Ulcer-Region Prior Improves Diabetic Foot Classification but Shows Comparator-Dependent Gains
An internal evaluation of 5,955 diabetic foot ulcer patches found that incorporating a soft ulcer-region prior improved full lesion-prior transfer ...
An internal evaluation of 5,955 diabetic foot ulcer patches found that incorporating a soft ulcer-region prior improved full lesion-prior transfer ...
A hierarchical XGBoost framework incorporating lithology-family information improved classification accuracy and consistency across three independent test wells from the Yanchang ...
A hybrid deep-learning model combining ConvNeXt and Swin Transformer architectures demonstrated 95.0 percent nine-class accuracy for diagnosing lumbar disc herniation ...
A study evaluating graph neural networks across multi-event conversational cascades found that cross-event rumour detection remains challenging and strongly dependent ...
A benchmark across 6,900 outputs revealed that reasoning modes substantially increased clinical text extraction accuracy while noticeably lengthening inference latency.
A tri-spectrogram fusion transformer called IAMPT achieves the highest accuracy, Macro-F1 and Macro-AUC among evaluated models for fine-grained multi-fault recognition ...
A retrieval-augmented BioGPT pipeline for biomedical yes/no/maybe question answering shows that cross-validated performance, evidence conditions and run-to-run variability all reshape ...
A new knowledge distillation method compresses large pre-trained models into lightweight detectors that identify vulnerabilities in Python code with Macro-F1 ...
A new dual-view machine learning framework called scTrans-SVD achieved a mean Macro-F1 of 0.8820 across 13 public single-cell transcriptomic datasets ...
A two-stage AI framework uses a large language model and evolutionary feature selection to halve the symptom set for Traditional ...
© 2026 Scienmag - Science Magazine
© 2026 Scienmag - Science Magazine