CMRxRecon: A publicly available k-space dataset and benchmark to advance deep learning for cardiac MRI.

Cardiac magnetic resonance imaging (CMR) has emerged as a valuable diagnostic tool for cardiac diseases. However, a significant drawback of CMR is its slow imaging speed, resulting in low patient throughput and compromised clinical diagnostic quality. The limited temporal resolution also causes patient discomfort and introduces artifacts in the images, further diminishing their overall quality and diagnostic value. There has been growing interest in deep learning-based CMR imaging algorithms that can reconstruct high-quality images from highly under-sampled k-space data. However, the development of deep learning methods requires large training datasets, which have so far not been made publicly available for CMR. To address this gap, we released a dataset that includes multi-contrast, multi-view, multi-slice and multi-coil CMR imaging data from 300 subjects. Imaging studies include cardiac cine and mapping sequences. The 'CMRxRecon' dataset contains raw k-space data and auto-calibration lines. Our aim is to facilitate the advancement of state-of-the-art CMR image reconstruction by introducing standardized evaluation criteria and making the dataset freely accessible to the research community.

STADNet: Spatial-Temporal Attention-Guided Dual-Path Network for cardiac cine MRI super-resolution.

Cardiac cine magnetic resonance imaging (MRI) is a commonly used clinical tool for evaluating cardiac function and morphology. However, its diagnostic accuracy may be compromised by the low spatial resolution. Current methods for cine MRI super-resolution reconstruction still have limitations. They typically rely on 3D convolutional neural networks or recurrent neural networks, which may not effectively capture long-range or non-local features due to their limited receptive fields. Optical flow estimators are also commonly used to align neighboring frames, which may cause information loss and inaccurate motion estimation. Additionally, pre-warping strategies may involve interpolation, leading to potential loss of texture details and complicated anatomical structures. To overcome these challenges, we propose a novel Spatial-Temporal Attention-Guided Dual-Path Network (STADNet) for cardiac cine MRI super-resolution. We utilize transformers to model long-range dependencies in cardiac cine MR images and design a cross-frame attention module in the location-aware spatial path, which enhances the spatial details of the current frame by using complementary information from neighboring frames. We also introduce a recurrent flow-enhanced attention module in the motion-aware temporal path that exploits the correlation between cine MRI frames and extracts the motion information of the heart. Experimental results demonstrate that STADNet outperforms SOTA approaches and has significant potential for clinical practice.

Iron nanoparticles surface decorated MXene via molten salts etching as selenium host for ultrafast sodium ion storage.

Sodium-selenium (Na-Se) batteries have gained attention due to their high energy density and power density, resulting from the liquid-liquid reaction at the interface in the dimethoxyethane electrolyte. Nevertheless, the pronounced shuttle effect of polyselenides causes low coulomb efficiency and inadequate cycling stability for Na-Se batteries. Herein, the iron nanoparticles surface modified accordion-like Ti3C2Tx MXene (MXene/Fe) synthesized via the molten salt etching is utilized as the host of Se species for high-performance Na-Se battery cathode. Benefiting from the layered structure and chemical adsorption of accordion-like MXene, the shuttle effect of the cathode is effectively inhibited. Simultaneously, electrochemical kinetics is boosted due to the catalytic effect of Fe nanoparticles, which facilitate the transformation of polyselenide from long-chain to short-chain, contributing to pseudocapacitive capacity. Consequently, the Se-based cathode delivers a steady capacity of 575.0 mA h g-1 at 0.2 A/g, and even a high capacity of 500 mAh/g at 50 A/g based on the mass of Se@MXene/Fe electrode, indicating the ultrafast Na+ ion storage. Most notably, this structure demonstrated remarkable long-term cycling stability for 5000 cycles with a high capacity retention of 97.4 %. The electrochemical energy storage mechanism is further revealed by in situ Raman. Herein, the confinement-catalysis structure shines light on inhibiting shuttling and facilitating ultrafast ion storage.

MoP quantum dots based multifunctional efficient electrocatalyst for stable and long-life flexible lithium-sulfur batteries.

