AI RESEARCH
Margin-Consistent Deep Subtyping of Invasive Lung Adenocarcinoma via Perturbation Fidelity in Whole-Slide Image Analysis
arXiv CS.CV
•
ArXi:2603.06650v1 Announce Type: new Whole-slide image classification for invasive lung adenocarcinoma subtyping remains vulnerable to real-world imaging perturbations that undermine model reliability at the decision boundary. We propose a margin consistency framework evaluated on 203,226 patches from 143 whole-slide images spanning five adenocarcinoma subtypes in the BMIRDS-LUAD dataset. By combining attention-weighted patch aggregation with margin-aware