The study evaluated whole‑slide image classifiers for gastric cancer across five Korean hospitals, using foundation‑model features and leave‑one‑hospital‑out validation. A hierarchical patch‑region‑slide graph aggregator (CHIME‑MIL) with site alignment and two‑scale fusion achieved a mean AUC of 0.837, outperforming six conventional attention and transformer aggregators that lacked these steps. However, when matched with the same alignment and fusion, CHIME‑MIL did not surpass simple attention‑based pooling, and global max pooling performed best (AUC 0.855). The alignment step, which subtracts each hospital’s mean feature vector without labels, enabled onboarding of a new hospital from only ten unlabeled slides but yielded only a modest 0.002 AUC improvement, and the model’s accuracy gains were inconsistent. "whyItMatters":"The findings show that adding complex aggregation methods does not necessarily improve cross‑hospital discrimination, highlighting the importance of simpler, label‑free onboarding and calibration checks for deployable pathology systems."
arXiv:2610.33965v1 Announce Type: new Abstract: Automated classification of gastrointestinal (GI) findings in endoscopic images could support early diagnosis, but architectures are often chosen on ac...
arXiv:2608.23574v1 Announce Type: cross Abstract: Each WSI slide contains thousands of candidate tissue patches, while supervision is usually available only at slide level. Existing bag-construction...
arXiv:2610.28007v1 Announce Type: new Abstract: Pathology foundation models (PFMs) support a wide range of histopathology tasks on whole-slide images through multiple instance learning (MIL). Despite...
HERO (Histology Encoder for Robust Representation in Oncology) is a ViT‑G/14 pathology foundation model trained with DINO and iBOT objectives and refined using high‑resolution Gram anchoring on a 500‑million‑tile corpus from about 575,000 clinical whole‑slide images. It demonstrates superior robustness to center, scanner, and stain variation compared to other state‑of‑the‑art foundation models, while maintaining competitive performance on tile‑level classification, segmentation, and gene‑expression prediction. Across 39 slide‑level clinical tasks, HERO ranks first on average and achieves the best average rank across six benchmark frameworks under an equal‑weighted analysis.
The study investigates whether pathology foundation models can enable training‑free, content‑based retrieval of similar cases from digitized archives of salivary gland tumors. Using a region of interest from a whole slide image, the models retrieve morphologically similar cases and their ICD‑O codes without task‑specific training. Evaluation on 137 cases across 12 morphology categories shows that Atlas2 achieves the highest retrieval accuracy, with macro‑averaged F1 scores ranging from 69.0% to 76.6% across models.
arXiv:2610.15669v2 Announce Type: replace-cross Abstract: Weak bag-level labels are widely available in computational pathology, whereas reliable instance-level annotations for whole-slide images rem...
arXiv:2606. 13135v1 Announce Type: cross Abstract: Purpose.
The paper introduces a pipeline that uses publicly available whole slide image foundation models (FMs) to automatically triage slides by ranking them based on zero‑shot classification predictions. This approach accurately identifies slides containing the most tumor, achieving top‑2 ranking for patients with up to 43 slides across multiple datasets. The study also proposes a ranked evaluation framework to benchmark FM performance in slide triage.
The study investigates whether pathology foundation models (PFMs) carry center-related biases into whole-slide image (WSI) classification. By training models with increasing class-center correlations and evaluating six PFMs across four datasets and two MIL aggregators, the authors introduce the Area Under the Cramér's V Curve (AUCC) to measure both accuracy and degradation due to spurious correlations. Results reveal that center information propagates to WSI predictions, with robustness varying by PFM and MIL strategy, and that ComBat harmonization does not consistently improve robustness.
arXiv:2609.00396v1 Announce Type: new Abstract: Histopathological whole slide images (WSIs) are central to cancer diagnosis, but their gigapixel scale, tissue heterogeneity, weak slide-level supervis...
arXiv:2606. 00092v1 Announce Type: cross Abstract: Weakly-supervised classification of whole-slide images with attention-based multiple instance learning (ABMIL) on top of foundation features now reaches near-saturation on Camelyon16 slide-level performance, but the corresponding attention maps are an imperfect localization signal: in clinical interpretation, a model that classifies correctly without firing on the actual lesion is hard to trust.
arXiv:2602. 14010v2 Announce Type: replace-cross Abstract: Pathology foundation models (PFMs) generalize well across computational pathology tasks but remain costly for gigapixel whole-slide image analysis.