AI learns to map children’s swallowing videos from sparse expert labels
Deep learning models trained on sparse expert annotations can segment anatomical structures in pediatric swallowing endoscopy videos from unseen patients, ...
Tag: U-Net
Deep learning model spots oil spills in drone photos with pixel-level precision
A new deep learning model called HyEff-UNet segments oil spills in drone-captured port images pixel by pixel, achieving a macro ...
Deep learning model fuses construction plans with monitoring data to forecast diaphragm wall deflections
A deep learning framework that fuses construction schedules with wall-deflection monitoring data cut long-term prediction errors for diaphragm walls at ...
Satellite dataset maps thousands of Arctic lake and pond water masks for segmentation
Researchers assembled 5513 Sentinel-2 image-mask pairs from 23 Arctic permafrost lowlands to train and compare models that map small thermokarst ...
AI Denoising Cuts PET Scan Time but Struggles with Radiomic Features
A clinical study of 90 patients found that GAN-enhanced short-acquisition PET images improved lesion detection and maintained metabolic metrics, though ...
AI Tooth Segmentation Review Maps a Decade of Deep Learning Progress
A new PRISMA-guided systematic review compiles 72 studies and twelve open datasets to map how deep learning, from U-Net to ...
AI Cleans Noisy Accelerator Images to Reveal Hidden Beam Halo
An unsupervised U-Net deep-learning framework denoises accelerator emittance images well enough to resolve beam halos below one ten-thousandth of the ...
U-Net Variants Compared for Automated Lung Tumor Segmentation on CT Scans
A comparative study of nine U-Net-based deep learning configurations found U-Net with a VGG16 backbone achieved the best overall lung ...
Transformer-convolution hybrid reaches near-perfect hyperspectral image classification
A four-stage hybrid framework combining convolutional spectral compression, a Transformer encoder, and U-Net spatial extraction reports 99.85% overall accuracy on ...
AI Emulator Replicates Ammonia Ensembles 3300 Times Faster Than Physics Model
A new AI emulator reproduces GEM-MACH ammonia ensembles about 3300 times faster while delivering nearly identical emissions inversion results.
-
Mothers who receive childcare support from maternal grandparents show more parental warmth, finds NTU Singapore study
-
University of Seville Breaks 120-Year-Old Mystery, Revises a Key Einstein Concept
-
Bee body mass, pathogens and local climate influence heat tolerance
-
Researchers record first-ever images and data of a shark experiencing a boat strike
-
Groundbreaking Clinical Trial Reveals Lubiprostone Enhances Kidney Function
Embark on a thrilling journey of discovery with Scienmag.com—your ultimate source for cutting-edge breakthroughs. Immerse yourself in a world where curiosity knows no limits and tomorrow’s possibilities become today’s reality!