The commercialization of lithium-sulfur (Li-S) batteries is challenging, owing to factors like the poor conductivity of S, the 'shuttle effect', and the slow reaction kinetics. To address these challenges, MoP quantum dots were decorated on hollow carbon spheres (MoPQDs/C) in this study and used as an efficient lithium polysulfides (LiPSs) adsorbents and catalysts. In this approach polysulfides are effectively trapped through strong chemisorption and physical adsorption while simultaneously facilitating LiPSs conversion by enhancing the reaction kinetics. MXene serves as a flexible physical barrier (MoPQDs/C@MXene), further enhancing the confinement of LiPSs. Moreover, both materials are conductive, significantly facilitating electron and charge transfer. Additionally, the flexible MoPQDs/C@MXene-S electrode offers a large specific surface area for sulfur loading and withstand volume expansion during electrochemical processes. As a result, the MoPQDs/C@MXene-S electrode exhibits excellent long-term cyclability and maintains a robust specific capacity of 992 mA h g-1 even after 800cycles at a rate of 1.0C (1C = 1675 mA g-1), with a minimal capacity decay rate of 0.034 % per cycle. This work proposes an efficient strategy to fabricate highly efficient electrocatalysts for advanced Li-S batteries.

Adaptive channel-modulated personalized federated learning for magnetic resonance image reconstruction.

Magnetic resonance imaging (MRI) is extensively utilized in clinical practice for diagnostic purposes, owing to its non-invasive nature and remarkable ability to provide detailed characterization of soft tissues. However, its drawback lies in the prolonged scanning time. To accelerate MR imaging, how to reconstruct MR images from under-sampled data quickly and accurately has drawn intensive research interest; it, however, remains a challenging task. While some deep learning models have achieved promising performance in MRI reconstruction, these models usually require a substantial quantity of paired data for training, which proves challenging to gather and share owing to high scanning costs and data privacy concerns. Federated learning (FL) is a potential tool to alleviate these difficulties. It enables multiple clinical clients to collaboratively train a global model without compromising privacy. However, it is extremely challenging to fit a single model to diverse data distributions of different clients. Moreover, existing FL algorithms treat the features of each channel equally, lacking discriminative learning ability across feature channels, and hence hindering their representational capability. In this study, we propose a novel Adaptive Channel-Modulated Federal learning framework for personalized MRI reconstruction, dubbed as ACM-FedMRI. Specifically, considering each local client may focus on features in different channels, we first design a client-specific hypernetwork to guide the channel selection operation in order to optimize the extracted features. Additionally, we introduce a performance-based channel decoupling scheme, which dynamically separates the global model at the channel level to facilitate personalized adjustments based on the performance of individual clients. This approach eliminates the need for heuristic design of specific personalization layers. Extensive experiments on four datasets under two different settings show that our ACM-FedMRI achieves outstanding results compared to other cutting-edge federated learning techniques in the field of MRI reconstruction.

Meta-Learning-Based Degradation Representation for Blind Super-Resolution.

Blind image super-resolution (blind SR) aims to generate high-resolution (HR) images from low-resolution (LR) input images with unknown degradations. To enhance the performance of SR, the majority of blind SR methods introduce an explicit degradation estimator, which helps the SR model adjust to unknown degradation scenarios. Unfortunately, it is impractical to provide concrete labels for the multiple combinations of degradations (e. g., blurring, noise, or JPEG compression) to guide the training of the degradation estimator. Moreover, the special designs for certain degradations hinder the models from being generalized for dealing with other degradations. Thus, it is imperative to devise an implicit degradation estimator that can extract discriminative degradation representations for all types of degradations without requiring the supervision of degradation ground-truth. To this end, we propose a Meta-Learning based Region Degradation Aware SR Network (MRDA), including Meta-Learning Network (MLN), Degradation Extraction Network (DEN), and Region Degradation Aware SR Network (RDAN). To handle the lack of ground-truth degradation, we use the MLN to rapidly adapt to the specific complex degradation after several iterations and extract implicit degradation information. Subsequently, a teacher network MRDAT is designed to further utilize the degradation information extracted by MLN for SR. However, MLN requires iterating on paired LR and HR images, which is unavailable in the inference phase. Therefore, we adopt knowledge distillation (KD) to make the student network learn to directly extract the same implicit degradation representation (IDR) as the teacher from LR images. Furthermore, we introduce an RDAN module that is capable of discerning regional degradations, allowing IDR to adaptively influence various texture patterns. Extensive experiments under classic and real-world degradation settings show that MRDA achieves SOTA performance and can generalize to various degradation processes.

