Hybrid Deep Learning Model Improves PM2.5 Forecasts in Anhui Province
A hybrid VMD-CBiLSTM model optimized by the shuffled frog-leaping algorithm achieved an average R2 of 0.981 when predicting PM2.5 concentrations ...
Tag: deep learning
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 ...
Review maps deep learning tools to real clinical tasks in pathology
A new review in Chinese Medical Journal maps deep learning architectures to core clinical tasks in pathology, highlighting accuracy gains, ...
Deep Learning Scans Earthquake Damage, But Review Finds the Evidence Shaky
A systematic review of 46 studies finds deep learning promising for post-earthquake structural damage assessment, but weak generalisation evidence and ...
Autoencoder Flags IoT Network Attacks with Near-Perfect Precision Under Extreme Imbalance
Most intrusion detection research is tested on artificially balanced datasets, where attack and benign records appear in equal numbers. A ...
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 ...
Vision-Language AI Model Detects Bengali Fake News with Explainable Reasoning
Researchers report a multimodal framework combining Bengali OCR, hybrid visual encoders, and vision-language reasoning that achieved 97% accuracy on a ...
CT-based model predicts early progression after TACE in liver cancer
A multicenter study reports that a model combining pretreatment arterial-phase CT features and clinical variables predicted time to progression after ...
AI Pipeline Turns Brain MRI Saliency Maps into Diagnostic Reports
A benchmark study couples CNN saliency maps, a cortical atlas and three LLMs to generate interpretable diagnostic reports from brain ...
AI Learns to Read Equine Emergency Ultrasound Windows in Proof-of-Concept Study
A deep learning model trained on ultrasound images from six healthy horses could recognize most standardized FLASH examination windows, though ...
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 ...
From Fourier to Deep Learning: How Bridge Health Monitoring Signals Are Evolving
A new review of 201 studies traces how bridge structural health monitoring evolved from classical Fourier and wavelet analysis to ...
Smarter Weather Inputs Cut Day-Ahead Solar Power Forecast Errors in Half
A two-stage framework that corrects coarse weather forecasts before predicting solar output halved day-ahead power forecast errors at a 20 ...
Dynamic Graph AI Model Hits Nearly 96% Accuracy in Traffic Congestion Prediction
A new dynamic graph neural framework called GAMMA-Traffic updates road-network connections in real time using cross-correlation analysis, reporting up to ...
Research report: Instance-level localization and similarity-aware contrastive learning for robust plant disease classification
A new AI framework first localizes individual disease regions in real-field leaf photographs and then applies similarity-aware contrastive learning, reaching ...
AI Model Forecasts Avian Influenza Outbreak Risk Across Canada
A pretrained vision transformer called RAVEN predicts highly pathogenic avian influenza outbreak risk across Canada up to two weeks in ...
TaichiNet reads whole-body Tai Chi motion from skeletons to sort subtle moves
A new skeleton-based framework, TaichiNet, combines local spatio-temporal graph modeling with a Transformer encoder to recognize Tai Chi movements, outperforming ...
Zero-Watermarking Scheme Aims to Protect Color Medical Images Without Altering Them
A new zero-watermarking scheme for color medical images uses VGG19 feature extraction, DCT and cyclic chaos encryption to verify copyright ...
Simple Machine Learning Outperforms Deep Models in Somali Dialect Test
A new study reports that a simple Naive Bayes classifier reached 99.05 percent accuracy in distinguishing Somali's MAXAA TIRI and ...
AI Takes On the Flawed Math of Measuring Leaky Heart Valves
A new review in Cardiovascular Ultrasound explains how artificial intelligence tools, including deep learning PISA segmentation and 3D Auto Color ...
POPULAR NEWS
-
Multitask Cambrian Model Improves Protein Melting Temperature Prediction
-
Vasopressors Exert Divergent Immunomodulatory Effects Across a Spectrum in Septic Shock
-
Study Identifies Key Drivers Shaping Salmonid Gill Microbiomes in Aquaculture
-
Chinese Cancer Association issues unified guideline for tumor-related lung injury care