Results 221 to 230 of about 10,610,136 (340)

In‐Sensor Reservoir Computing Using Ferroelectric Optoelectronic Synapse for Near‐Infrared Face Recognition

open access: yesAdvanced Intelligent Systems, EarlyView.
Near‐infrared reservoir computing system has been realized by using α‐In2Se3 optoelectronic device with concise device and algorithm architecture, achieving high accuracy near 100% and noise robustness in face recognition tasks. These findings demonstrate 2D ferroelectric device‐based optoelectronic RC system as a compact and efficient computing ...
Wenyu Songlu   +15 more
wiley   +1 more source

Small but Strong: Lightweight Architecture Improves Lung Nodule Detection and Segmentation

open access: yesAdvanced Intelligent Systems, EarlyView.
(a) Separated network architecture, where the lung nodule detection and segmentation networks are separated. (b) Separated head architecture, where an anchor‐based head is responsible for detection and another head for segmentation. (c) The study's architecture, containing a lightweight backbone network based on deformable Transformer for feature ...
Lilu Liu   +6 more
wiley   +1 more source

A Periodic Error Correction Method for Terahertz Coded‐Aperture Imaging

open access: yesAdvanced Intelligent Systems, EarlyView.
A physics‐constrained unsupervised learning framework with periodic error correction is proposed for terahertz coded‐aperture imaging. The method improves reconstruction robustness against noise and model mismatch, while reducing reliance on large‐scale labeled data.
Hongxu Wu   +8 more
wiley   +1 more source

A Novel Semi‐Automated Pipeline for Optimizing 3D‐Printed Drug Formulations

open access: yesAdvanced Intelligent Systems, EarlyView.
A novel, semi‐automated deep learning and differential evolution pipeline designed for generating medicines for 3D‐printed medicines is introduced. This pipeline successfully generates a diverse range of drug formulations with an 80% printing success rate and 92% accuracy in predicting printing parameters.
Youssef Abdalla   +8 more
wiley   +1 more source

RefineCatDiff: Toward High‐Quality Medical Image Segmentation via a Categorical Diffusion Refinement Framework

open access: yesAdvanced Intelligent Systems, EarlyView.
This study proposes RefineCatDiff, a refinement framework for high‐quality medical image segmentation. By developing a categorical distribution‐based discrete diffusion process for refinement, the framework aligns well with the characteristics of image segmentation tasks. Experimental results on multiple datasets across different modalities demonstrate
Feng Liu   +8 more
wiley   +1 more source

Home - About - Disclaimer - Privacy