Tag: deep learning

Pretrained convolutional neural networks classified biological sex from smartphone photographs of palatal rugae with up to 77 percent accuracy in ...

A systematic review of 75 studies finds ECG-based AI models reporting 90 to 99 percent accuracy, but warns that external ...

A new ensemble deep learning framework called EnPreNiNet, paired with a night image enhancement technique, achieved a reported mean Average ...

A new attention-based neural network that reconfigures its risk logic according to macroeconomic conditions outperformed standard machine learning and statistical ...

A comprehensive narrative review finds that predictive AI can accelerate pollutant hazard assessment, monitoring, and exposure modeling, but only as ...

Researchers have built a federated, explainable deep learning framework that detects attacks on Internet of Medical Things devices with detection ...

A customized VGG19 deep learning model trained on 2,100 chest X-rays from Ethiopian hospitals classified pneumonia and tuberculosis with 99% ...

A hybrid CNN-LSTM-attention model predicts risky digital usage patterns in adults aged 60 and older with 0.947 accuracy and 0.93 ...

A new systematic review and meta-analysis of 20 studies finds that AI models show high pooled accuracy for detecting pediatric ...

A study tested nine deep learning models on time-frequency images of EEG signals, with Mel-spectrogram features paired with AlexNet reaching ...

Deep learning models trained on video, equipment logs and clinical data from 222 real coronary angiograms recognized catheterization lab workflow ...

Researchers report an AI and IoT framework that detects ambulances with cameras and audio, achieving 98.4 percent accuracy and flipping ...

A controlled benchmark of five CNN architectures on 2,000 intraoral photographs found EfficientNet-B0 led with 98.18 percent accuracy, though no ...

A multicenter study of 603 rectal cancer patients found that a deep learning network segmented rectal tumors on MRI with ...

An enhanced PointNet++ with attention mechanisms and model scaling reaches up to 96.26 percent accuracy in separating ground from vegetation ...

A new open-access review finds that machine learning and deep learning models using real-time sensor data improve electric vehicle remaining ...

A new open-access study presents a modular web platform that classifies brain tumor MRI scans in real time using a ...

An automated deep learning pipeline measured skeletal muscle and fat with high agreement from same-day whole-body MRI and low-dose PET/CT ...

A new generative AI framework called MOVIE synthesizes late-phase amyloid PET images from one-minute early-phase scans, outperforming a UNet baseline ...

A lightweight hybrid transformer model combining wavelet transforms with MobileViT achieved over 99% accuracy in classifying plant diseases across four ...