STDAN: Deformable Attention Network for Space-Time Video Super-Resolution.

The target of space-time video super-resolution (STVSR) is to increase the spatial-temporal resolution of low-resolution (LR) and low-frame-rate (LFR) videos. Recent approaches based on deep learning have made significant improvements, but most of them only use two adjacent frames, that is, short-term features, to synthesize the missing frame embedding, which cannot fully explore the information flow of consecutive input LR frames. In addition, existing STVSR models hardly exploit the temporal contexts explicitly to assist high-resolution (HR) frame reconstruction. To address these issues, in this article, we propose a deformable attention network called STDAN for STVSR. First, we devise a long short-term feature interpolation (LSTFI) module that is capable of excavating abundant content from more neighboring input frames for the interpolation process through a bidirectional recurrent neural network (RNN) structure. Second, we put forward a spatial-temporal deformable feature aggregation (STDFA) module, in which spatial and temporal contexts in dynamic video frames are adaptively captured and aggregated to enhance SR reconstruction. Experimental results on several datasets demonstrate that our approach outperforms state-of-the-art STVSR methods. The code is available at https://github.com/littlewhitesea/STDAN.

Residual Dense Network for Image Restoration.

Recently, deep convolutional neural network (CNN) has achieved great success for image restoration (IR) and provided hierarchical features at the same time. However, most deep CNN based IR models do not make full use of the hierarchical features from the original low-quality images; thereby, resulting in relatively-low performance. In this work, we propose a novel and efficient residual dense network (RDN) to address this problem in IR, by making a better tradeoff between efficiency and effectiveness in exploiting the hierarchical features from all the convolutional layers. Specifically, we propose residual dense block (RDB) to extract abundant local features via densely connected convolutional layers. RDB further allows direct connections from the state of preceding RDB to all the layers of current RDB, leading to a contiguous memory mechanism. To adaptively learn more effective features from preceding and current local features and stabilize the training of wider network, we proposed local feature fusion in RDB. After fully obtaining dense local features, we use global feature fusion to jointly and adaptively learn global hierarchical features in a holistic way. We demonstrate the effectiveness of RDN with several representative IR applications, single image super-resolution, Gaussian image denoising, image compression artifact reduction, and image deblurring. Experiments on benchmark and real-world datasets show that our RDN achieves favorable performance against state-of-the-art methods for each IR task quantitatively and visually.

Ultrathin δ-MnO2 nanoflakes with Na+ intercalation as a high-capacity cathode for aqueous zinc-ion batteries.

Pristine δ-MnO2 as the typical cathode for rechargeable zinc-ion batteries (ZIBs) suffers from sluggish reaction kinetics, which is the key issue to prepare high-performance manganese-based materials. In this work, Na+ incorporated into layered δ-MnO2 (NMO) was prepared for ZIB cathodes with high capacity, high energy density, and excellent durable stability. By an effective fabricated strategy of hydrothermal synthesis, a three-dimensional interconnected δ-MnO2 nanoflake network with Na+ intercalation showed a uniform array arrangement and high conductivity. Also, the H+ insertion contribution in the NMO cathode to the discharge capacity confirmed the fast electrochemical charge transfer kinetics due to the enhanced ion conductivity from the insertion of Na+ into the interlayers of the host material. Consequently, a neutral aqueous NMO-based ZIB revealed a superior reversible capacity of 335 mA h g-1, and an impressive durability over 1000 cycles, and a peak gravimetric energy output of 459 W h kg-1. As a proof of concept, the as-fabricated quasi-solid-state ZIB exhibited a remarkable capacity of 284 mA h g-1 at a current density of 0.5 A g-1, and good practicability. This research demonstrated a significant enhancement of the electrochemical performance of MnO2-based ZIBs by the intercalation of Na+ to regulate the microstructure and boost the electrochemical kinetics of the δ-MnO2 cathode, thus providing a new insight for high-performance aqueous ZIBs.

