Audit Reveals PCOS Ultrasound AI Relies on Image Shortcuts Rather Than Morphology
A systematic audit of a widely used ultrasound archive reveals that top-performing federated deep learning models classify polycystic ovary syndrome ...
A systematic audit of a widely used ultrasound archive reveals that top-performing federated deep learning models classify polycystic ovary syndrome ...
A phase-adaptive deep learning pipeline automatically quantified five liver indices from gadoxetic acid-enhanced MRI without inter-phase co-registration, matching manual measurements ...
Researchers present CADS, a framework pairing a 22,022-scan dataset with complete annotations of 167 anatomical structures and a whole-body CT ...
A hybrid deep learning framework for classifying cataract, diabetic retinopathy, glaucoma, and normal retinas dropped from 99.24 percent internal accuracy ...
A standardized 3D convolutional neural network identified cervical spinal cord injury on CT scans with AUC-ROC values near 0.95 across ...
A knowledge-distilled convolutional network reached 92.64 percent accuracy on an independent breast ultrasound dataset, outperforming chief physicians in a retrospective ...
A new graph neural network framework makes skin lesion classification more transparent, but candid reporting reveals serious weaknesses in melanoma ...
A systematic review of 11 studies finds that deep learning models using ultrasound show moderate sensitivity and high specificity for ...
A new deep learning architecture called TexTNet integrates local binary pattern texture features into a Swin Transformer's attention mechanism to ...
A vision-capable large language model detected coronary artery calcification on routine chest X-rays with promising accuracy, suggesting ordinary radiographs could ...
© 2026 Scienmag - Science Magazine
© 2026 Scienmag - Science Magazine