Head-to-head benchmark ranks EfficientNet-B0 top for spotting enamel caries in intraoral photos
A controlled benchmark of five CNN architectures on 2,000 intraoral photographs found EfficientNet-B0 led with 98.18 percent accuracy, though no ...
Tag: Transfer Learning
Web Platform Puts Three Deep Learning Models to Work on Brain Tumor MRI Scans
A new open-access study presents a modular web platform that classifies brain tumor MRI scans in real time using a ...
Most EEG emotion datasets are hard to reuse, review of 27 finds
A systematic review of 27 EEG emotion-recognition datasets finds that nearly half lack a dataset-level identifier, three heavily cited corpora ...
Dual-Stream AI Network Predicts Battery Heat and Health to Warn of Thermal Risks Early
A dual-stream AI network that jointly predicts lithium-ion battery temperature and state of health achieved a 92.6 percent timely warning ...
Causal Inference Boosts Privacy-Preserving Federated Learning Across Organizations
A Scientific Reports study presents a causal-enhanced federated transfer learning framework that improved cross-domain generalization robustness by 43% over conventional ...
Stacking deep learning reaches 97.2% accuracy for olive disease images
A stacking ensemble combining EfficientNet, ResNet and MobileNet classified healthy and diseased olive leaf images with 97.21% accuracy on a ...
Deep Learning Model Reaches 98.39% Accuracy in Cotton Leaf Disease Classification
A hybrid ConvNeXt and Random Walk Grey Wolf Optimizer framework classified six categories of cotton leaf conditions with 98.39% accuracy ...
Super-resolution GAN and swarm-optimized transfer learning sharpen Alzheimer’s MRI classification
A new deep learning framework that pairs super-resolution image enhancement with swarm-optimized transfer learning classified Alzheimer's disease stages from MRI ...
Hybrid AI model finds balanced mix for marine coral concrete
A response-surface-guided neural network predicted a balanced silica-fume and nano-silica formulation for coral aggregate concrete, reducing prediction error without additional ...
AI Learns How Far Cultures Sit Apart to Rewrite Content That Truly Travels
A new transfer learning framework measures cultural distance between audiences and uses it to rewrite film, tourism, and heritage content ...
More Detail in AI Prompts Can Backfire When Chatbots Build Safety Models
A controlled study of four frontier chatbots building helmet-detection models finds that the most detailed prompts can trigger catastrophic training ...
Self-Supervised Transformer Reads Resting-State EEG to Detect Alzheimer’s Disease With 96.42% Accuracy
Alzheimer's disease affects tens of millions of people worldwide, and the World Health Organization estimates that over 55 million people ...
Learnable Activation Head Pushes Alzheimer’s MRI Staging to 94.5 Percent Accuracy
A hybrid model pairing a ConvNeXt backbone with a Kolmogorov–Arnold Network head reaches 94.5 percent accuracy in four-way Alzheimer's staging ...
AI Chooses Its Own Strategy to Predict How Fabrics Fade in Sunlight
Every dyed fabric tells a story of slow defeat. Sunlight, moisture, and heat conspire to break the chemical bonds that ...
AI Reads Brain Scans Twice: Dual-Task Deep Learning Spots Tumors and Names Them in Milliseconds
Researchers have built a modified dual-task DenseNet121 that classifies brain tumors from MRI scans with over 99 percent accuracy while ...
AI Trick Turns 76 Soil Samples Into a Powerful Carbon-Mapping Tool
A systematic 264-experiment study in China found that a Wasserstein GAN augmenting just 76 soil samples improved drone-based soil organic ...
Self-Supervised AI Reads Heart Rhythms With Hospital-Scale Accuracy Using Mamba
A new self-supervised framework combining joint-embedding prediction and bidirectional Mamba learns diagnostic ECG representations without labeled data, outperforming prior baselines ...
Clustering Across Domains: New Method Sharpens Graph Transfer Learning
Researchers at Central South University have developed a cluster-based domain adaptation method that aligns only semantically similar portions of source ...
Deep Learning Is Transforming How the World Spots Sick Plants, Landmark Review Finds
A sweeping review of 1,081 studies from 2014 to 2024 shows convolutional neural networks rapidly becoming the world's leading tool ...
New R Package matrans Brings Transfer Learning to Multi-Source Cancer Prognosis Studies
A newly released open-access R package called matrans makes parameter-transfer learning based on frequentist model averaging accessible for building cancer ...
POPULAR NEWS
-
Modeling Reveals How Silica Gold Nanoparticles Reshape Optical Energy and Fields
-
Intravenous Iron Formulations Spur Divergent Muscle Recovery in Iron Deficiency Anaemia
-
Pararectal Leiomyosarcoma Diagnosed After Laparoscopic Hysterectomy Without Morcellation
-
Bone and Soft Tissue Combined: New Imaging Method Builds 3D Shape Models of the Foot and Ankle