Results 231 to 240 of about 397,237 (311)

Cytological Classification Diagnosis for Thyroid Nodules via Multimodal Model Deep Learning

open access: yesAdvanced Science, EarlyView.
This study introduces AI‐TFNA, an innovative artificial intelligence model designed to assist cytopathologists in classifying thyroid nodules based on The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC). The model effectively differentiates between benign and malignant thyroid nodules, demonstrating significant potential as a screening ...
Yuanzheng Lou   +27 more
wiley   +1 more source

Artificial Nervous Systems

open access: yesAdvanced Science, EarlyView.
By drawing inspiration from biological neural hierarchies and the working mechanisms of plasticity, researchers have constructed a series of bionic devices, including sensory devices, synapse devices, and artificial neural systems. They committed to simulating and surpassing the biological information processing function, thereby realizing the ...
Lu Yang   +5 more
wiley   +1 more source

Orchestrating Multi‐Ångstrom Spaced Cu─Ni Dual‐Atom Pair for Synergistic C─H Activation in Direct Methane Oxidation to Methanol

open access: yesAdvanced Science, EarlyView.
This work constructs a sub‐ångstrom‐spaced Cu−Ni dual‐atom pair over defect‐rich indium oxide support for methane oxidation into methanol. With Cu/Ni showing affinity to C−H bond's H/C end, an outstanding primary oxygenates (CH3OH and CH3OOH) productivity of 106 mmol gcat h−1 and a high oxygenates selectivity of 84.55% is achieved under mild conditions.
Jingting Jin   +4 more
wiley   +1 more source

Rethinking the AI Paradigm for Solubility Prediction of Drug‑Like Compounds with Dual‐Perspective Modeling and Experimental Validation

open access: yesAdvanced Science, EarlyView.
Aqueous solubility governs drug delivery and bioavailability, yet its prediction remains challenging. Utilizing the largest curated dataset of drug‐like molecules, prediction models are constructed through dual‐perspective benchmarking. Stacking methods outperform advanced deep learning due to data constraints. These validated models enabled systematic
Qilin Zhu   +8 more
wiley   +1 more source

On the Design of Effective Water‐Soluble Actinide‐Masking Ligands Through Ligand Structure Modulation

open access: yesAdvanced Science, EarlyView.
Ligand topology and non‐coordination atoms are proved to be important for the design of effective water‐soluble actinide masking agents. Abstract Water‐soluble actinide‐masking ligands are fundamentally important for achieving efficient lanthanide/actinide separation and for the development of water‐soluble f‐block complexes for bioimaging and ...
Bin Li   +9 more
wiley   +1 more source

Home - About - Disclaimer - Privacy