Browse Articles
Discover research articles across all indexed journals
Sensory-guided human-machine joint learning accelerates the acquisition of motor imagery brain computer interface control
pH-Dependent Vibrational Dynamics Drives Excited-State Quenching in the Phycobiliprotein Complex PC645
AI-driven feature descriptor design using CNNs and transfer learning for enhanced image retrieval accuracy and efficiency
CenSegNet: a generalist high-throughput deep learning framework for centrosome phenotyping at spatial and single-cell resolution in heterogeneous tissues
Abstract Centrosome abnormalities (CA) are a hallmark of epithelial cancers, yet their spatial complexity and phenotypic heterogeneity remain poorly understood due to limitations in conventional image analysis. Here we present CenSegNet (Centrosome Segmentation Network), a modular deep learning framework for high-throughput segmentation of centrosomes and epithelial architecture, enabling accurate and generalisable centrosome phenotyping at spatial and single-cell resolution across imaging modalities and tissue contexts. Applied to tissue microarrays comprising 911 breast cancer cores from 127 patients, CenSegNet enables large-scale, spatially resolved quantification of numerical and structural CA. We show that these CA subtypes are mechanistically uncoupled, exhibiting distinct spatial distributions, age-dependent dynamics, and associations with tumour grade, hormone receptor status, genomic alterations and nodal involvement. Structural CA are associated with overall survival, whereas discordant CA profiles at tumour margins correlate with local tumour aggressiveness and stromal remodelling. These findings establish CenSegNet as a scalable platform for spatially resolved centrosome phenotyping, enabling systematic investigation of centrosome biology and its dysregulation in cancer and other epithelial diseases.
Interfacial Reverse Charge Transfer Enabling Near 100% Selectivity of Glycerol Photooxidation into Formate over Schottky’s Au/TiO <sub>2</sub>
A methodology for qualitative and quantitative assessment of compressional deformation from growth strata: the Dibei structural zone, Kuqa Depression, Northwestern China
Fluctuation-driven mass-selective transport in dynamic nanopores
Abstract Precise molecular separation is essential in nanotechnology. However, static designs based on pore size and host-guest interactions often fail for species with nearly identical properties. To overcome this, we elucidate how nanopore dynamics govern transport and separation by establishing a general, predictive framework. We demonstrate that diffusion in a fluctuating periodic potential is maximized at an optimal fluctuation rate. Critically, this rate depends on molecular mass, translating subtle mass differences into significant kinetic disparities. By integrating this framework with quantum-chemical energy landscapes for archetypal soft porous crystals, we identify two factors governing selectivity: the magnitude of energy-barrier fluctuations and the alignment between nanopore dynamics and the energy-fluctuation rate that maximizes selectivity. These results establish nanopore dynamics as a key design dimension for controlling separation by tuning structural dynamics via ligand substitution or external fields, providing a theoretical foundation for the rational design of separation materials.
Efficient Light-Driven CO <sub>2</sub> Capture and Reversible Release Enabled by Metastable Photoacid-Decorated Metal–Organic Frameworks
Interpretable machine learning for predicting mild cognitive impairment in elderly patients with type 2 diabetes mellitus: model development and performance assessment
Spin-coordinated mixed-valence transport via indium-induced perovskite-spinel heterointerface in oxygen electrode for reversible protonic ceramic cells
Abstract Achieving efficient oxygen electrocatalysis in protonic ceramic electrochemical cells requires more than accelerating surface reactions, which demands precise coordination of electronic and spin states across the bulk-interface continuum. Conventional strategies based on vacancy engineering or catalyst infiltration increase active sites, yet they do not intrinsically couple bulk charge transport with interfacial redox kinetics. Here, we introduce spin-coordinated heterointerface engineering as a general design principle, exemplified by In-induced self-assembly of a Co 3 O 4 spinel layer on Pr 0.5 Ba 0.5 Co 0.7 Fe 0.3 O 3-δ (PBCF), forming PBCFIn05. Indium incorporation triggers selective Co migration and interfacial spinel reconstruction, which in turn drives a bulk spin-valence reconfiguration from predominantly high-spin Co 3 ⁺ to mixed-spin Co 3 ⁺/Co 4 ⁺. This evolution establishes complementary double-exchange conduction in the perovskite bulk and small-polaron hopping within the Co 3 O 4 spinel, creating a continuous mixed-valence pathway that synergistically lowers the activation barriers for both oxygen reduction and oxygen evolution. These results demonstrate that self-assembled spin-active heterointerfaces provide a powerful route to overcome intrinsic spin-selection bottlenecks in ceramic electrocatalysis and offer a broadly applicable platform for advanced solid-state energy conversion systems.