SAR and optical data fuse in dynamic graph network to classify sea ice with 96.69% accuracy
A dual-branch deep learning framework called MAG-Net fuses Sentinel-1 radar and Sentinel-2 optical imagery over Hudson Bay, reaching 96.69% mean ...
Tag: deep learning
Deep Learning Framework CancerSubtyper Automates DNA Methylation Analysis for Tumor Subtyping
Researchers have developed CancerSubtyper, an end-to-end deep learning framework with an interactive web interface that automates preprocessing, batch effect correction, ...
Artificial Intelligence Models Predict Cancer Diagnoses and Gene Expression from Tissue Slides
A systematic review examines how deep learning algorithms analyze gigapixel whole-slide images to classify tumors and predict transcriptomic profiles directly ...
Dual-Modal Neural Network Accelerates Automated Road Crack Detection for Edge Hardware
A lightweight dual-modal neural network combining wavelet-based frequency fusion and knowledge distillation achieves accurate road crack detection and real-time inference ...
Dual-Domain Swin Framework Restores Projections to Enhance Sparse-View CT Quality
A dual-domain Swin framework restoring projection data before filtered back projection and refining images improves sparse-view computed tomography reconstruction quality.
Machine Learning Model Integrates Multimodal Data to Stratify Glioblastoma Survival Risk
A multimodal machine learning nomogram combining clinical, radiomics, and pathomics features improved overall survival risk stratification in patients with IDH-wildtype ...
AI blood glucose predictor built on physiology explains its own forecasts
A new physiology-informed neural network with attributable branches predicts blood glucose in type 1 diabetes while revealing which factors drive ...
Benchmarking Deep Transfer Learning Models for COVID-19 Lung Ultrasound Detection
A systematic benchmarking study demonstrates that DenseNet-121 paired with the Adam optimizer achieves exceptional diagnostic accuracy in detecting COVID-19 from ...
Novel Architecture Enhances Computer Vision for Complex Construction Site Object Detection
Researchers developed SFMDet, an augmented YOLOv12-based framework combining state-space scanning and cross-attention to improve object detection across cluttered construction environments.
Metasurface Sensor and Deep Learning Enable Real-Time Hyperspectral Polarimetric Imaging
Researchers developed a deep learning-empowered metasurface sensor that achieves real-time hyperspectral and full-Stokes polarimetric imaging across 167 spectral channels.
Robot Learns Where to Stick Its Soil Sensors Safely
A Chinese research team built a soil-sensing robot that scans the ground, scores candidate insertion points and matched expert judgment ...
Lightweight Edge Computing Model Dynamically Segments Citrus Flowers and Developing Fruitlets
Agricultural engineering researchers have developed a lightweight deep learning model capable of segmenting citrus flowers and young fruitlets in real ...
Retinal Deep Learning Biomarker Reconstructs Individual Lifespan Trajectories and Mortality Risk
A deep learning biomarker using retinal photographs maps personalized lifespan trajectories to predict all-cause mortality and systemic disease risk across ...
Deep Learning Network MSCF-DM Predicts DNA Methylation Sites Across Multiple Species
Researchers have developed MSCF-DM, a multi-scale species-conditioned deep learning network that predicts cross-species DNA methylation sites across three modification types.
AI model predicts liver tumor ablation volumes better than manufacturer estimates
A patient-specific AI-powered biophysical model predicted radiofrequency ablation volumes in hepatocellular carcinoma with a median relative volume difference of 0.39, ...
Multi-omics attention model classifies breast cancer subtypes with explicit cross-omics interactions
A modality-aware external attention framework integrating mRNA, methylation, and miRNA data reached 86.78% accuracy in cross-validated breast cancer subtype classification ...
Mamba AI model reads Sentinel-2 satellite data to map seed maize yields
An agronomic-knowledge guided deep learning model called AKGMamba estimates plot-scale seed maize yield from Sentinel-2 imagery, reaching an R2 of ...
Multimodal IoT Framework Blends Gesture Control and Deep Learning for Adaptive Art
Researchers have developed IoT ArtNet, an intelligent framework combining deep learning and Internet of Things gesture sensing to dynamically drive ...
Deep Learning Network Reconstructs Missing Geometry in Maize Leaf Point Clouds
Researchers have developed PFL-Net, a deep learning network that reconstructs incomplete three-dimensional point clouds of maize leaves to improve structural ...
Deep Learning Synthesizes Cross-Contrast MRI Scans Across Thirty Reviewed Medical Imaging Studies
A systematic review of thirty studies reveals that generative adversarial networks and diffusion models lead deep learning efforts to synthesize ...
POPULAR NEWS
-
Chinese Cancer Association issues unified guideline for tumor-related lung injury care
-
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