Results 151 to 160 of about 324,009 (309)
Euclid: Searches for strong gravitational lenses using convolutional neural nets in Early Release Observations of the Perseus field [PDF]
R. Pearce-Casey+11 more
openalex +2 more sources
Liver tumor segmentation CT data based on Alexnet-like convolution neural nets
Alexandr N. Korabelnikov+5 more
openalex +2 more sources
This study uses single‐nucleus RNA sequencing and spatial transcriptomics to investigate penile squamous cell carcinoma (PSCC). It reveals that PSCC tumor cells mimic normal penile epithelium differentiation, independent of HPV status. The Tum_1 subtype shows basal stem‐like characteristics and promotes invasiveness. HPV‐positive basal stem‐like tumors
Hongjian Song+37 more
wiley +1 more source
Human skin inherently forms a body‐coupled triboelectric circuit that enables the generation of triboelectric charges through contact between the skin surface and various materials. Integrated with an on‐finger, substrate‐free microfiber electrode, the material‐specific and personalized triboelectric signals can be imperceptibly collected for machine ...
Junting Huang+7 more
wiley +1 more source
Probability Voting-based Ensemble of Convolutional Neural nets Classifiers for Image Classification
Sarwo+3 more
openalex +1 more source
Colloidal nanoparticles self‐assembly advances towards intelligent, customized assembly through precise control of binary co‐assemblies. This review explores the evolution from monolithic to binary assemblies, highlighting how the AI‐guided programmable assembly approach has the potential to shift from passive assembly to active intelligent design.
Cancan Li+5 more
wiley +1 more source
Quantitative Electron Beam‐Single Atom Interactions Enabled by Sub‐20‐pm Precision Targeting
Atomic lock‐on (ALO) is a rapid, low‐dose in situ technique in scanning transmission electron microscopy (STEM) that achieves sub‐20 picometer (pm) beam positioning. A sparse annular scan collects positional information while avoiding irradiation of the target site.
Kevin M. Roccapriore+2 more
wiley +1 more source
PRO‐LDM is a modular latent diffusion framework that learns hierarchical biological representations to enable efficient and high‐fidelity protein design. PRO‐LDM can design species that resemble natural sequences with enhanced diversity or optimize protein functions.
Sitao Zhang+15 more
wiley +1 more source
Drug Resistance Predictions Based on a Directed Flag Transformer
CAPTURE, a deep learning framework combining structural and sequence data to predict drug‐resistant Mpro mutations, revealing that PAXLOVID use may be accelerating resistance—supporting real‐time viral surveillance and therapeutic design. Abstract The evolving SARS‐CoV‐2 virus threatens global public health, particularly with potential resistance to ...
Dong Chen+8 more
wiley +1 more source
This study presents an explainable multimodal AI model, MAIGGT (Multimodal Artificial Intelligence Germline Genetic Testing), that combines whole‐slide histopathology with clinical data for accurate germline BRCA1/2 mutation prescreening. By integrating digital pathology and EHR phenotypes, MAIGCT enables cost‐effective, scalable hereditary breast ...
Zijian Yang+23 more
wiley +1 more source