Tag
This paper presents a modular medical imaging agent that verifies spatial relations in CT scans by decomposing tasks into parsing, localization, and geometric rules, achieving 94.1% accuracy and outperforming end-to-end vision-language models by 42.5 percentage points on a benchmark while ensuring auditable reasoning.
The paper introduces RibAssist 3D, a method for biplanar rib-fracture detection and 3D localization from CT projections, identifying cross-view correspondence as a key bottleneck in accuracy.
A new AI model (REDMOD) can detect pancreatic cancer up to three years earlier than human doctors by analyzing CT scans for subtle irregularities, potentially improving early diagnosis and survival rates.