Tag: deep learning

A new federated learning framework called NeuroSono-Net classifies neonatal brain ultrasound conditions across simulated institutions while keeping patient data local, ...

Researchers in South Africa combined deep learning, causal discovery and explainability to predict temperature breaks in apple cold rooms before ...

A deep learning model analyzing eye positioning and video-based attention achieved over 97% accuracy in distinguishing autistic from non-autistic children ...

A unified deep learning framework called MyoSpectrumNet combines pathologic myopia screening, maculopathy grading, refraction estimation and lesion segmentation from heterogeneous ...

A customised convolutional neural network detected mitotic figures in breast cancer histopathology images with a reported accuracy of 96.6 percent, ...

A new MLP-based segmentation method combining mixer layers with channel attention achieved a Dice coefficient of 81.45 percent on the ...

A new YOLOv8-based algorithm called DARF improves visible-infrared object detection by enhancing shallow features and adaptively routing cross-modal information, achieving ...

A new study maps major sign languages across five continents and proposes a hierarchical transformer architecture that reaches 91 percent ...

A new deep learning network called MTF-Net combines multi-scale transformer attention with cross-branch feature fusion to segment brain tumors from ...

PVFire-YOLO, a real-time detection framework with attention-enhanced multiscale fusion, reaches 87.4% mean average precision on a new photovoltaic fire dataset ...

Researchers report a test-time fine-tuning approach that adapts large-scale vision foundation models to segment electronic waste, cutting end-to-end processing time ...

Researchers trained stage-specific StyleGAN3 models to synthesize human blastocyst images and validated them with distributional, spectral, and classifier-based tests, finding ...

Adding a stride-4 P2 detection branch to YOLOv11n raised tiny fabric defect detection mAP@0.5 by 5.8 percentage points on the ...

Researchers combined YOLOv5, DeepSORT, OpenPose, BiLSTM and Dynamic Time Warping into a long jump teaching system that scored take-off movements ...

A new study reports a bumble bee mating-optimized convolutional neural network that classifies seven student emotional states from facial expressions ...

A prototype attendance system called Smart Click recognizes enrolled faces in group photographs with 95.26 percent accuracy, rejecting unknown faces ...

A new deep learning model combines bilingual text, cultural pragmatics and wireless sensor data from factory floors to assess whether ...

A new informatics workflow transfers fracture-segmentation knowledge from a large dataset to small UAV survey sites, improving crack maps where ...

A new lightweight detector, YOLO-FWF, combines Fourier-domain enhancement, wavelet-guided attention and orientation-aware feature fusion to improve scene text detection accuracy ...

A new cross-modal attention network combines match video, commentary audio, EEG and eye-tracking data to assess viewer immersion in VR ...