Coarse‐grained (left) and atomistic (right) models of the shape memory polymer ESTANE ETE 75DT3 are shown schematically. The two representations bridge molecular detail and mesoscopic description. Both models capture shape memory behavior, linking segmental mobility and conformational relaxation of anisotropic chains to macroscopic recovery, and ...
Fathollah Varnik
wiley +1 more source
A multi-scale transformer-enhanced YOLO framework for unified road damage detection and boundary-aware segmentation. [PDF]
Kulambayev B, Olzhayev O, Suliman A.
europepmc +1 more source
Low‐voltage FIB‐SEM tomography combined with a image preprocessing pipeline improves phase contrast and enables reliable machine‐learning segmentation of conductive networks in lithium‐ion battery electrodes. Structural descriptors are extracted from segmented images, done semimanually and automated, and compared.
Lisa Beran +6 more
wiley +1 more source
FEDI-CODE: A federated and causally-informed framework for dementia risk prediction using multi-site patient data. [PDF]
Moniruzzaman M +5 more
europepmc +1 more source
Comparison of Triply Periodic Minimal Surface Energy Absorbers Under Uniaxial Compressive Loading
This study investigates LCD 3D printed Triply Periodic Minimal Surface (TPMS) structures as mechanical energy absorbers. By comparing various base designs and layered combinations under uniaxial compression, it identifies that a Diamond‐Gyroid sandwich structure offers superior performance.
Sergej Grednev +2 more
wiley +1 more source
Ancient architecture image classification with progressive stacking pseudoinverse learning. [PDF]
Cai Z, Sun X, Zhang S, Zhao Z, Wu P.
europepmc +1 more source
A Practical Noise2Noise Denoising Pipeline for High‐Throughput Raman Spectroscopy
A lightweight and reproducible denoising pipeline for high‐throughput Raman spectroscopy is introduced, based on a 1D convolutional autoencoder trained with a Noise2Noise strategy. Using only repeated short‐exposure acquisitions, the method suppresses stochastic noise without reference spectra, enabling reliable spectral reconstruction while preserving
David Martin‐Calle +5 more
wiley +1 more source
LAIOR: a hyperbolic neural ODE variational framework for interpretable single-cell manifold learning and trajectory inference. [PDF]
Fu Z, Fu J, Zhang K, Ran T, Chen C.
europepmc +1 more source
Design and Analysis of Compression–Torsion Coupling Metamaterials Using the Golden Section Method
A novel compression–torsion metamaterial is engineered using inclined rods and symmetry breaking. To optimize its torsional performance, the golden section method is employed. The mechanical response of the metamaterial is validated through both numerical analysis and experimental validation.
Amirhossein Hassani, Sara Bagherifard
wiley +1 more source
Representation learning of crystal materials using Graph Neural Networks: Passive symmetry challenges and advances. [PDF]
Cui J, Han C, Liang J, Li L, Wang F.
europepmc +1 more source