Porous N-doped carbon nanoflakes supported hybridized SnO2/Co3O4 nanocomposites as high-performance anode for lithium-ion batteries.

Alloy-/conversion-type metal oxides usually exhibit high theoretical lithium storage capacities but suffer from the large volume change induced electrode pulverization and the poor electric conductivity, which limit their practical applications. Hybrid/mixed metal oxides with different working mechanisms/potentials can display advantageous synergistic enhancement effect if delicate structure engineering is performed. Herein, atomically hybridized SnO2/Co3O4 nanocomposites with amorphous nature are successfully cast onto the porous N-doped carbon (denoted as NC) nanoflakes through facile pyrolysis of the tin (II) 2-ethylhexanoate (C16H30O4Sn) and cobalt (II) 2-ethylhexanoate (C16H30O4Co) mixture within NC nanoflakes in air at 300 °C for 1 h. The Sn/Co atomic ratio and the loading amount of SnO2/Co3O4 can be readily controlled, whose effect on lithium storage are investigated as anodes for lithium ion batteries (LIBs). Notably, SnO2/Co3O4@NC (RSn/Co = 1.25) nanoflakes exhibit the most excellent lithium storage properties, delivering a reversible capacity of 1450.3 mA h g-1 after 300 cycles at 200 mA g-1, which is much higher than that of the single metal oxide SnO2@NC and Co3O4@NC electrodes.

LCSCNet: Linear Compressing Based Skip-Connecting Network for Image Super-Resolution.

In this paper, we develop a concise but efficient network architecture called linear compressing based skipconnecting network (LCSCNet) for image super-resolution. Compared with two representative network architectures with skip connections, ResNet and DenseNet, a linear compressing layer is designed in LCSCNet for skip connection, which connects former feature maps and distinguishes them from newly-explored feature maps. In this way, the proposed LCSCNet enjoys the merits of the distinguish feature treatment of DenseNet and the parametereconomic form of ResNet. Moreover, to better exploit hierarchical information from both low and high levels of various receptive fields in deep models, inspired by gate units in LSTM, we also propose an adaptive element-wise fusion strategy with multisupervised training. Experimental results in comparison with state-of-the-art algorithms validate the effectiveness of LCSCNet.

Facile preparation of protonated hexaniobate nanosheets and its enhanced photocatalytic activity.

Exfoliated hexaniobate nanosheets E-H2K2Nb6O17-x (E-HKNO) with broad light absorption (up to 850 nm) and high adsorption properties were prepared via ion exchange and transient annealing processes with micron-size K4Nb6O17 powders as the precursor. The as-prepared E-HKNO nanosheets show excellent visible light photodegradation performances when compared to degussa P25, which was evaluated in terms of degradation of Rhodamine B (Rh B). High adsorption and broad light absorption characteristics could be attributed to the exfoliation behavior and the reduction of surface Nb5+ to Nb4+, which was confirmed by x-ray photoelectron spectroscopy (XPS) and Raman spectra. From the Mott-Schottky analysis, the E-HKNO is an n-type semiconductor and has a higher flat band voltage (-0.46 V versus RHE at pH = 7), compared with K4Nb6O17. In addition, the electrochemical impedance spectroscopy (EIS) indicates that the E-HKNO nanosheets have an increased semiconductor-electrolyte charge transfer resistance, which is not conducive to the separation of photogenerated carriers (e--h+). Accordingly, a small amount of holes scavenger (EDTA) was added to improve the photodegradation performance of the E-HKNO, since the holes scavenger can inhibit the recombination of the photogenerated carriers. This work provides not only a facile method for the preparation of an efficient E-HKNO nanosheets photocatalyst, but also new insights for further enhancing the photodegradation performance by adding trace scavenger.