Results 111 to 120 of about 14,848 (289)

Enhanced Endoscopic Internal Drainage of Gastric Abscess Through Additively Manufactured Stents

open access: yesAdvanced Healthcare Materials, EarlyView.
Postoperative gastric leaks are often treated with off‐label biliary double‐pigtail stents, yet conventional extruded designs are not optimized for leak anatomy, can migrate, and may limit abscess evacuation. PETALS is introduced to optimize transmural drainage geometry and enable patient‐specific 3D‐printable stents.
Parima Phowarasoontorn   +14 more
wiley   +1 more source

A Lightweight Network With Embedded Soft Constraints on Approximate Spectral Features for Real-Time Water Body Segmentation in Remote Sensing Images

open access: yesIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Real-time extraction of water bodies from remote sensing images captured by unmanned aerial vehicles (UAVs) or satellites is challenging due to the difficulty in performing precise atmospheric correction and other preprocessing steps. In this article, we
Qingqing Cao   +4 more
doaj   +1 more source

EM-Network: Oracle Guided Self-distillation for Sequence Learning

open access: yes, 2023
We introduce EM-Network, a novel self-distillation approach that effectively leverages target information for supervised sequence-to-sequence (seq2seq) learning.
Ahn, Sunghwan   +5 more
core  

Oxygen Supply of Islets of Langerhans by Photosynthetically Active Microalgae in Bioprinted Co‐Cultures Maintains Their Function in a Hypoxic Environment

open access: yesAdvanced Healthcare Materials, EarlyView.
This study developed a bioprinted co‐culture system embedding rat pancreatic islets and Scenedesmus sp. microalgae spatially defined in close vicinity. Red light was found optimal to ensure microalgal photosynthesis while maintaining islet viability and functionality. A tailored co‐culture medium supported both cell types.
Finn Dani   +7 more
wiley   +1 more source

Efficient and secure µ-Training and µ-Fine-Tuning for edge-based TinyML with future-guided self-distillation

open access: yesMachine Learning: Health
This study presents a novel, computationally efficient training framework demonstrated through bio-signal processing on edge medical devices. The approach integrates conventional full training with an innovative µ -Training technique, wherein the encoder
Zhaojing Huang   +3 more
doaj   +1 more source

A Novel Multibranch Self-Distillation Framework for Optimizing Remote Sensing Change Detection

open access: yesIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Deep learning (DL) has achieved remarkable success in the field of change detection (CD) for remote sensing images. Existing training methods for DL-based CD models are predominantly single-stage, single-stream, and end-to-end.
Ziyuan Liu   +3 more
doaj   +1 more source

Self Distillation via Iterative Constructive Perturbations

open access: yes
Deep Neural Networks have achieved remarkable achievements across various domains, however balancing performance and generalization still remains a challenge while training these networks. In this paper, we propose a novel framework that uses a cyclic optimization strategy to concurrently optimize the model and its input data for better training ...
Dave, Maheak   +5 more
openaire   +2 more sources

Engineered Protein‐Based Ionic Conductors for Sustainable Energy Storage Applications

open access: yesAdvanced Materials, EarlyView.
Rational incorporation of charged residues into an engineered, self‐assembling protein scaffold yields solid‐state protein films with outstanding ionic conductivity. Salt‐doping further enhances conductivity, an effect amplified in the engineered variants. These properties enable the material integration into an efficient supercapacitor.
Juan David Cortés‐Ossa   +14 more
wiley   +1 more source

Research on a Cross-Domain Few-Shot Adaptive Classification Algorithm Based on Knowledge Distillation Technology

open access: yesSensors
With the development of deep learning and sensors and sensor collection methods, computer vision inspection technology has developed rapidly.
Jiuyang Gao   +4 more
doaj   +1 more source

Label‐Free SERS Fingerprinting of Neuroprotein Conformational Dynamics in Human Saliva

open access: yesAdvanced Materials, EarlyView.
Galvanic molecular entrapment (GME) is a label‐free method for detecting and quantifying neuroprotein conformational states. This technique enables direct surface binding and in situ hotspot generation around molecules, effectively overcoming challenges related to target localization and mismatched hotspot geometries.
Muhammad Shalahuddin Al Ja'farawy   +10 more
wiley   +1 more source

Home - About - Disclaimer - Privacy