clopidogrel-aspirin was determined with subpopulation treatment effect pattern plot. The primary efficacy and safety outcome was recurrent stroke and severe or moderate bleeding within 90 days, respectively. CHANCE-2 is registered with ClinicalTrials.gov, NCT04078737. The cutoff to define elevated RC was 0.91mmol/L. Ticagrelor-aspirin versus clopidogrel-aspirin was associated with a reduced risk of recurrent stroke in patients with non-elevated RC levels (122 [5.3%] versus 179 [7.8%]; hazard ratio [HR], 0.68; 95% confidence interval [CI], 0.54-0.85), but this benefit was absent in those with elevated RC levels (58 [8.4%] versus 48 [7.3%]; HR, 1.10; 95% CI, 0.73-1.65; P-interaction=0.03). When analyzed as a continuous variable, the benefit of ticagrelor-aspirin on recurrent stroke decreased as RC levels increased. The rates of severe or moderate bleeding between treatment groups were similar across RC categories (0.3% versus 0.3%, P-interaction=0.95). Our post-hoc findings suggest that RC could be a potential biomarker to discriminate patients who received more benefits from ticagrelor-aspirin versus clopidogrel-aspirin therapy in CYP2C19 loss-of-function carriers with minor stroke or TIA. These findings need to be validated in an independent study. The National Key Research and Development Program of China, Beijing Natural Science Foundation Haidian original innovation joint fund, Fund for Young Talents of Beijing Medical Management Center, the high-level public health talents, Training Fund for Open Projects at Clinical Institutes and Departments of Capital Medical University; and Salubris.

Asymmetric convolution Swin transformer for medical image super-resolution

Medical Image Super-Resolution plays a pivotal role in enhancing diagnostic accuracy. Transformer-based methods, such as Image Restoration Using Swin Transformer (SwinIR) and Swin transformer for fast Magnetic Resonance Imaging (SwinMR), have shown prowess in this area but also exhibit limitations. Specifically, LayerNorm channel normalization diminishes high-frequency detail, while the Multilayer Perceptron prioritizes global information over local information. Moreover, low-resolution inputs contain substantial low-frequency information, treated uniformly in self-attention, hampering the effectiveness of shifted window-based self-attention. To address these challenges, this study proposes a novel Asymmetric convolution Swin Transformer Layer that leverages global and local information within adjacent windows or pixels. Furthermore, this study presents an innovative joint attention mechanism that integrates channel attention into the Swin Transformer architecture, allowing for the simultaneous capture of local and global information and enhancing the network's ability to adapt and represent information effectively. Based on these components, this study develops a joint Channel Attention and Swin Transformer Residual Network (CSRNet) for medical image super-resolution. To evaluate its performance, this study conducts comprehensive experiments on established datasets consisting of high-resolution medical images. Compared to the state-of-the-art medical image super-resolution methods, the proposed method outperforms in reconstructing medical images across qualitative and quantitative metrics.

Ensemble data assimilation for open channel water level forecasting under flow disturbances: A synthetic data study

Accurate water level forecasts are essential for effective river management. Data assimilation (DA), involving real-time model correction through observational data, serves as a pivotal approach to enhance the predictive precision of model-based state forecasting. Nevertheless, prevailing DA methodologies encounter constraints in swiftly and accurately forecasting water levels amidst substantial uncertainties stemming from flow disturbances. To surmount this challenge, an innovative joint ensemble Kalman filter (EnKF) framework that facilitates concurrent estimation of flow disturbances and water levels is proposed. By incorporating the flow disturbance state variable into the EnKF framework for water level assimilation, a joint EnKF framework that effectively estimates flow disturbances at known disturbance points through assimilation techniques is established. Additionally, for scenarios where the precise location of the flow disturbance point remains unknown, the introduction of a fictitious lateral outflow in proximity to the observation point allows estimation of the flow disturbance process of this fictitious outflow. Synthetic case study results demonstrate that the incorporation of flow disturbance assimilation yields superior water level estimations compared to scenarios disregarding flow disturbances, regardless of the concordance between estimated and actual disturbance flow values. Furthermore, even when the location of the disturbance point is known, utilization of a fictitious lateral outflow and and subsequent flow correction exhibits heightened robustness and stability in comparison to direct correction of the disturbance point's flow. Therefore, in practical applications, utilizing a fictitious lateral outflow and flow correction within the joint EnKF framework emerges as a more suitable approach, irrespective of the knowledge of disturbance location. The proposed method constitutes a substantial contribution to the field of water level forecasting in river systems under conditions of uncertain flow disturbances.