mlsanityData to 14 Oct 2026

DINOv3-MIL: Per-Kidney Multi-Label Tumour and Cyst Detection from Foundation-Model Patch Tokens on KiTS23

arXiv 2607.22687 ↗First posted 16 Jul 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Foundation vision models trained on natural images transfer to medical tasks without domain pre-training, but volumetric classification requires aggregating tens of thousands of patch tokens per study, and the aggregator constrains how the resulting model can be interpreted. We compare three aggregators on identical frozen DINOv3 ViT-H/16+ features for renal tumour/cyst detection on KiTS23 (966 kidneys; n=97 test): a CLS-token linear probe, gated attention multiple instance learning (MIL) over 55,296 patch tokens, and a prototype head following ProtoViT. Attention MIL achieves the highest AUROC for tumour (0.74, 95% CI 0.64-0.83) and cyst (0.80, 0.70-0.88), with attention enriched 7.5-9.8x over chance within annotated lesions. The prototype head does not transfer to cyst detection (AUROC 0.51), exposing an interpretability-performance trade-off at this token scale.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 4 of its 7 references that are in the map. First posted on arXiv 16 Jul 2026.

Builds on

4 of 7 references.

References · 4

  1. 2018Sanity Checks for Saliency Mapscited by 1,156
  2. 2018Attention-based Deep Multiple Instance Learningcited by 1,095
  3. 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 737
  4. 2024Interpretable Image Classification with Adaptive Prototype-based Vision Transformerscited by 25

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025A Disease-Centric Vision-Language Foundation Model for Precision Oncology in Kidney Cancer
  2. 2025CytoFM: The first cytology foundation model
  3. 2026Attend what matters: Leveraging vision foundational models for breast cancer classification using mammograms
  4. 2025From Classification to Cross-Modal Understanding: Leveraging Vision-Language Models for Fine-Grained Renal Pathology
  5. 2026Organ-Aware Attention Improves CT Triage and Classification
  6. 2023NephroNet: A Novel Program for Identifying Renal Cell Carcinoma and Generating Synthetic Training Images with Convolutional Neural Networks and Diffusion Models
  7. 2026CytoSyn: a Foundation Diffusion Model for Histopathology -- Tech Report
  8. 2026MagViT: Interpretable Multi-Magnification Transformers with Patient-Level Model Selection for Breast Histopathology