Browse Articles

Nanoscale Core–Shell Catalysts for H ₂ Production by Methane Decomposition: Supported Nickel Nanoparticles Ensheathed in Metal Oxides

Effects of growth hormone and anabolic steroids, in critically ill patients admitted to the intensive care unit: a systematic review and meta-analysis

Here there be

Iridium-Catalyzed, Regio- and Stereoselective Silylation of Primary and Secondary C( <i>sp</i> ³ )–H Bonds in Primary Amines

Multi-omics insights into the molecular basis of powdery mildew resistance and root metabolic variation in Astragalus membranaceus var. mongholicus

Correction: Integrated spectroscopic and morphological analyses reveal cellular shifts in gene-silenced melanoma CSCs

Phosphate Boosts Catalytic Hydrodeoxygenation by Facilitating Proton/Electron Transfer at the Metal–Support Interface

Bacterial wollastonite concrete for sustainable high performance construction

Mechanistic Insights into the Light-Driven Difunctionalization of Alkenes with a Sulfonyl-Based Reagent: A Catalyst-Free Approach

BDS-Adam optimizer integrating adaptive variance rectification with semi-adaptive gradient smoothing

Abstract In this work, an enhanced variant of the Adam optimizer, termed BDS-Adam, is proposed to address two critical limitations of the original Adam algorithm: biased gradient estimation and training instability during early optimization. To overcome these issues, a dual-path framework is adopted. In the first path, a nonlinear gradient mapping module (adaptive reshaping of raw gradients using hyperbolic tangent) is applied to adaptively reshape raw gradients, enabling the optimizer to better capture local geometric structures. In the second path, a semi-adaptive gradient smoothing controller–based on real-time gradient variance–is incorporated to suppress abrupt parameter updates and stabilize training dynamics. These two outputs are integrated through a gradient fusion mechanism (combining smoothed and transformed gradients before updates), in which smoothed and transformed gradients are combined prior to parameter updates. Moreover, an adaptive second-order moment correction technique is employed to mitigate cold-start effects caused by inaccurate variance estimates in the early training phase. A convergence analysis under non-convex settings is provided, and it is theoretically demonstrated that the expected gradient norm is bounded under standard assumptions, indicating improved robustness and long-term stability. This adaptive bias-correction formulation further improves training stability. Empirical evaluations on three benchmark datasets–CIFAR-10, MNIST, and a gastric pathology image dataset–reveal test accuracy improvements of 9.27%, 0.08%, and 3.00%, respectively, compared to Adam. These results confirm that the proposed dual-mechanism optimizer effectively enhances both convergence speed and generalization performance across diverse tasks.

Homologous Imide Bonds to Build Polymer-Covalent Organic Framework Electrolytes for Efficient Ion Transport

Artificial intelligence driven intraocular lens power calculation in extreme axial myopia

Abstract Accurate intraocular lens (IOL) power calculation is critical in cataract surgery, especially in patients with extreme axial myopia where traditional formulas often yield inaccurate results. This study retrospectively evaluated the accuracy of two AI-driven IOL formulas (Hill-RBF, Kane), the Barrett Universal II formula, and the traditional SRK/T formula in patients with axial lengths ≥ 30.0 mm. Data from 80 eyes of 51 patients treated at the Institute of Science Tokyo were analyzed. Postoperative refractive errors were recalculated, and accuracy was assessed using mean error (ME), mean absolute error (MAE), and median absolute error (MedAE). Statistical analyses included the Wilcoxon signed-rank test and chi-square test. The Kane and Hill-RBF formulas demonstrated significantly lower MAE (0.51 D and 0.52 D, respectively) compared to SRK/T ( P  &lt; 0.05). MAE of the Barrett Universal II formula was 0.66D, which was not significantly different from SRK/T. In eyes with axial lengths ≥ 32.0 mm, Kane achieved the lowest MAE and MedAE (0.44 D and 0.40 D). Both Kane and Hill-RBF showed lower refractive errors &gt; ± 1.0 D (7.5%) compared to SRK/T (42.5%). AI-driven formulas, particularly Kane and Hill-RBF, significantly improve refractive accuracy in extreme axial myopia. Their clinical adoption may enhance postoperative visual outcomes and reduce the need for corrective interventions.

Ligand-Controlled Stereodivergent α-Vinylation and α-Arylation of Peptide Backbones

Determinants of chronic malnutrition among under-five children in Ethiopia using simultaneous quantile regression

Intracranial aneurysm risk stratification in acute ocular motor nerve palsy based on clinical features

Tumours might be sensitized to immune therapy by COVID mRNA vaccines

High-resolution climate prediction in mountainous terrain using a ConvLSTM-XGBoost hybrid model with dynamic bayesian weighting

Abstract To address the challenge where the interplay between spatiotemporal dynamics and topographic effects complicates climate modeling over complex terrain, we propose a hybrid ConvLSTM-XGBoost model incorporating dynamic Bayesian weighting, and demonstrate its capacity for high-precision climate prediction through a case study in the Hongyuan Mountain region of Yunnan, China (22.5°–23.5°N, 102.5°–103.5°E); specifically, the ConvLSTM network captures spatiotemporal evolution patterns (e.g., propagation of the southwest monsoon front) from the 0.25° resolution CN05.1 climate dataset, while XGBoost quantifies the nonlinear modulation effects of 90-m SRTM DEM-derived topographic features (elevation, aspect) on precipitation phases, with an innovatively integrated Bayesian Model Averaging (BMA) framework dynamically calibrating model weights—XGBoost at 0.68 ± 0.05 during dry seasons and ConvLSTM at 0.72 ± 0.07 during monsoons—to enhance responsiveness to extreme events. Validation using 1961–2022 climate data shows the hybrid model reduces precipitation prediction mean absolute error (MAE) by 30.5% compared to CMIP6 (achieving an MAE of 0.0089 [specify units, e.g., mm/day]), improves the F1-score for identifying extreme precipitation (&gt; 50 mm/day) by 20%, achieves 96.53% accuracy in maximum temperature (Tmax) predictions (errors ≤ 3%), and reduces high-temperature dispersion by 52%, thereby serving as a 1-km resolution decision-support tool for mountain climate risk management, supporting drought warning and hydropower scheduling in Yunnan’s Climate Adaptation Plan 2035, and offering a scalable framework for global mountain climate modeling.

Structure-Guided Design of a Bioactive Covalent Small Molecule Targeting a Riboswitch

Risk factors for refracture by different cement leakage after percutaneous vertebroplasty in patients with thoracolumbar compression fractures

Ion-Specific Interfacial Electric Fields on Water Microdroplets for Tuning Menshutkin Reactions